src/backend/storage/buffer/bufmgr.c - master - balmaster/postgres

src/backend/storage/buffer/bufmgr.c
8 967 строк260 KB
Michael Paquier
Fix MarkBufferDirtyHint() to not call GetBufferDescriptor() for local buffers
10 июн 2026, 07:49
10 июн 2026, 07:49e18b0cb
/*-------------------------------------------------------------------------
 *
 * bufmgr.c
 *	  buffer manager interface routines
 *
 * Portions Copyright (c) 1996-2026, PostgreSQL Global Development Group
 * Portions Copyright (c) 1994, Regents of the University of California
 *
 *
 * IDENTIFICATION
 *	  src/backend/storage/buffer/bufmgr.c
 *
 *-------------------------------------------------------------------------
 */
/*
 * Principal entry points:
 *
 * ReadBuffer() -- find or create a buffer holding the requested page,
 *		and pin it so that no one can destroy it while this process
 *		is using it.
 *
 * StartReadBuffer() -- as above, with separate wait step
 * StartReadBuffers() -- multiple block version
 * WaitReadBuffers() -- second step of above
 *
 * ReleaseBuffer() -- unpin a buffer
 *
 * MarkBufferDirty() -- mark a pinned buffer's contents as "dirty".
 *		The disk write is delayed until buffer replacement or checkpoint.
 *
 * See also these files:
 *		freelist.c -- chooses victim for buffer replacement
 *		buf_table.c -- manages the buffer lookup table
 */
#include "postgres.h"

#include <sys/file.h>
#include <unistd.h>

#include "access/tableam.h"
#include "access/xloginsert.h"
#include "access/xlogutils.h"
#ifdef USE_ASSERT_CHECKING
#include "catalog/pg_tablespace_d.h"
#endif
#include "catalog/storage.h"
#include "catalog/storage_xlog.h"
#include "common/hashfn.h"
#include "executor/instrument.h"
#include "lib/binaryheap.h"
#include "miscadmin.h"
#include "pg_trace.h"
#include "pgstat.h"
#include "postmaster/bgwriter.h"
#include "storage/aio.h"
#include "storage/buf_internals.h"
#include "storage/bufmgr.h"
#include "storage/fd.h"
#include "storage/ipc.h"
#include "storage/lmgr.h"
#include "storage/proc.h"
#include "storage/proclist.h"
#include "storage/procsignal.h"
#include "storage/read_stream.h"
#include "storage/smgr.h"
#include "storage/standby.h"
#include "utils/memdebug.h"
#include "utils/ps_status.h"
#include "utils/rel.h"
#include "utils/resowner.h"
#include "utils/timestamp.h"
#include "utils/wait_event.h"


/* Note: these two macros only work on shared buffers, not local ones! */
#define BufHdrGetBlock(bufHdr)	((Block) (BufferBlocks + ((Size) (bufHdr)->buf_id) * BLCKSZ))
#define BufferGetLSN(bufHdr)	(PageGetLSN(BufHdrGetBlock(bufHdr)))

/* Note: this macro only works on local buffers, not shared ones! */
#define LocalBufHdrGetBlock(bufHdr) \
	LocalBufferBlockPointers[-((bufHdr)->buf_id + 2)]

/* Bits in SyncOneBuffer's return value */
#define BUF_WRITTEN				0x01
#define BUF_REUSABLE			0x02

#define RELS_BSEARCH_THRESHOLD		20

/*
 * This is the size (in the number of blocks) above which we scan the
 * entire buffer pool to remove the buffers for all the pages of relation
 * being dropped. For the relations with size below this threshold, we find
 * the buffers by doing lookups in BufMapping table.
 */
#define BUF_DROP_FULL_SCAN_THRESHOLD		(uint64) (NBuffers / 32)

/*
 * This is separated out from PrivateRefCountEntry to allow for copying all
 * the data members via struct assignment.
 */
typedef struct PrivateRefCountData
{
	/*
	 * How many times has the buffer been pinned by this backend.
	 */
	int32		refcount;

	/*
	 * Is the buffer locked by this backend? BUFFER_LOCK_UNLOCK indicates that
	 * the buffer is not locked.
	 */
	BufferLockMode lockmode;
} PrivateRefCountData;

typedef struct PrivateRefCountEntry
{
	/*
	 * Note that this needs to be same as the entry's corresponding
	 * PrivateRefCountArrayKeys[i], if the entry is stored in the array. We
	 * store it in both places as this is used for the hashtable key and
	 * because it is more convenient (passing around a PrivateRefCountEntry
	 * suffices to identify the buffer) and faster (checking the keys array is
	 * faster when checking many entries, checking the entry is faster if just
	 * checking a single entry).
	 */
	Buffer		buffer;

	char		status;

	PrivateRefCountData data;
} PrivateRefCountEntry;

#define SH_PREFIX refcount
#define SH_ELEMENT_TYPE PrivateRefCountEntry
#define SH_KEY_TYPE Buffer
#define SH_KEY buffer
#define SH_HASH_KEY(tb, key) murmurhash32((uint32) (key))
#define SH_EQUAL(tb, a, b) ((a) == (b))
#define SH_SCOPE static inline
#define SH_DECLARE
#define SH_DEFINE
#include "lib/simplehash.h"

/* 64 bytes, about the size of a cache line on common systems */
#define REFCOUNT_ARRAY_ENTRIES 8

/*
 * Status of buffers to checkpoint for a particular tablespace, used
 * internally in BufferSync.
 */
typedef struct CkptTsStatus
{
	/* oid of the tablespace */
	Oid			tsId;

	/*
	 * Checkpoint progress for this tablespace. To make progress comparable
	 * between tablespaces the progress is, for each tablespace, measured as a
	 * number between 0 and the total number of to-be-checkpointed pages. Each
	 * page checkpointed in this tablespace increments this space's progress
	 * by progress_slice.
	 */
	float8		progress;
	float8		progress_slice;

	/* number of to-be checkpointed pages in this tablespace */
	int			num_to_scan;
	/* already processed pages in this tablespace */
	int			num_scanned;

	/* current offset in CkptBufferIds for this tablespace */
	int			index;
} CkptTsStatus;

/*
 * Type for array used to sort SMgrRelations
 *
 * FlushRelationsAllBuffers shares the same comparator function with
 * DropRelationsAllBuffers. Pointer to this struct and RelFileLocator must be
 * compatible.
 */
typedef struct SMgrSortArray
{
	RelFileLocator rlocator;	/* This must be the first member */
	SMgrRelation srel;
} SMgrSortArray;

/* GUC variables */
bool		zero_damaged_pages = false;
int			bgwriter_lru_maxpages = 100;
double		bgwriter_lru_multiplier = 2.0;
bool		track_io_timing = false;

/*
 * How many buffers PrefetchBuffer callers should try to stay ahead of their
 * ReadBuffer calls by.  Zero means "never prefetch".  This value is only used
 * for buffers not belonging to tablespaces that have their
 * effective_io_concurrency parameter set.
 */
int			effective_io_concurrency = DEFAULT_EFFECTIVE_IO_CONCURRENCY;

/*
 * Like effective_io_concurrency, but used by maintenance code paths that might
 * benefit from a higher setting because they work on behalf of many sessions.
 * Overridden by the tablespace setting of the same name.
 */
int			maintenance_io_concurrency = DEFAULT_MAINTENANCE_IO_CONCURRENCY;

/*
 * Limit on how many blocks should be handled in single I/O operations.
 * StartReadBuffers() callers should respect it, as should other operations
 * that call smgr APIs directly.  It is computed as the minimum of underlying
 * GUCs io_combine_limit_guc and io_max_combine_limit.
 */
int			io_combine_limit = DEFAULT_IO_COMBINE_LIMIT;
int			io_combine_limit_guc = DEFAULT_IO_COMBINE_LIMIT;
int			io_max_combine_limit = DEFAULT_IO_COMBINE_LIMIT;

/*
 * GUC variables about triggering kernel writeback for buffers written; OS
 * dependent defaults are set via the GUC mechanism.
 */
int			checkpoint_flush_after = DEFAULT_CHECKPOINT_FLUSH_AFTER;
int			bgwriter_flush_after = DEFAULT_BGWRITER_FLUSH_AFTER;
int			backend_flush_after = DEFAULT_BACKEND_FLUSH_AFTER;

/* local state for LockBufferForCleanup */
static BufferDesc *PinCountWaitBuf = NULL;

/*
 * Backend-Private refcount management:
 *
 * Each buffer also has a private refcount that keeps track of the number of
 * times the buffer is pinned in the current process.  This is so that the
 * shared refcount needs to be modified only once if a buffer is pinned more
 * than once by an individual backend.  It's also used to check that no
 * buffers are still pinned at the end of transactions and when exiting. We
 * also use this mechanism to track whether this backend has a buffer locked,
 * and, if so, in what mode.
 *
 *
 * To avoid - as we used to - requiring an array with NBuffers entries to keep
 * track of local buffers, we use a small sequentially searched array
 * (PrivateRefCountArrayKeys, with the corresponding data stored in
 * PrivateRefCountArray) and an overflow hash table (PrivateRefCountHash) to
 * keep track of backend local pins.
 *
 * Until no more than REFCOUNT_ARRAY_ENTRIES buffers are pinned at once, all
 * refcounts are kept track of in the array; after that, new array entries
 * displace old ones into the hash table. That way a frequently used entry
 * can't get "stuck" in the hashtable while infrequent ones clog the array.
 *
 * Note that in most scenarios the number of pinned buffers will not exceed
 * REFCOUNT_ARRAY_ENTRIES.
 *
 *
 * To enter a buffer into the refcount tracking mechanism first reserve a free
 * entry using ReservePrivateRefCountEntry() and then later, if necessary,
 * fill it with NewPrivateRefCountEntry(). That split lets us avoid doing
 * memory allocations in NewPrivateRefCountEntry() which can be important
 * because in some scenarios it's called with a spinlock held...
 */
static Buffer PrivateRefCountArrayKeys[REFCOUNT_ARRAY_ENTRIES];
static struct PrivateRefCountEntry PrivateRefCountArray[REFCOUNT_ARRAY_ENTRIES];
static refcount_hash *PrivateRefCountHash = NULL;
static int32 PrivateRefCountOverflowed = 0;
static uint32 PrivateRefCountClock = 0;
static int	ReservedRefCountSlot = -1;
static int	PrivateRefCountEntryLast = -1;

static uint32 MaxProportionalPins;

static void ReservePrivateRefCountEntry(void);
static PrivateRefCountEntry *NewPrivateRefCountEntry(Buffer buffer);
static PrivateRefCountEntry *GetPrivateRefCountEntry(Buffer buffer, bool do_move);
static inline int32 GetPrivateRefCount(Buffer buffer);
static void ForgetPrivateRefCountEntry(PrivateRefCountEntry *ref);

/* ResourceOwner callbacks to hold in-progress I/Os and buffer pins */
static void ResOwnerReleaseBufferIO(Datum res);
static char *ResOwnerPrintBufferIO(Datum res);
static void ResOwnerReleaseBuffer(Datum res);
static char *ResOwnerPrintBuffer(Datum res);

const ResourceOwnerDesc buffer_io_resowner_desc =
{
	.name = "buffer io",
	.release_phase = RESOURCE_RELEASE_BEFORE_LOCKS,
	.release_priority = RELEASE_PRIO_BUFFER_IOS,
	.ReleaseResource = ResOwnerReleaseBufferIO,
	.DebugPrint = ResOwnerPrintBufferIO
};

const ResourceOwnerDesc buffer_resowner_desc =
{
	.name = "buffer",
	.release_phase = RESOURCE_RELEASE_BEFORE_LOCKS,
	.release_priority = RELEASE_PRIO_BUFFER_PINS,
	.ReleaseResource = ResOwnerReleaseBuffer,
	.DebugPrint = ResOwnerPrintBuffer
};

/*
 * Ensure that the PrivateRefCountArray has sufficient space to store one more
 * entry. This has to be called before using NewPrivateRefCountEntry() to fill
 * a new entry - but it's perfectly fine to not use a reserved entry.
 */
static void
ReservePrivateRefCountEntry(void)
{
	/* Already reserved (or freed), nothing to do */
	if (ReservedRefCountSlot != -1)
		return;

	/*
	 * First search for a free entry the array, that'll be sufficient in the
	 * majority of cases.
	 */
	{
		int			i;

		for (i = 0; i < REFCOUNT_ARRAY_ENTRIES; i++)
		{
			if (PrivateRefCountArrayKeys[i] == InvalidBuffer)
			{
				ReservedRefCountSlot = i;

				/*
				 * We could return immediately, but iterating till the end of
				 * the array allows compiler-autovectorization.
				 */
			}
		}

		if (ReservedRefCountSlot != -1)
			return;
	}

	/*
	 * No luck. All array entries are full. Move one array entry into the hash
	 * table.
	 */
	{
		/*
		 * Move entry from the current clock position in the array into the
		 * hashtable. Use that slot.
		 */
		int			victim_slot;
		PrivateRefCountEntry *victim_entry;
		PrivateRefCountEntry *hashent;
		bool		found;

		/* select victim slot */
		victim_slot = PrivateRefCountClock++ % REFCOUNT_ARRAY_ENTRIES;
		victim_entry = &PrivateRefCountArray[victim_slot];
		ReservedRefCountSlot = victim_slot;

		/* Better be used, otherwise we shouldn't get here. */
		Assert(PrivateRefCountArrayKeys[victim_slot] != InvalidBuffer);
		Assert(PrivateRefCountArray[victim_slot].buffer != InvalidBuffer);
		Assert(PrivateRefCountArrayKeys[victim_slot] == PrivateRefCountArray[victim_slot].buffer);

		/* enter victim array entry into hashtable */
		hashent = refcount_insert(PrivateRefCountHash,
								  PrivateRefCountArrayKeys[victim_slot],
								  &found);
		Assert(!found);
		/* move data from the entry in the array to the hash entry */
		hashent->data = victim_entry->data;

		/* clear the now free array slot */
		PrivateRefCountArrayKeys[victim_slot] = InvalidBuffer;
		victim_entry->buffer = InvalidBuffer;

		/* clear the whole data member, just for future proofing */
		memset(&victim_entry->data, 0, sizeof(victim_entry->data));
		victim_entry->data.refcount = 0;
		victim_entry->data.lockmode = BUFFER_LOCK_UNLOCK;

		PrivateRefCountOverflowed++;
	}
}

/*
 * Fill a previously reserved refcount entry.
 */
static PrivateRefCountEntry *
NewPrivateRefCountEntry(Buffer buffer)
{
	PrivateRefCountEntry *res;

	/* only allowed to be called when a reservation has been made */
	Assert(ReservedRefCountSlot != -1);

	/* use up the reserved entry */
	res = &PrivateRefCountArray[ReservedRefCountSlot];

	/* and fill it */
	PrivateRefCountArrayKeys[ReservedRefCountSlot] = buffer;
	res->buffer = buffer;
	res->data.refcount = 0;
	res->data.lockmode = BUFFER_LOCK_UNLOCK;

	/* update cache for the next lookup */
	PrivateRefCountEntryLast = ReservedRefCountSlot;

	ReservedRefCountSlot = -1;

	return res;
}

/*
 * Slow-path for GetPrivateRefCountEntry(). This is big enough to not be worth
 * inlining. This particularly seems to be true if the compiler is capable of
 * auto-vectorizing the code, as that imposes additional stack-alignment
 * requirements etc.
 */
static pg_noinline PrivateRefCountEntry *
GetPrivateRefCountEntrySlow(Buffer buffer, bool do_move)
{
	PrivateRefCountEntry *res;
	int			match = -1;
	int			i;

	/*
	 * First search for references in the array, that'll be sufficient in the
	 * majority of cases.
	 */
	for (i = 0; i < REFCOUNT_ARRAY_ENTRIES; i++)
	{
		if (PrivateRefCountArrayKeys[i] == buffer)
		{
			match = i;
			/* see ReservePrivateRefCountEntry() for why we don't return */
		}
	}

	if (likely(match != -1))
	{
		/* update cache for the next lookup */
		PrivateRefCountEntryLast = match;

		return &PrivateRefCountArray[match];
	}

	/*
	 * By here we know that the buffer, if already pinned, isn't residing in
	 * the array.
	 *
	 * Only look up the buffer in the hashtable if we've previously overflowed
	 * into it.
	 */
	if (PrivateRefCountOverflowed == 0)
		return NULL;

	res = refcount_lookup(PrivateRefCountHash, buffer);

	if (res == NULL)
		return NULL;
	else if (!do_move)
	{
		/* caller doesn't want us to move the hash entry into the array */
		return res;
	}
	else
	{
		/* move buffer from hashtable into the free array slot */
		PrivateRefCountEntry *free;
		PrivateRefCountData data;

		/* Save data and delete from hashtable while res is still valid */
		data = res->data;
		refcount_delete_item(PrivateRefCountHash, res);
		Assert(PrivateRefCountOverflowed > 0);
		PrivateRefCountOverflowed--;

		/* Ensure there's a free array slot */
		ReservePrivateRefCountEntry();

		/* Use up the reserved slot */
		Assert(ReservedRefCountSlot != -1);
		free = &PrivateRefCountArray[ReservedRefCountSlot];
		Assert(PrivateRefCountArrayKeys[ReservedRefCountSlot] == free->buffer);
		Assert(free->buffer == InvalidBuffer);

		/* and fill it */
		free->buffer = buffer;
		free->data = data;
		PrivateRefCountArrayKeys[ReservedRefCountSlot] = buffer;
		/* update cache for the next lookup */
		PrivateRefCountEntryLast = ReservedRefCountSlot;

		ReservedRefCountSlot = -1;

		return free;
	}
}

/*
 * Return the PrivateRefCount entry for the passed buffer.
 *
 * Returns NULL if a buffer doesn't have a refcount entry. Otherwise, if
 * do_move is true, and the entry resides in the hashtable the entry is
 * optimized for frequent access by moving it to the array.
 */
static inline PrivateRefCountEntry *
GetPrivateRefCountEntry(Buffer buffer, bool do_move)
{
	Assert(BufferIsValid(buffer));
	Assert(!BufferIsLocal(buffer));

	/*
	 * It's very common to look up the same buffer repeatedly. To make that
	 * fast, we have a one-entry cache.
	 *
	 * In contrast to the loop in GetPrivateRefCountEntrySlow(), here it
	 * faster to check PrivateRefCountArray[].buffer, as in the case of a hit
	 * fewer addresses are computed and fewer cachelines are accessed. Whereas
	 * in GetPrivateRefCountEntrySlow()'s case, checking
	 * PrivateRefCountArrayKeys saves a lot of memory accesses.
	 */
	if (likely(PrivateRefCountEntryLast != -1) &&
		likely(PrivateRefCountArray[PrivateRefCountEntryLast].buffer == buffer))
	{
		return &PrivateRefCountArray[PrivateRefCountEntryLast];
	}

	/*
	 * The code for the cached lookup is small enough to be worth inlining
	 * into the caller. In the miss case however, that empirically doesn't
	 * seem worth it.
	 */
	return GetPrivateRefCountEntrySlow(buffer, do_move);
}

/*
 * Returns how many times the passed buffer is pinned by this backend.
 *
 * Only works for shared memory buffers!
 */
static inline int32
GetPrivateRefCount(Buffer buffer)
{
	PrivateRefCountEntry *ref;

	Assert(BufferIsValid(buffer));
	Assert(!BufferIsLocal(buffer));

	/*
	 * Not moving the entry - that's ok for the current users, but we might
	 * want to change this one day.
	 */
	ref = GetPrivateRefCountEntry(buffer, false);

	if (ref == NULL)
		return 0;
	return ref->data.refcount;
}

/*
 * Release resources used to track the reference count of a buffer which we no
 * longer have pinned and don't want to pin again immediately.
 */
static void
ForgetPrivateRefCountEntry(PrivateRefCountEntry *ref)
{
	Assert(ref->data.refcount == 0);
	Assert(ref->data.lockmode == BUFFER_LOCK_UNLOCK);

	if (ref >= &PrivateRefCountArray[0] &&
		ref < &PrivateRefCountArray[REFCOUNT_ARRAY_ENTRIES])
	{
		ref->buffer = InvalidBuffer;
		PrivateRefCountArrayKeys[ref - PrivateRefCountArray] = InvalidBuffer;


		/*
		 * Mark the just used entry as reserved - in many scenarios that
		 * allows us to avoid ever having to search the array/hash for free
		 * entries.
		 */
		ReservedRefCountSlot = ref - PrivateRefCountArray;
	}
	else
	{
		refcount_delete_item(PrivateRefCountHash, ref);
		Assert(PrivateRefCountOverflowed > 0);
		PrivateRefCountOverflowed--;
	}
}

/*
 * BufferIsPinned
 *		True iff the buffer is pinned (also checks for valid buffer number).
 *
 *		NOTE: what we check here is that *this* backend holds a pin on
 *		the buffer.  We do not care whether some other backend does.
 */
#define BufferIsPinned(bufnum) \
( \
	!BufferIsValid(bufnum) ? \
		false \
	: \
		BufferIsLocal(bufnum) ? \
			(LocalRefCount[-(bufnum) - 1] > 0) \
		: \
	(GetPrivateRefCount(bufnum) > 0) \
)


static Buffer ReadBuffer_common(Relation rel,
								SMgrRelation smgr, char smgr_persistence,
								ForkNumber forkNum, BlockNumber blockNum,
								ReadBufferMode mode, BufferAccessStrategy strategy);
static BlockNumber ExtendBufferedRelCommon(BufferManagerRelation bmr,
										   ForkNumber fork,
										   BufferAccessStrategy strategy,
										   uint32 flags,
										   uint32 extend_by,
										   BlockNumber extend_upto,
										   Buffer *buffers,
										   uint32 *extended_by);
static BlockNumber ExtendBufferedRelShared(BufferManagerRelation bmr,
										   ForkNumber fork,
										   BufferAccessStrategy strategy,
										   uint32 flags,
										   uint32 extend_by,
										   BlockNumber extend_upto,
										   Buffer *buffers,
										   uint32 *extended_by);
static bool PinBuffer(BufferDesc *buf, BufferAccessStrategy strategy,
					  bool skip_if_not_valid);
static void PinBuffer_Locked(BufferDesc *buf);
static void UnpinBuffer(BufferDesc *buf);
static void UnpinBufferNoOwner(BufferDesc *buf);
static void BufferSync(int flags);
static int	SyncOneBuffer(int buf_id, bool skip_recently_used,
						  WritebackContext *wb_context);
static void WaitIO(BufferDesc *buf);
static void AbortBufferIO(Buffer buffer);
static void shared_buffer_write_error_callback(void *arg);
static void local_buffer_write_error_callback(void *arg);
static inline BufferDesc *BufferAlloc(SMgrRelation smgr,
									  char relpersistence,
									  ForkNumber forkNum,
									  BlockNumber blockNum,
									  BufferAccessStrategy strategy,
									  bool *foundPtr, IOContext io_context);
static bool AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress);
static void CheckReadBuffersOperation(ReadBuffersOperation *operation, bool is_complete);

static pg_attribute_always_inline void TrackBufferHit(IOObject io_object,
													  IOContext io_context,
													  Relation rel, char persistence, SMgrRelation smgr,
													  ForkNumber forknum, BlockNumber blocknum);
static Buffer GetVictimBuffer(BufferAccessStrategy strategy, IOContext io_context);
static void FlushUnlockedBuffer(BufferDesc *buf, SMgrRelation reln,
								IOObject io_object, IOContext io_context);
static void FlushBuffer(BufferDesc *buf, SMgrRelation reln,
						IOObject io_object, IOContext io_context);
static void FindAndDropRelationBuffers(RelFileLocator rlocator,
									   ForkNumber forkNum,
									   BlockNumber nForkBlock,
									   BlockNumber firstDelBlock);
static void RelationCopyStorageUsingBuffer(RelFileLocator srclocator,
										   RelFileLocator dstlocator,
										   ForkNumber forkNum, bool permanent);
static void AtProcExit_Buffers(int code, Datum arg);
static void CheckForBufferLeaks(void);
#ifdef USE_ASSERT_CHECKING
static void AssertNotCatalogBufferLock(Buffer buffer, BufferLockMode mode);
#endif
static int	rlocator_comparator(const void *p1, const void *p2);
static inline int buffertag_comparator(const BufferTag *ba, const BufferTag *bb);
static inline int ckpt_buforder_comparator(const CkptSortItem *a, const CkptSortItem *b);
static int	ts_ckpt_progress_comparator(Datum a, Datum b, void *arg);

static void BufferLockAcquire(Buffer buffer, BufferDesc *buf_hdr, BufferLockMode mode);
static void BufferLockUnlock(Buffer buffer, BufferDesc *buf_hdr);
static bool BufferLockConditional(Buffer buffer, BufferDesc *buf_hdr, BufferLockMode mode);
static bool BufferLockHeldByMeInMode(BufferDesc *buf_hdr, BufferLockMode mode);
static bool BufferLockHeldByMe(BufferDesc *buf_hdr);
static inline void BufferLockDisown(Buffer buffer, BufferDesc *buf_hdr);
static inline int BufferLockDisownInternal(Buffer buffer, BufferDesc *buf_hdr);
static inline bool BufferLockAttempt(BufferDesc *buf_hdr, BufferLockMode mode);
static void BufferLockQueueSelf(BufferDesc *buf_hdr, BufferLockMode mode);
static void BufferLockDequeueSelf(BufferDesc *buf_hdr);
static void BufferLockWakeup(BufferDesc *buf_hdr, bool wake_exclusive);
static void BufferLockProcessRelease(BufferDesc *buf_hdr, BufferLockMode mode, uint64 lockstate);
static inline uint64 BufferLockReleaseSub(BufferLockMode mode);


/*
 * Implementation of PrefetchBuffer() for shared buffers.
 */
PrefetchBufferResult
PrefetchSharedBuffer(SMgrRelation smgr_reln,
					 ForkNumber forkNum,
					 BlockNumber blockNum)
{
	PrefetchBufferResult result = {InvalidBuffer, false};
	BufferTag	newTag;			/* identity of requested block */
	uint32		newHash;		/* hash value for newTag */
	LWLock	   *newPartitionLock;	/* buffer partition lock for it */
	int			buf_id;

	Assert(BlockNumberIsValid(blockNum));

	/* create a tag so we can lookup the buffer */
	InitBufferTag(&newTag, &smgr_reln->smgr_rlocator.locator,
				  forkNum, blockNum);

	/* determine its hash code and partition lock ID */
	newHash = BufTableHashCode(&newTag);
	newPartitionLock = BufMappingPartitionLock(newHash);

	/* see if the block is in the buffer pool already */
	LWLockAcquire(newPartitionLock, LW_SHARED);
	buf_id = BufTableLookup(&newTag, newHash);
	LWLockRelease(newPartitionLock);

	/* If not in buffers, initiate prefetch */
	if (buf_id < 0)
	{
#ifdef USE_PREFETCH
		/*
		 * Try to initiate an asynchronous read.  This returns false in
		 * recovery if the relation file doesn't exist.
		 */
		if ((io_direct_flags & IO_DIRECT_DATA) == 0 &&
			smgrprefetch(smgr_reln, forkNum, blockNum, 1))
		{
			result.initiated_io = true;
		}
#endif							/* USE_PREFETCH */
	}
	else
	{
		/*
		 * Report the buffer it was in at that time.  The caller may be able
		 * to avoid a buffer table lookup, but it's not pinned and it must be
		 * rechecked!
		 */
		result.recent_buffer = buf_id + 1;
	}

	/*
	 * If the block *is* in buffers, we do nothing.  This is not really ideal:
	 * the block might be just about to be evicted, which would be stupid
	 * since we know we are going to need it soon.  But the only easy answer
	 * is to bump the usage_count, which does not seem like a great solution:
	 * when the caller does ultimately touch the block, usage_count would get
	 * bumped again, resulting in too much favoritism for blocks that are
	 * involved in a prefetch sequence. A real fix would involve some
	 * additional per-buffer state, and it's not clear that there's enough of
	 * a problem to justify that.
	 */

	return result;
}

/*
 * PrefetchBuffer -- initiate asynchronous read of a block of a relation
 *
 * This is named by analogy to ReadBuffer but doesn't actually allocate a
 * buffer.  Instead it tries to ensure that a future ReadBuffer for the given
 * block will not be delayed by the I/O.  Prefetching is optional.
 *
 * There are three possible outcomes:
 *
 * 1.  If the block is already cached, the result includes a valid buffer that
 * could be used by the caller to avoid the need for a later buffer lookup, but
 * it's not pinned, so the caller must recheck it.
 *
 * 2.  If the kernel has been asked to initiate I/O, the initiated_io member is
 * true.  Currently there is no way to know if the data was already cached by
 * the kernel and therefore didn't really initiate I/O, and no way to know when
 * the I/O completes other than using synchronous ReadBuffer().
 *
 * 3.  Otherwise, the buffer wasn't already cached by PostgreSQL, and
 * USE_PREFETCH is not defined (this build doesn't support prefetching due to
 * lack of a kernel facility), direct I/O is enabled, or the underlying
 * relation file wasn't found and we are in recovery.  (If the relation file
 * wasn't found and we are not in recovery, an error is raised).
 */
PrefetchBufferResult
PrefetchBuffer(Relation reln, ForkNumber forkNum, BlockNumber blockNum)
{
	Assert(RelationIsValid(reln));
	Assert(BlockNumberIsValid(blockNum));

	if (RelationUsesLocalBuffers(reln))
	{
		/* see comments in ReadBuffer_common */
		if (RELATION_IS_OTHER_TEMP(reln))
			ereport(ERROR,
					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
					 errmsg("cannot access temporary tables of other sessions")));

		/* pass it off to localbuf.c */
		return PrefetchLocalBuffer(RelationGetSmgr(reln), forkNum, blockNum);
	}
	else
	{
		/* pass it to the shared buffer version */
		return PrefetchSharedBuffer(RelationGetSmgr(reln), forkNum, blockNum);
	}
}

/*
 * ReadRecentBuffer -- try to pin a block in a recently observed buffer
 *
 * Compared to ReadBuffer(), this avoids a buffer mapping lookup when it's
 * successful.  Return true if the buffer is valid and still has the expected
 * tag.  In that case, the buffer is pinned and the usage count is bumped.
 */
bool
ReadRecentBuffer(RelFileLocator rlocator, ForkNumber forkNum, BlockNumber blockNum,
				 Buffer recent_buffer)
{
	BufferDesc *bufHdr;
	BufferTag	tag;
	uint64		buf_state;

	Assert(BufferIsValid(recent_buffer));

	ResourceOwnerEnlarge(CurrentResourceOwner);
	ReservePrivateRefCountEntry();
	InitBufferTag(&tag, &rlocator, forkNum, blockNum);

	if (BufferIsLocal(recent_buffer))
	{
		int			b = -recent_buffer - 1;

		bufHdr = GetLocalBufferDescriptor(b);
		buf_state = pg_atomic_read_u64(&bufHdr->state);

		/* Is it still valid and holding the right tag? */
		if ((buf_state & BM_VALID) && BufferTagsEqual(&tag, &bufHdr->tag))
		{
			PinLocalBuffer(bufHdr, true);

			pgBufferUsage.local_blks_hit++;

			return true;
		}
	}
	else
	{
		bufHdr = GetBufferDescriptor(recent_buffer - 1);

		/*
		 * Is it still valid and holding the right tag?  We do an unlocked tag
		 * comparison first, to make it unlikely that we'll increment the
		 * usage counter of the wrong buffer, if someone calls us with a very
		 * out of date recent_buffer.  Then we'll check it again if we get the
		 * pin.
		 */
		if (BufferTagsEqual(&tag, &bufHdr->tag) &&
			PinBuffer(bufHdr, NULL, true))
		{
			if (BufferTagsEqual(&tag, &bufHdr->tag))
			{
				pgBufferUsage.shared_blks_hit++;
				return true;
			}
			UnpinBuffer(bufHdr);
		}
	}

	return false;
}

/*
 * ReadBuffer -- a shorthand for ReadBufferExtended, for reading from main
 *		fork with RBM_NORMAL mode and default strategy.
 */
Buffer
ReadBuffer(Relation reln, BlockNumber blockNum)
{
	return ReadBufferExtended(reln, MAIN_FORKNUM, blockNum, RBM_NORMAL, NULL);
}

/*
 * ReadBufferExtended -- returns a buffer containing the requested
 *		block of the requested relation.  If the blknum
 *		requested is P_NEW, extend the relation file and
 *		allocate a new block.  (Caller is responsible for
 *		ensuring that only one backend tries to extend a
 *		relation at the same time!)
 *
 * Returns: the buffer number for the buffer containing
 *		the block read.  The returned buffer has been pinned.
 *		Does not return on error --- elog's instead.
 *
 * Assume when this function is called, that reln has been opened already.
 *
 * In RBM_NORMAL mode, the page is read from disk, and the page header is
 * validated.  An error is thrown if the page header is not valid.  (But
 * note that an all-zero page is considered "valid"; see
 * PageIsVerified().)
 *
 * RBM_ZERO_ON_ERROR is like the normal mode, but if the page header is not
 * valid, the page is zeroed instead of throwing an error. This is intended
 * for non-critical data, where the caller is prepared to repair errors.
 *
 * In RBM_ZERO_AND_LOCK mode, if the page isn't in buffer cache already, it's
 * filled with zeros instead of reading it from disk.  Useful when the caller
 * is going to fill the page from scratch, since this saves I/O and avoids
 * unnecessary failure if the page-on-disk has corrupt page headers.
 * The page is returned locked to ensure that the caller has a chance to
 * initialize the page before it's made visible to others.
 * Caution: do not use this mode to read a page that is beyond the relation's
 * current physical EOF; that is likely to cause problems in md.c when
 * the page is modified and written out. P_NEW is OK, though.
 *
 * RBM_ZERO_AND_CLEANUP_LOCK is the same as RBM_ZERO_AND_LOCK, but acquires
 * a cleanup-strength lock on the page.
 *
 * RBM_NORMAL_NO_LOG mode is treated the same as RBM_NORMAL here.
 *
 * If strategy is not NULL, a nondefault buffer access strategy is used.
 * See buffer/README for details.
 */
inline Buffer
ReadBufferExtended(Relation reln, ForkNumber forkNum, BlockNumber blockNum,
				   ReadBufferMode mode, BufferAccessStrategy strategy)
{
	Buffer		buf;

	/*
	 * Read the buffer, and update pgstat counters to reflect a cache hit or
	 * miss.  The other-session temp-relation check is enforced by
	 * ReadBuffer_common().
	 */
	buf = ReadBuffer_common(reln, RelationGetSmgr(reln), 0,
							forkNum, blockNum, mode, strategy);

	return buf;
}


/*
 * ReadBufferWithoutRelcache -- like ReadBufferExtended, but doesn't require
 *		a relcache entry for the relation.
 *
 * Pass permanent = true for a RELPERSISTENCE_PERMANENT relation, and
 * permanent = false for a RELPERSISTENCE_UNLOGGED relation. This function
 * cannot be used for temporary relations (and making that work might be
 * difficult, unless we only want to read temporary relations for our own
 * ProcNumber).
 */
Buffer
ReadBufferWithoutRelcache(RelFileLocator rlocator, ForkNumber forkNum,
						  BlockNumber blockNum, ReadBufferMode mode,
						  BufferAccessStrategy strategy, bool permanent)
{
	SMgrRelation smgr = smgropen(rlocator, INVALID_PROC_NUMBER);

	return ReadBuffer_common(NULL, smgr,
							 permanent ? RELPERSISTENCE_PERMANENT : RELPERSISTENCE_UNLOGGED,
							 forkNum, blockNum,
							 mode, strategy);
}

/*
 * Convenience wrapper around ExtendBufferedRelBy() extending by one block.
 */
Buffer
ExtendBufferedRel(BufferManagerRelation bmr,
				  ForkNumber forkNum,
				  BufferAccessStrategy strategy,
				  uint32 flags)
{
	Buffer		buf;
	uint32		extend_by = 1;

	ExtendBufferedRelBy(bmr, forkNum, strategy, flags, extend_by,
						&buf, &extend_by);

	return buf;
}

/*
 * Extend relation by multiple blocks.
 *
 * Tries to extend the relation by extend_by blocks. Depending on the
 * availability of resources the relation may end up being extended by a
 * smaller number of pages (unless an error is thrown, always by at least one
 * page). *extended_by is updated to the number of pages the relation has been
 * extended to.
 *
 * buffers needs to be an array that is at least extend_by long. Upon
 * completion, the first extend_by array elements will point to a pinned
 * buffer.
 *
 * If EB_LOCK_FIRST is part of flags, the first returned buffer is
 * locked. This is useful for callers that want a buffer that is guaranteed to
 * be empty.
 */
BlockNumber
ExtendBufferedRelBy(BufferManagerRelation bmr,
					ForkNumber fork,
					BufferAccessStrategy strategy,
					uint32 flags,
					uint32 extend_by,
					Buffer *buffers,
					uint32 *extended_by)
{
	Assert((bmr.rel != NULL) != (bmr.smgr != NULL));
	Assert(bmr.smgr == NULL || bmr.relpersistence != '\0');
	Assert(extend_by > 0);

	if (bmr.relpersistence == '\0')
		bmr.relpersistence = bmr.rel->rd_rel->relpersistence;

	return ExtendBufferedRelCommon(bmr, fork, strategy, flags,
								   extend_by, InvalidBlockNumber,
								   buffers, extended_by);
}

/*
 * Extend the relation so it is at least extend_to blocks large, return buffer
 * (extend_to - 1).
 *
 * This is useful for callers that want to write a specific page, regardless
 * of the current size of the relation (e.g. useful for visibilitymap and for
 * crash recovery).
 */
Buffer
ExtendBufferedRelTo(BufferManagerRelation bmr,
					ForkNumber fork,
					BufferAccessStrategy strategy,
					uint32 flags,
					BlockNumber extend_to,
					ReadBufferMode mode)
{
	BlockNumber current_size;
	uint32		extended_by = 0;
	Buffer		buffer = InvalidBuffer;
	Buffer		buffers[64];

	Assert((bmr.rel != NULL) != (bmr.smgr != NULL));
	Assert(bmr.smgr == NULL || bmr.relpersistence != '\0');
	Assert(extend_to != InvalidBlockNumber && extend_to > 0);

	if (bmr.relpersistence == '\0')
		bmr.relpersistence = bmr.rel->rd_rel->relpersistence;

	/*
	 * If desired, create the file if it doesn't exist.  If
	 * smgr_cached_nblocks[fork] is positive then it must exist, no need for
	 * an smgrexists call.
	 */
	if ((flags & EB_CREATE_FORK_IF_NEEDED) &&
		(BMR_GET_SMGR(bmr)->smgr_cached_nblocks[fork] == 0 ||
		 BMR_GET_SMGR(bmr)->smgr_cached_nblocks[fork] == InvalidBlockNumber) &&
		!smgrexists(BMR_GET_SMGR(bmr), fork))
	{
		LockRelationForExtension(bmr.rel, ExclusiveLock);

		/* recheck, fork might have been created concurrently */
		if (!smgrexists(BMR_GET_SMGR(bmr), fork))
			smgrcreate(BMR_GET_SMGR(bmr), fork, flags & EB_PERFORMING_RECOVERY);

		UnlockRelationForExtension(bmr.rel, ExclusiveLock);
	}

	/*
	 * If requested, invalidate size cache, so that smgrnblocks asks the
	 * kernel.
	 */
	if (flags & EB_CLEAR_SIZE_CACHE)
		BMR_GET_SMGR(bmr)->smgr_cached_nblocks[fork] = InvalidBlockNumber;

	/*
	 * Estimate how many pages we'll need to extend by. This avoids acquiring
	 * unnecessarily many victim buffers.
	 */
	current_size = smgrnblocks(BMR_GET_SMGR(bmr), fork);

	/*
	 * Since no-one else can be looking at the page contents yet, there is no
	 * difference between an exclusive lock and a cleanup-strength lock. Note
	 * that we pass the original mode to ReadBuffer_common() below, when
	 * falling back to reading the buffer to a concurrent relation extension.
	 */
	if (mode == RBM_ZERO_AND_LOCK || mode == RBM_ZERO_AND_CLEANUP_LOCK)
		flags |= EB_LOCK_TARGET;

	while (current_size < extend_to)
	{
		uint32		num_pages = lengthof(buffers);
		BlockNumber first_block;

		if ((uint64) current_size + num_pages > extend_to)
			num_pages = extend_to - current_size;

		first_block = ExtendBufferedRelCommon(bmr, fork, strategy, flags,
											  num_pages, extend_to,
											  buffers, &extended_by);

		current_size = first_block + extended_by;
		Assert(num_pages != 0 || current_size >= extend_to);

		for (uint32 i = 0; i < extended_by; i++)
		{
			if (first_block + i != extend_to - 1)
				ReleaseBuffer(buffers[i]);
			else
				buffer = buffers[i];
		}
	}

	/*
	 * It's possible that another backend concurrently extended the relation.
	 * In that case read the buffer.
	 *
	 * XXX: Should we control this via a flag?
	 */
	if (buffer == InvalidBuffer)
	{
		Assert(extended_by == 0);
		buffer = ReadBuffer_common(bmr.rel, BMR_GET_SMGR(bmr), bmr.relpersistence,
								   fork, extend_to - 1, mode, strategy);
	}

	return buffer;
}

/*
 * Lock and optionally zero a buffer, as part of the implementation of
 * RBM_ZERO_AND_LOCK or RBM_ZERO_AND_CLEANUP_LOCK.  The buffer must be already
 * pinned.  If the buffer is not already valid, it is zeroed and made valid.
 */
static void
ZeroAndLockBuffer(Buffer buffer, ReadBufferMode mode, bool already_valid)
{
	BufferDesc *bufHdr;
	bool		need_to_zero;
	bool		isLocalBuf = BufferIsLocal(buffer);
	StartBufferIOResult sbres;

	Assert(mode == RBM_ZERO_AND_LOCK || mode == RBM_ZERO_AND_CLEANUP_LOCK);

	if (already_valid)
	{
		/*
		 * If the caller already knew the buffer was valid, we can skip some
		 * header interaction.  The caller just wants to lock the buffer.
		 */
		need_to_zero = false;
	}
	else
	{
		if (isLocalBuf)
		{
			/* Simple case for non-shared buffers. */
			bufHdr = GetLocalBufferDescriptor(-buffer - 1);
			sbres = StartLocalBufferIO(bufHdr, true, true, NULL);
		}
		else
		{
			/*
			 * Take BM_IO_IN_PROGRESS, or discover that BM_VALID has been set
			 * concurrently.  Even though we aren't doing I/O, that ensures
			 * that we don't zero a page that someone else has pinned.  An
			 * exclusive content lock wouldn't be enough, because readers are
			 * allowed to drop the content lock after determining that a tuple
			 * is visible (see buffer access rules in README).
			 */
			bufHdr = GetBufferDescriptor(buffer - 1);
			sbres = StartSharedBufferIO(bufHdr, true, true, NULL);
		}

		Assert(sbres != BUFFER_IO_IN_PROGRESS);
		need_to_zero = sbres == BUFFER_IO_READY_FOR_IO;
	}

	if (need_to_zero)
	{
		memset(BufferGetPage(buffer), 0, BLCKSZ);

		/*
		 * Grab the buffer content lock before marking the page as valid, to
		 * make sure that no other backend sees the zeroed page before the
		 * caller has had a chance to initialize it.
		 *
		 * Since no-one else can be looking at the page contents yet, there is
		 * no difference between an exclusive lock and a cleanup-strength
		 * lock. (Note that we cannot use LockBuffer() or
		 * LockBufferForCleanup() here, because they assert that the buffer is
		 * already valid.)
		 */
		if (!isLocalBuf)
			LockBuffer(buffer, BUFFER_LOCK_EXCLUSIVE);

		/* Set BM_VALID, terminate IO, and wake up any waiters */
		if (isLocalBuf)
			TerminateLocalBufferIO(bufHdr, false, BM_VALID, false);
		else
			TerminateBufferIO(bufHdr, false, BM_VALID, true, false);
	}
	else if (!isLocalBuf)
	{
		/*
		 * The buffer is valid, so we can't zero it.  The caller still expects
		 * the page to be locked on return.
		 */
		if (mode == RBM_ZERO_AND_LOCK)
			LockBuffer(buffer, BUFFER_LOCK_EXCLUSIVE);
		else
			LockBufferForCleanup(buffer);
	}
}

/*
 * Pin a buffer for a given block.  *foundPtr is set to true if the block was
 * already present, or false if more work is required to either read it in or
 * zero it.
 */
static pg_attribute_always_inline Buffer
PinBufferForBlock(Relation rel,
				  SMgrRelation smgr,
				  char persistence,
				  ForkNumber forkNum,
				  BlockNumber blockNum,
				  BufferAccessStrategy strategy,
				  IOObject io_object,
				  IOContext io_context,
				  bool *foundPtr)
{
	BufferDesc *bufHdr;

	Assert(blockNum != P_NEW);

	/* Persistence should be set before */
	Assert((persistence == RELPERSISTENCE_TEMP ||
			persistence == RELPERSISTENCE_PERMANENT ||
			persistence == RELPERSISTENCE_UNLOGGED));

	TRACE_POSTGRESQL_BUFFER_READ_START(forkNum, blockNum,
									   smgr->smgr_rlocator.locator.spcOid,
									   smgr->smgr_rlocator.locator.dbOid,
									   smgr->smgr_rlocator.locator.relNumber,
									   smgr->smgr_rlocator.backend);

	if (persistence == RELPERSISTENCE_TEMP)
		bufHdr = LocalBufferAlloc(smgr, forkNum, blockNum, foundPtr);
	else
		bufHdr = BufferAlloc(smgr, persistence, forkNum, blockNum,
							 strategy, foundPtr, io_context);

	if (*foundPtr)
		TrackBufferHit(io_object, io_context, rel, persistence, smgr, forkNum, blockNum);

	if (rel)
	{
		/*
		 * While pgBufferUsage's "read" counter isn't bumped unless we reach
		 * WaitReadBuffers() (so, not for hits, and not for buffers that are
		 * zeroed instead), the per-relation stats always count them.
		 */
		pgstat_count_buffer_read(rel);
	}

	return BufferDescriptorGetBuffer(bufHdr);
}

/*
 * ReadBuffer_common -- common logic for all ReadBuffer variants
 *
 * smgr is required, rel is optional unless using P_NEW.
 */
static pg_attribute_always_inline Buffer
ReadBuffer_common(Relation rel, SMgrRelation smgr, char smgr_persistence,
				  ForkNumber forkNum,
				  BlockNumber blockNum, ReadBufferMode mode,
				  BufferAccessStrategy strategy)
{
	ReadBuffersOperation operation;
	Buffer		buffer;
	int			flags;
	char		persistence;

	/*
	 * Reject attempts to read non-local temporary relations; we would be
	 * likely to get wrong data since we have no visibility into the owning
	 * session's local buffers.  This is the canonical place for the check,
	 * covering the ReadBufferExtended() entry point and any other caller that
	 * supplies a Relation.
	 */
	if (rel && RELATION_IS_OTHER_TEMP(rel))
		ereport(ERROR,
				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
				 errmsg("cannot access temporary tables of other sessions")));

	/*
	 * Backward compatibility path, most code should use ExtendBufferedRel()
	 * instead, as acquiring the extension lock inside ExtendBufferedRel()
	 * scales a lot better.
	 */
	if (unlikely(blockNum == P_NEW))
	{
		uint32		flags = EB_SKIP_EXTENSION_LOCK;

		/*
		 * Since no-one else can be looking at the page contents yet, there is
		 * no difference between an exclusive lock and a cleanup-strength
		 * lock.
		 */
		if (mode == RBM_ZERO_AND_LOCK || mode == RBM_ZERO_AND_CLEANUP_LOCK)
			flags |= EB_LOCK_FIRST;

		return ExtendBufferedRel(BMR_REL(rel), forkNum, strategy, flags);
	}

	if (rel)
		persistence = rel->rd_rel->relpersistence;
	else
		persistence = smgr_persistence;

	if (unlikely(mode == RBM_ZERO_AND_CLEANUP_LOCK ||
				 mode == RBM_ZERO_AND_LOCK))
	{
		bool		found;
		IOContext	io_context;
		IOObject	io_object;

		if (persistence == RELPERSISTENCE_TEMP)
		{
			io_context = IOCONTEXT_NORMAL;
			io_object = IOOBJECT_TEMP_RELATION;
		}
		else
		{
			io_context = IOContextForStrategy(strategy);
			io_object = IOOBJECT_RELATION;
		}

		buffer = PinBufferForBlock(rel, smgr, persistence,
								   forkNum, blockNum, strategy,
								   io_object, io_context, &found);
		ZeroAndLockBuffer(buffer, mode, found);
		return buffer;
	}

	/*
	 * Signal that we are going to immediately wait. If we're immediately
	 * waiting, there is no benefit in actually executing the IO
	 * asynchronously, it would just add dispatch overhead.
	 */
	flags = READ_BUFFERS_SYNCHRONOUSLY;
	if (mode == RBM_ZERO_ON_ERROR)
		flags |= READ_BUFFERS_ZERO_ON_ERROR;
	operation.smgr = smgr;
	operation.rel = rel;
	operation.persistence = persistence;
	operation.forknum = forkNum;
	operation.strategy = strategy;
	if (StartReadBuffer(&operation,
						&buffer,
						blockNum,
						flags))
		WaitReadBuffers(&operation);

	return buffer;
}

static pg_attribute_always_inline bool
StartReadBuffersImpl(ReadBuffersOperation *operation,
					 Buffer *buffers,
					 BlockNumber blockNum,
					 int *nblocks,
					 int flags,
					 bool allow_forwarding)
{
	int			actual_nblocks = *nblocks;
	int			maxcombine = 0;
	bool		did_start_io;
	IOContext	io_context;
	IOObject	io_object;

	Assert(*nblocks == 1 || allow_forwarding);
	Assert(*nblocks > 0);
	Assert(*nblocks <= MAX_IO_COMBINE_LIMIT);

	/* see comments in ReadBuffer_common */
	if (operation->rel && RELATION_IS_OTHER_TEMP(operation->rel))
		ereport(ERROR,
				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
				 errmsg("cannot access temporary tables of other sessions")));

	if (operation->persistence == RELPERSISTENCE_TEMP)
	{
		io_context = IOCONTEXT_NORMAL;
		io_object = IOOBJECT_TEMP_RELATION;
	}
	else
	{
		io_context = IOContextForStrategy(operation->strategy);
		io_object = IOOBJECT_RELATION;
	}

	for (int i = 0; i < actual_nblocks; ++i)
	{
		bool		found;

		if (allow_forwarding && buffers[i] != InvalidBuffer)
		{
			BufferDesc *bufHdr;

			/*
			 * This is a buffer that was pinned by an earlier call to
			 * StartReadBuffers(), but couldn't be handled in one operation at
			 * that time.  The operation was split, and the caller has passed
			 * an already pinned buffer back to us to handle the rest of the
			 * operation.  It must continue at the expected block number.
			 */
			Assert(BufferGetBlockNumber(buffers[i]) == blockNum + i);

			/*
			 * It might be an already valid buffer (a hit) that followed the
			 * final contiguous block of an earlier I/O (a miss) marking the
			 * end of it, or a buffer that some other backend has since made
			 * valid by performing the I/O for us, in which case we can handle
			 * it as a hit now.  It is safe to check for a BM_VALID flag with
			 * a relaxed load, because we got a fresh view of it while pinning
			 * it in the previous call.
			 *
			 * On the other hand if we don't see BM_VALID yet, it must be an
			 * I/O that was split by the previous call and we need to try to
			 * start a new I/O from this block.  We're also racing against any
			 * other backend that might start the I/O or even manage to mark
			 * it BM_VALID after this check, but StartBufferIO() will handle
			 * those cases.
			 */
			if (BufferIsLocal(buffers[i]))
				bufHdr = GetLocalBufferDescriptor(-buffers[i] - 1);
			else
				bufHdr = GetBufferDescriptor(buffers[i] - 1);
			Assert(pg_atomic_read_u64(&bufHdr->state) & BM_TAG_VALID);
			found = pg_atomic_read_u64(&bufHdr->state) & BM_VALID;
		}
		else
		{
			buffers[i] = PinBufferForBlock(operation->rel,
										   operation->smgr,
										   operation->persistence,
										   operation->forknum,
										   blockNum + i,
										   operation->strategy,
										   io_object, io_context,
										   &found);
		}

		if (found)
		{
			/*
			 * We have a hit.  If it's the first block in the requested range,
			 * we can return it immediately and report that WaitReadBuffers()
			 * does not need to be called.  If the initial value of *nblocks
			 * was larger, the caller will have to call again for the rest.
			 */
			if (i == 0)
			{
				*nblocks = 1;

#ifdef USE_ASSERT_CHECKING

				/*
				 * Initialize enough of ReadBuffersOperation to make
				 * CheckReadBuffersOperation() work. Outside of assertions
				 * that's not necessary when no IO is issued.
				 */
				operation->buffers = buffers;
				operation->blocknum = blockNum;
				operation->nblocks = 1;
				operation->nblocks_done = 1;
				CheckReadBuffersOperation(operation, true);
#endif
				return false;
			}

			/*
			 * Otherwise we already have an I/O to perform, but this block
			 * can't be included as it is already valid.  Split the I/O here.
			 * There may or may not be more blocks requiring I/O after this
			 * one, we haven't checked, but they can't be contiguous with this
			 * one in the way.  We'll leave this buffer pinned, forwarding it
			 * to the next call, avoiding the need to unpin it here and re-pin
			 * it in the next call.
			 */
			actual_nblocks = i;
			break;
		}
		else
		{
			/*
			 * Check how many blocks we can cover with the same IO. The smgr
			 * implementation might e.g. be limited due to a segment boundary.
			 */
			if (i == 0 && actual_nblocks > 1)
			{
				maxcombine = smgrmaxcombine(operation->smgr,
											operation->forknum,
											blockNum);
				if (unlikely(maxcombine < actual_nblocks))
				{
					elog(DEBUG2, "limiting nblocks at %u from %u to %u",
						 blockNum, actual_nblocks, maxcombine);
					actual_nblocks = maxcombine;
				}
			}
		}
	}
	*nblocks = actual_nblocks;

	/* Populate information needed for I/O. */
	operation->buffers = buffers;
	operation->blocknum = blockNum;
	operation->flags = flags;
	operation->nblocks = actual_nblocks;
	operation->nblocks_done = 0;
	pgaio_wref_clear(&operation->io_wref);

	/*
	 * When using AIO, start the IO in the background. If not, issue prefetch
	 * requests if desired by the caller.
	 *
	 * The reason we have a dedicated path for IOMETHOD_SYNC here is to
	 * de-risk the introduction of AIO somewhat. It's a large architectural
	 * change, with lots of chances for unanticipated performance effects.
	 *
	 * Use of IOMETHOD_SYNC already leads to not actually performing IO
	 * asynchronously, but without the check here we'd execute IO earlier than
	 * we used to. Eventually this IOMETHOD_SYNC specific path should go away.
	 */
	if (io_method != IOMETHOD_SYNC)
	{
		/*
		 * Try to start IO asynchronously. It's possible that no IO needs to
		 * be started, if another backend already performed the IO.
		 *
		 * Note that if an IO is started, it might not cover the entire
		 * requested range, e.g. because an intermediary block has been read
		 * in by another backend.  In that case any "trailing" buffers we
		 * already pinned above will be "forwarded" by read_stream.c to the
		 * next call to StartReadBuffers().
		 *
		 * This is signalled to the caller by decrementing *nblocks *and*
		 * reducing operation->nblocks. The latter is done here, but not below
		 * WaitReadBuffers(), as in WaitReadBuffers() we can't "shorten" the
		 * overall read size anymore, we need to retry until done in its
		 * entirety or until failed.
		 */
		did_start_io = AsyncReadBuffers(operation, nblocks);

		operation->nblocks = *nblocks;
	}
	else
	{
		operation->flags |= READ_BUFFERS_SYNCHRONOUSLY;

		if (flags & READ_BUFFERS_ISSUE_ADVICE)
		{
			/*
			 * In theory we should only do this if PinBufferForBlock() had to
			 * allocate new buffers above.  That way, if two calls to
			 * StartReadBuffers() were made for the same blocks before
			 * WaitReadBuffers(), only the first would issue the advice.
			 * That'd be a better simulation of true asynchronous I/O, which
			 * would only start the I/O once, but isn't done here for
			 * simplicity.
			 */
			smgrprefetch(operation->smgr,
						 operation->forknum,
						 blockNum,
						 actual_nblocks);
		}

		/*
		 * Indicate that WaitReadBuffers() should be called. WaitReadBuffers()
		 * will initiate the necessary IO.
		 */
		did_start_io = true;
	}

	CheckReadBuffersOperation(operation, !did_start_io);

	return did_start_io;
}

/*
 * Begin reading a range of blocks beginning at blockNum and extending for
 * *nblocks.  *nblocks and the buffers array are in/out parameters.  On entry,
 * the buffers elements covered by *nblocks must hold either InvalidBuffer or
 * buffers forwarded by an earlier call to StartReadBuffers() that was split
 * and is now being continued.  On return, *nblocks holds the number of blocks
 * accepted by this operation.  If it is less than the original number then
 * this operation has been split, but buffer elements up to the original
 * requested size may hold forwarded buffers to be used for a continuing
 * operation.  The caller must either start a new I/O beginning at the block
 * immediately following the blocks accepted by this call and pass those
 * buffers back in, or release them if it chooses not to.  It shouldn't make
 * any other use of or assumptions about forwarded buffers.
 *
 * If false is returned, no I/O is necessary and the buffers covered by
 * *nblocks on exit are valid and ready to be accessed.  If true is returned,
 * an I/O has been started, and WaitReadBuffers() must be called with the same
 * operation object before the buffers covered by *nblocks on exit can be
 * accessed.  Along with the operation object, the caller-supplied array of
 * buffers must remain valid until WaitReadBuffers() is called, and any
 * forwarded buffers must also be preserved for a continuing call unless
 * they are explicitly released.
 */
bool
StartReadBuffers(ReadBuffersOperation *operation,
				 Buffer *buffers,
				 BlockNumber blockNum,
				 int *nblocks,
				 int flags)
{
	return StartReadBuffersImpl(operation, buffers, blockNum, nblocks, flags,
								true /* expect forwarded buffers */ );
}

/*
 * Single block version of the StartReadBuffers().  This might save a few
 * instructions when called from another translation unit, because it is
 * specialized for nblocks == 1.
 *
 * This version does not support "forwarded" buffers: they cannot be created
 * by reading only one block and *buffer is ignored on entry.
 */
bool
StartReadBuffer(ReadBuffersOperation *operation,
				Buffer *buffer,
				BlockNumber blocknum,
				int flags)
{
	int			nblocks = 1;
	bool		result;

	result = StartReadBuffersImpl(operation, buffer, blocknum, &nblocks, flags,
								  false /* single block, no forwarding */ );
	Assert(nblocks == 1);		/* single block can't be short */

	return result;
}

/*
 * Perform sanity checks on the ReadBuffersOperation.
 */
static void
CheckReadBuffersOperation(ReadBuffersOperation *operation, bool is_complete)
{
#ifdef USE_ASSERT_CHECKING
	Assert(operation->nblocks_done <= operation->nblocks);
	Assert(!is_complete || operation->nblocks == operation->nblocks_done);

	for (int i = 0; i < operation->nblocks; i++)
	{
		Buffer		buffer = operation->buffers[i];
		BufferDesc *buf_hdr = BufferIsLocal(buffer) ?
			GetLocalBufferDescriptor(-buffer - 1) :
			GetBufferDescriptor(buffer - 1);

		Assert(BufferGetBlockNumber(buffer) == operation->blocknum + i);
		Assert(pg_atomic_read_u64(&buf_hdr->state) & BM_TAG_VALID);

		if (i < operation->nblocks_done)
			Assert(pg_atomic_read_u64(&buf_hdr->state) & BM_VALID);
	}
#endif
}

/*
 * We track various stats related to buffer hits. Because this is done in a
 * few separate places, this helper exists for convenience.
 */
static pg_attribute_always_inline void
TrackBufferHit(IOObject io_object, IOContext io_context,
			   Relation rel, char persistence, SMgrRelation smgr,
			   ForkNumber forknum, BlockNumber blocknum)
{
	TRACE_POSTGRESQL_BUFFER_READ_DONE(forknum,
									  blocknum,
									  smgr->smgr_rlocator.locator.spcOid,
									  smgr->smgr_rlocator.locator.dbOid,
									  smgr->smgr_rlocator.locator.relNumber,
									  smgr->smgr_rlocator.backend,
									  true);

	if (persistence == RELPERSISTENCE_TEMP)
		pgBufferUsage.local_blks_hit += 1;
	else
		pgBufferUsage.shared_blks_hit += 1;

	pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);

	if (VacuumCostActive)
		VacuumCostBalance += VacuumCostPageHit;

	if (rel)
		pgstat_count_buffer_hit(rel);
}

/*
 * Helper for WaitReadBuffers() that processes the results of a readv
 * operation, raising an error if necessary.
 */
static void
ProcessReadBuffersResult(ReadBuffersOperation *operation)
{
	PgAioReturn *aio_ret = &operation->io_return;
	PgAioResultStatus rs = aio_ret->result.status;
	int			newly_read_blocks = 0;

	Assert(pgaio_wref_valid(&operation->io_wref));
	Assert(aio_ret->result.status != PGAIO_RS_UNKNOWN);

	/*
	 * SMGR reports the number of blocks successfully read as the result of
	 * the IO operation. Thus we can simply add that to ->nblocks_done.
	 */

	if (likely(rs != PGAIO_RS_ERROR))
		newly_read_blocks = aio_ret->result.result;

	if (rs == PGAIO_RS_ERROR || rs == PGAIO_RS_WARNING)
		pgaio_result_report(aio_ret->result, &aio_ret->target_data,
							rs == PGAIO_RS_ERROR ? ERROR : WARNING);
	else if (aio_ret->result.status == PGAIO_RS_PARTIAL)
	{
		/*
		 * We'll retry, so we just emit a debug message to the server log (or
		 * not even that in prod scenarios).
		 */
		pgaio_result_report(aio_ret->result, &aio_ret->target_data, DEBUG1);
		elog(DEBUG3, "partial read, will retry");
	}

	Assert(newly_read_blocks > 0);
	Assert(newly_read_blocks <= MAX_IO_COMBINE_LIMIT);

	operation->nblocks_done += newly_read_blocks;

	Assert(operation->nblocks_done <= operation->nblocks);
}

/*
 * Wait for the IO operation initiated by StartReadBuffers() et al to
 * complete.
 *
 * Returns true if we needed to wait for the IO operation, false otherwise.
 */
bool
WaitReadBuffers(ReadBuffersOperation *operation)
{
	PgAioReturn *aio_ret = &operation->io_return;
	IOContext	io_context;
	IOObject	io_object;
	bool		needed_wait = false;

	if (operation->persistence == RELPERSISTENCE_TEMP)
	{
		io_context = IOCONTEXT_NORMAL;
		io_object = IOOBJECT_TEMP_RELATION;
	}
	else
	{
		io_context = IOContextForStrategy(operation->strategy);
		io_object = IOOBJECT_RELATION;
	}

	/*
	 * If we get here without an IO operation having been issued, the
	 * io_method == IOMETHOD_SYNC path must have been used. Otherwise the
	 * caller should not have called WaitReadBuffers().
	 *
	 * In the case of IOMETHOD_SYNC, we start - as we used to before the
	 * introducing of AIO - the IO in WaitReadBuffers(). This is done as part
	 * of the retry logic below, no extra code is required.
	 *
	 * This path is expected to eventually go away.
	 */
	if (!pgaio_wref_valid(&operation->io_wref) && io_method != IOMETHOD_SYNC)
		elog(ERROR, "waiting for read operation that didn't read");

	/*
	 * To handle partial reads, and IOMETHOD_SYNC, we re-issue IO until we're
	 * done. We may need multiple retries, not just because we could get
	 * multiple partial reads, but also because some of the remaining
	 * to-be-read buffers may have been read in by other backends, limiting
	 * the IO size.
	 */
	while (true)
	{
		int			ignored_nblocks_progress;

		CheckReadBuffersOperation(operation, false);

		/*
		 * If there is an IO associated with the operation, we may need to
		 * wait for it.
		 */
		if (pgaio_wref_valid(&operation->io_wref))
		{
			/*
			 * Track the time spent waiting for the IO to complete. As
			 * tracking a wait even if we don't actually need to wait
			 *
			 * a) is not cheap, due to the timestamping overhead
			 *
			 * b) reports some time as waiting, even if we never waited
			 *
			 * we first check if we already know the IO is complete.
			 *
			 * Note that operation->io_return is uninitialized for foreign IO,
			 * so we cannot use the cheaper PGAIO_RS_UNKNOWN pre-check.
			 */
			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
				!pgaio_wref_check_done(&operation->io_wref))
			{
				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);

				pgaio_wref_wait(&operation->io_wref);
				needed_wait = true;

				/*
				 * The IO operation itself was already counted earlier, in
				 * AsyncReadBuffers(), this just accounts for the wait time.
				 */
				pgstat_count_io_op_time(io_object, io_context, IOOP_READ,
										io_start, 0, 0);
			}
			else
			{
				Assert(pgaio_wref_check_done(&operation->io_wref));
			}

			if (unlikely(operation->foreign_io))
			{
				Buffer		buffer = operation->buffers[operation->nblocks_done];
				BufferDesc *desc = BufferIsLocal(buffer) ?
					GetLocalBufferDescriptor(-buffer - 1) :
					GetBufferDescriptor(buffer - 1);
				uint64		buf_state = pg_atomic_read_u64(&desc->state);

				if (buf_state & BM_VALID)
				{
					BlockNumber blocknum = operation->blocknum + operation->nblocks_done;

					operation->nblocks_done += 1;
					Assert(operation->nblocks_done <= operation->nblocks);

					/*
					 * Track this as a 'hit' for this backend. The backend
					 * performing the IO will track it as a 'read'.
					 */
					TrackBufferHit(io_object, io_context,
								   operation->rel, operation->persistence,
								   operation->smgr, operation->forknum,
								   blocknum);
				}

				/*
				 * If the foreign IO failed and left the buffer invalid,
				 * nblocks_done is not incremented. The retry loop below will
				 * call AsyncReadBuffers() which will attempt the IO itself.
				 */
			}
			else
			{
				/*
				 * We now are sure the IO completed. Check the results. This
				 * includes reporting on errors if there were any.
				 */
				ProcessReadBuffersResult(operation);
			}
		}

		/*
		 * Most of the time, the one IO we already started, will read in
		 * everything.  But we need to deal with partial reads and buffers not
		 * needing IO anymore.
		 */
		if (operation->nblocks_done == operation->nblocks)
			break;

		CHECK_FOR_INTERRUPTS();

		/*
		 * If the IO completed only partially, we need to perform additional
		 * work, consider that a form of having had to wait.
		 */
		needed_wait = true;

		/*
		 * This may only complete the IO partially, either because some
		 * buffers were already valid, or because of a partial read.
		 *
		 * NB: In contrast to after the AsyncReadBuffers() call in
		 * StartReadBuffers(), we do *not* reduce
		 * ReadBuffersOperation->nblocks here, callers expect the full
		 * operation to be completed at this point (as more operations may
		 * have been queued).
		 */
		AsyncReadBuffers(operation, &ignored_nblocks_progress);
	}

	CheckReadBuffersOperation(operation, true);

	/* NB: READ_DONE tracepoint was already executed in completion callback */
	return needed_wait;
}

/*
 * Initiate IO for the ReadBuffersOperation
 *
 * This function only starts a single IO at a time. The size of the IO may be
 * limited to below the to-be-read blocks, if one of the buffers has
 * concurrently been read in. If the first to-be-read buffer is already valid,
 * no IO will be issued.
 *
 * To support retries after partial reads, the first operation->nblocks_done
 * buffers are skipped.
 *
 * On return *nblocks_progress is updated to reflect the number of buffers
 * affected by the call. If the first buffer is valid, *nblocks_progress is
 * set to 1 and operation->nblocks_done is incremented.
 *
 * Returns true if IO was initiated or is already in progress (foreign IO),
 * false if the buffer was already valid.
 */
static bool
AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
{
	Buffer	   *buffers = &operation->buffers[0];
	int			flags = operation->flags;
	ForkNumber	forknum = operation->forknum;
	char		persistence = operation->persistence;
	int16		nblocks_done = operation->nblocks_done;
	BlockNumber blocknum = operation->blocknum + nblocks_done;
	Buffer	   *io_buffers = &operation->buffers[nblocks_done];
	int			io_buffers_len = 0;
	PgAioHandle *ioh;
	uint32		ioh_flags = 0;
	void	   *io_pages[MAX_IO_COMBINE_LIMIT];
	IOContext	io_context;
	IOObject	io_object;
	instr_time	io_start;
	StartBufferIOResult status;

	if (persistence == RELPERSISTENCE_TEMP)
	{
		io_context = IOCONTEXT_NORMAL;
		io_object = IOOBJECT_TEMP_RELATION;
	}
	else
	{
		io_context = IOContextForStrategy(operation->strategy);
		io_object = IOOBJECT_RELATION;
	}

	/*
	 * When this IO is executed synchronously, either because the caller will
	 * immediately block waiting for the IO or because IOMETHOD_SYNC is used,
	 * the AIO subsystem needs to know.
	 */
	if (flags & READ_BUFFERS_SYNCHRONOUSLY)
		ioh_flags |= PGAIO_HF_SYNCHRONOUS;

	if (persistence == RELPERSISTENCE_TEMP)
		ioh_flags |= PGAIO_HF_REFERENCES_LOCAL;

	/*
	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
	 * set globally, but on a per-session basis. The completion callback,
	 * which may be run in other processes, e.g. in IO workers, may have a
	 * different value of the zero_damaged_pages GUC.
	 *
	 * XXX: We probably should eventually use a different flag for
	 * zero_damaged_pages, so we can report different log levels / error codes
	 * for zero_damaged_pages and ZERO_ON_ERROR.
	 */
	if (zero_damaged_pages)
		flags |= READ_BUFFERS_ZERO_ON_ERROR;

	/*
	 * For the same reason as with zero_damaged_pages we need to use this
	 * backend's ignore_checksum_failure value.
	 */
	if (ignore_checksum_failure)
		flags |= READ_BUFFERS_IGNORE_CHECKSUM_FAILURES;


	/*
	 * To be allowed to report stats in the local completion callback we need
	 * to prepare to report stats now. This ensures we can safely report the
	 * checksum failure even in a critical section.
	 */
	pgstat_prepare_report_checksum_failure(operation->smgr->smgr_rlocator.locator.dbOid);

	/*
	 * We must get an IO handle before StartBufferIO(), as pgaio_io_acquire()
	 * might block, which we don't want after setting IO_IN_PROGRESS. If we
	 * don't need to do the IO, we'll release the handle.
	 *
	 * If we need to wait for IO before we can get a handle, submit
	 * already-staged IO first, so that other backends don't need to wait.
	 * There wouldn't be a deadlock risk, as pgaio_io_acquire() just needs to
	 * wait for already submitted IO, which doesn't require additional locks,
	 * but it could still cause undesirable waits.
	 *
	 * A secondary benefit is that this would allow us to measure the time in
	 * pgaio_io_acquire() without causing undue timer overhead in the common,
	 * non-blocking, case.  However, currently the pgstats infrastructure
	 * doesn't really allow that, as it a) asserts that an operation can't
	 * have time without operations b) doesn't have an API to report
	 * "accumulated" time.
	 */
	ioh = pgaio_io_acquire_nb(CurrentResourceOwner, &operation->io_return);
	if (unlikely(!ioh))
	{
		pgaio_submit_staged();
		ioh = pgaio_io_acquire(CurrentResourceOwner, &operation->io_return);
	}

	operation->foreign_io = false;
	pgaio_wref_clear(&operation->io_wref);

	/*
	 * Try to start IO on the first buffer in a new run of blocks. If AIO is
	 * in progress, be it in this backend or another backend, we just
	 * associate the wait reference with the operation and wait in
	 * WaitReadBuffers(). This turns out to be important for performance in
	 * two workloads:
	 *
	 * 1) A read stream that has to read the same block multiple times within
	 * the readahead distance. This can happen e.g. for the table accesses of
	 * an index scan.
	 *
	 * 2) Concurrent scans by multiple backends on the same relation.
	 *
	 * If we were to synchronously wait for the in-progress IO, we'd not be
	 * able to keep enough I/O in flight.
	 *
	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
	 *
	 * It's possible that another backend has started IO on the buffer but not
	 * yet set its wait reference. In this case, we have no choice but to wait
	 * for either the wait reference to be valid or the IO to be done.
	 */
	status = StartBufferIO(buffers[nblocks_done], true, true,
						   &operation->io_wref);
	if (status != BUFFER_IO_READY_FOR_IO)
	{
		pgaio_io_release(ioh);
		*nblocks_progress = 1;
		if (status == BUFFER_IO_ALREADY_DONE)
		{
			/*
			 * Someone has already completed this block, we're done.
			 *
			 * When IO is necessary, ->nblocks_done is updated in
			 * ProcessReadBuffersResult(), but that is not called if no IO is
			 * necessary. Thus update here.
			 */
			operation->nblocks_done += 1;
			Assert(operation->nblocks_done <= operation->nblocks);

			Assert(!pgaio_wref_valid(&operation->io_wref));

			/*
			 * Report and track this as a 'hit' for this backend, even though
			 * it must have started out as a miss in PinBufferForBlock(). The
			 * other backend will track this as a 'read'.
			 */
			TrackBufferHit(io_object, io_context,
						   operation->rel, operation->persistence,
						   operation->smgr, operation->forknum,
						   blocknum);
			return false;
		}

		/* The IO is already in-progress */
		Assert(status == BUFFER_IO_IN_PROGRESS);
		Assert(pgaio_wref_valid(&operation->io_wref));
		operation->foreign_io = true;

		return true;
	}

	Assert(io_buffers[0] == buffers[nblocks_done]);
	io_pages[0] = BufferGetBlock(buffers[nblocks_done]);
	io_buffers_len = 1;

	/*
	 * NB: As little code as possible should be added between the
	 * StartBufferIO() above, the further StartBufferIO()s below and the
	 * smgrstartreadv(), as some of the buffers are now marked as
	 * IO_IN_PROGRESS and will thus cause other backends to wait.
	 */

	/*
	 * How many neighboring-on-disk blocks can we scatter-read into other
	 * buffers at the same time?  In this case we don't wait if we see an I/O
	 * already in progress (see comment above).
	 */
	for (int i = nblocks_done + 1; i < operation->nblocks; i++)
	{
		/* Must be consecutive block numbers. */
		Assert(BufferGetBlockNumber(buffers[i - 1]) ==
			   BufferGetBlockNumber(buffers[i]) - 1);

		status = StartBufferIO(buffers[i], true, false, NULL);
		if (status != BUFFER_IO_READY_FOR_IO)
			break;

		Assert(io_buffers[io_buffers_len] == buffers[i]);

		io_pages[io_buffers_len++] = BufferGetBlock(buffers[i]);
	}

	/* get a reference to wait for in WaitReadBuffers() */
	pgaio_io_get_wref(ioh, &operation->io_wref);

	/* provide the list of buffers to the completion callbacks */
	pgaio_io_set_handle_data_32(ioh, (uint32 *) io_buffers, io_buffers_len);

	pgaio_io_register_callbacks(ioh,
								persistence == RELPERSISTENCE_TEMP ?
								PGAIO_HCB_LOCAL_BUFFER_READV :
								PGAIO_HCB_SHARED_BUFFER_READV,
								flags);

	pgaio_io_set_flag(ioh, ioh_flags);

	/* ---
	 * Even though we're trying to issue IO asynchronously, track the time
	 * in smgrstartreadv():
	 * - if io_method == IOMETHOD_SYNC, we will always perform the IO
	 *   immediately
	 * - the io method might not support the IO (e.g. worker IO for a temp
	 *   table)
	 * ---
	 */
	io_start = pgstat_prepare_io_time(track_io_timing);
	smgrstartreadv(ioh, operation->smgr, forknum,
				   blocknum,
				   io_pages, io_buffers_len);
	pgstat_count_io_op_time(io_object, io_context, IOOP_READ,
							io_start, 1, io_buffers_len * BLCKSZ);

	if (persistence == RELPERSISTENCE_TEMP)
		pgBufferUsage.local_blks_read += io_buffers_len;
	else
		pgBufferUsage.shared_blks_read += io_buffers_len;

	/*
	 * Track vacuum cost when issuing IO, not after waiting for it. Otherwise
	 * we could end up issuing a lot of IO in a short timespan, despite a low
	 * cost limit.
	 */
	if (VacuumCostActive)
		VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;

	*nblocks_progress = io_buffers_len;

	return true;
}

/*
 * BufferAlloc -- subroutine for PinBufferForBlock.  Handles lookup of a shared
 *		buffer.  If no buffer exists already, selects a replacement victim and
 *		evicts the old page, but does NOT read in new page.
 *
 * "strategy" can be a buffer replacement strategy object, or NULL for
 * the default strategy.  The selected buffer's usage_count is advanced when
 * using the default strategy, but otherwise possibly not (see PinBuffer).
 *
 * The returned buffer is pinned and is already marked as holding the
 * desired page.  If it already did have the desired page, *foundPtr is
 * set true.  Otherwise, *foundPtr is set false.
 *
 * io_context is passed as an output parameter to avoid calling
 * IOContextForStrategy() when there is a shared buffers hit and no IO
 * statistics need be captured.
 *
 * No locks are held either at entry or exit.
 */
static pg_attribute_always_inline BufferDesc *
BufferAlloc(SMgrRelation smgr, char relpersistence, ForkNumber forkNum,
			BlockNumber blockNum,
			BufferAccessStrategy strategy,
			bool *foundPtr, IOContext io_context)
{
	BufferTag	newTag;			/* identity of requested block */
	uint32		newHash;		/* hash value for newTag */
	LWLock	   *newPartitionLock;	/* buffer partition lock for it */
	int			existing_buf_id;
	Buffer		victim_buffer;
	BufferDesc *victim_buf_hdr;
	uint64		victim_buf_state;
	uint64		set_bits = 0;

	/* Make sure we will have room to remember the buffer pin */
	ResourceOwnerEnlarge(CurrentResourceOwner);
	ReservePrivateRefCountEntry();

	/* create a tag so we can lookup the buffer */
	InitBufferTag(&newTag, &smgr->smgr_rlocator.locator, forkNum, blockNum);

	/* determine its hash code and partition lock ID */
	newHash = BufTableHashCode(&newTag);
	newPartitionLock = BufMappingPartitionLock(newHash);

	/* see if the block is in the buffer pool already */
	LWLockAcquire(newPartitionLock, LW_SHARED);
	existing_buf_id = BufTableLookup(&newTag, newHash);
	if (existing_buf_id >= 0)
	{
		BufferDesc *buf;
		bool		valid;

		/*
		 * Found it.  Now, pin the buffer so no one can steal it from the
		 * buffer pool, and check to see if the correct data has been loaded
		 * into the buffer.
		 */
		buf = GetBufferDescriptor(existing_buf_id);

		valid = PinBuffer(buf, strategy, false);

		/* Can release the mapping lock as soon as we've pinned it */
		LWLockRelease(newPartitionLock);

		*foundPtr = true;

		if (!valid)
		{
			/*
			 * We can only get here if (a) someone else is still reading in
			 * the page, (b) a previous read attempt failed, or (c) someone
			 * called StartReadBuffers() but not yet WaitReadBuffers().
			 */
			*foundPtr = false;
		}

		return buf;
	}

	/*
	 * Didn't find it in the buffer pool.  We'll have to initialize a new
	 * buffer.  Remember to unlock the mapping lock while doing the work.
	 */
	LWLockRelease(newPartitionLock);

	/*
	 * Acquire a victim buffer. Somebody else might try to do the same, we
	 * don't hold any conflicting locks. If so we'll have to undo our work
	 * later.
	 */
	victim_buffer = GetVictimBuffer(strategy, io_context);
	victim_buf_hdr = GetBufferDescriptor(victim_buffer - 1);

	/*
	 * Try to make a hashtable entry for the buffer under its new tag. If
	 * somebody else inserted another buffer for the tag, we'll release the
	 * victim buffer we acquired and use the already inserted one.
	 */
	LWLockAcquire(newPartitionLock, LW_EXCLUSIVE);
	existing_buf_id = BufTableInsert(&newTag, newHash, victim_buf_hdr->buf_id);
	if (existing_buf_id >= 0)
	{
		BufferDesc *existing_buf_hdr;
		bool		valid;

		/*
		 * Got a collision. Someone has already done what we were about to do.
		 * We'll just handle this as if it were found in the buffer pool in
		 * the first place.  First, give up the buffer we were planning to
		 * use.
		 *
		 * We could do this after releasing the partition lock, but then we'd
		 * have to call ResourceOwnerEnlarge() & ReservePrivateRefCountEntry()
		 * before acquiring the lock, for the rare case of such a collision.
		 */
		UnpinBuffer(victim_buf_hdr);

		/* remaining code should match code at top of routine */

		existing_buf_hdr = GetBufferDescriptor(existing_buf_id);

		valid = PinBuffer(existing_buf_hdr, strategy, false);

		/* Can release the mapping lock as soon as we've pinned it */
		LWLockRelease(newPartitionLock);

		*foundPtr = true;

		if (!valid)
		{
			/*
			 * We can only get here if (a) someone else is still reading in
			 * the page, (b) a previous read attempt failed, or (c) someone
			 * called StartReadBuffers() but not yet WaitReadBuffers().
			 */
			*foundPtr = false;
		}

		return existing_buf_hdr;
	}

	/*
	 * Need to lock the buffer header too in order to change its tag.
	 */
	victim_buf_state = LockBufHdr(victim_buf_hdr);

	/* some sanity checks while we hold the buffer header lock */
	Assert(BUF_STATE_GET_REFCOUNT(victim_buf_state) == 1);
	Assert(!(victim_buf_state & (BM_TAG_VALID | BM_VALID | BM_DIRTY | BM_IO_IN_PROGRESS)));

	victim_buf_hdr->tag = newTag;

	/*
	 * Make sure BM_PERMANENT is set for buffers that must be written at every
	 * checkpoint.  Unlogged buffers only need to be written at shutdown
	 * checkpoints, except for their "init" forks, which need to be treated
	 * just like permanent relations.
	 */
	set_bits |= BM_TAG_VALID | BUF_USAGECOUNT_ONE;
	if (relpersistence == RELPERSISTENCE_PERMANENT || forkNum == INIT_FORKNUM)
		set_bits |= BM_PERMANENT;

	UnlockBufHdrExt(victim_buf_hdr, victim_buf_state,
					set_bits, 0, 0);

	LWLockRelease(newPartitionLock);

	/*
	 * Buffer contents are currently invalid.
	 */
	*foundPtr = false;

	return victim_buf_hdr;
}

/*
 * InvalidateBuffer -- mark a shared buffer invalid.
 *
 * The buffer header spinlock must be held at entry.  We drop it before
 * returning.  (This is sane because the caller must have locked the
 * buffer in order to be sure it should be dropped.)
 *
 * This is used only in contexts such as dropping a relation.  We assume
 * that no other backend could possibly be interested in using the page,
 * so the only reason the buffer might be pinned is if someone else is
 * trying to write it out.  We have to let them finish before we can
 * reclaim the buffer.
 *
 * The buffer could get reclaimed by someone else while we are waiting
 * to acquire the necessary locks; if so, don't mess it up.
 */
static void
InvalidateBuffer(BufferDesc *buf)
{
	BufferTag	oldTag;
	uint32		oldHash;		/* hash value for oldTag */
	LWLock	   *oldPartitionLock;	/* buffer partition lock for it */
	uint32		oldFlags;
	uint64		buf_state;

	/* Save the original buffer tag before dropping the spinlock */
	oldTag = buf->tag;

	UnlockBufHdr(buf);

	/*
	 * Need to compute the old tag's hashcode and partition lock ID. XXX is it
	 * worth storing the hashcode in BufferDesc so we need not recompute it
	 * here?  Probably not.
	 */
	oldHash = BufTableHashCode(&oldTag);
	oldPartitionLock = BufMappingPartitionLock(oldHash);

retry:

	/*
	 * Acquire exclusive mapping lock in preparation for changing the buffer's
	 * association.
	 */
	LWLockAcquire(oldPartitionLock, LW_EXCLUSIVE);

	/* Re-lock the buffer header */
	buf_state = LockBufHdr(buf);

	/* If it's changed while we were waiting for lock, do nothing */
	if (!BufferTagsEqual(&buf->tag, &oldTag))
	{
		UnlockBufHdr(buf);
		LWLockRelease(oldPartitionLock);
		return;
	}

	/*
	 * We assume the reason for it to be pinned is that either we were
	 * asynchronously reading the page in before erroring out or someone else
	 * is flushing the page out.  Wait for the IO to finish.  (This could be
	 * an infinite loop if the refcount is messed up... it would be nice to
	 * time out after awhile, but there seems no way to be sure how many loops
	 * may be needed.  Note that if the other guy has pinned the buffer but
	 * not yet done StartBufferIO, WaitIO will fall through and we'll
	 * effectively be busy-looping here.)
	 */
	if (BUF_STATE_GET_REFCOUNT(buf_state) != 0)
	{
		UnlockBufHdr(buf);
		LWLockRelease(oldPartitionLock);
		/* safety check: should definitely not be our *own* pin */
		if (GetPrivateRefCount(BufferDescriptorGetBuffer(buf)) > 0)
			elog(ERROR, "buffer is pinned in InvalidateBuffer");
		WaitIO(buf);
		goto retry;
	}

	/*
	 * An invalidated buffer should not have any backends waiting to lock the
	 * buffer, therefore BM_LOCK_WAKE_IN_PROGRESS should not be set.
	 */
	Assert(!(buf_state & BM_LOCK_WAKE_IN_PROGRESS));

	/*
	 * Clear out the buffer's tag and flags.  We must do this to ensure that
	 * linear scans of the buffer array don't think the buffer is valid.
	 */
	oldFlags = buf_state & BUF_FLAG_MASK;
	ClearBufferTag(&buf->tag);

	UnlockBufHdrExt(buf, buf_state,
					0,
					BUF_FLAG_MASK | BUF_USAGECOUNT_MASK,
					0);

	/*
	 * Remove the buffer from the lookup hashtable, if it was in there.
	 */
	if (oldFlags & BM_TAG_VALID)
		BufTableDelete(&oldTag, oldHash);

	/*
	 * Done with mapping lock.
	 */
	LWLockRelease(oldPartitionLock);
}

/*
 * Helper routine for GetVictimBuffer()
 *
 * Needs to be called on a buffer with a valid tag, pinned, but without the
 * buffer header spinlock held.
 *
 * Returns true if the buffer can be reused, in which case the buffer is only
 * pinned by this backend and marked as invalid, false otherwise.
 */
static bool
InvalidateVictimBuffer(BufferDesc *buf_hdr)
{
	uint64		buf_state;
	uint32		hash;
	LWLock	   *partition_lock;
	BufferTag	tag;

	Assert(GetPrivateRefCount(BufferDescriptorGetBuffer(buf_hdr)) == 1);

	/* have buffer pinned, so it's safe to read tag without lock */
	tag = buf_hdr->tag;

	hash = BufTableHashCode(&tag);
	partition_lock = BufMappingPartitionLock(hash);

	LWLockAcquire(partition_lock, LW_EXCLUSIVE);

	/* lock the buffer header */
	buf_state = LockBufHdr(buf_hdr);

	/*
	 * We have the buffer pinned nobody else should have been able to unset
	 * this concurrently.
	 */
	Assert(buf_state & BM_TAG_VALID);
	Assert(BUF_STATE_GET_REFCOUNT(buf_state) > 0);
	Assert(BufferTagsEqual(&buf_hdr->tag, &tag));

	/*
	 * If somebody else pinned the buffer since, or even worse, dirtied it,
	 * give up on this buffer: It's clearly in use.
	 */
	if (BUF_STATE_GET_REFCOUNT(buf_state) != 1 || (buf_state & BM_DIRTY))
	{
		Assert(BUF_STATE_GET_REFCOUNT(buf_state) > 0);

		UnlockBufHdr(buf_hdr);
		LWLockRelease(partition_lock);

		return false;
	}

	/*
	 * An invalidated buffer should not have any backends waiting to lock the
	 * buffer, therefore BM_LOCK_WAKE_IN_PROGRESS should not be set.
	 */
	Assert(!(buf_state & BM_LOCK_WAKE_IN_PROGRESS));

	/*
	 * Clear out the buffer's tag and flags and usagecount.  This is not
	 * strictly required, as BM_TAG_VALID/BM_VALID needs to be checked before
	 * doing anything with the buffer. But currently it's beneficial, as the
	 * cheaper pre-check for several linear scans of shared buffers use the
	 * tag (see e.g. FlushDatabaseBuffers()).
	 */
	ClearBufferTag(&buf_hdr->tag);
	UnlockBufHdrExt(buf_hdr, buf_state,
					0,
					BUF_FLAG_MASK | BUF_USAGECOUNT_MASK,
					0);

	Assert(BUF_STATE_GET_REFCOUNT(buf_state) > 0);

	/* finally delete buffer from the buffer mapping table */
	BufTableDelete(&tag, hash);

	LWLockRelease(partition_lock);

	buf_state = pg_atomic_read_u64(&buf_hdr->state);
	Assert(!(buf_state & (BM_DIRTY | BM_VALID | BM_TAG_VALID)));
	Assert(BUF_STATE_GET_REFCOUNT(buf_state) > 0);
	Assert(BUF_STATE_GET_REFCOUNT(pg_atomic_read_u64(&buf_hdr->state)) > 0);

	return true;
}

static Buffer
GetVictimBuffer(BufferAccessStrategy strategy, IOContext io_context)
{
	BufferDesc *buf_hdr;
	Buffer		buf;
	uint64		buf_state;
	bool		from_ring;

	/*
	 * Ensure, before we pin a victim buffer, that there's a free refcount
	 * entry and resource owner slot for the pin.
	 */
	ReservePrivateRefCountEntry();
	ResourceOwnerEnlarge(CurrentResourceOwner);

	/* we return here if a prospective victim buffer gets used concurrently */
again:

	/*
	 * Select a victim buffer.  The buffer is returned pinned and owned by
	 * this backend.
	 */
	buf_hdr = StrategyGetBuffer(strategy, &buf_state, &from_ring);
	buf = BufferDescriptorGetBuffer(buf_hdr);

	/*
	 * We shouldn't have any other pins for this buffer.
	 */
	CheckBufferIsPinnedOnce(buf);

	/*
	 * If the buffer was dirty, try to write it out.  There is a race
	 * condition here, another backend could dirty the buffer between
	 * StrategyGetBuffer() checking that it is not in use and invalidating the
	 * buffer below. That's addressed by InvalidateVictimBuffer() verifying
	 * that the buffer is not dirty.
	 */
	if (buf_state & BM_DIRTY)
	{
		Assert(buf_state & BM_TAG_VALID);
		Assert(buf_state & BM_VALID);

		/*
		 * We need a share-exclusive lock on the buffer contents to write it
		 * out (else we might write invalid data, eg because someone else is
		 * compacting the page contents while we write).  We must use a
		 * conditional lock acquisition here to avoid deadlock.  Even though
		 * the buffer was not pinned (and therefore surely not locked) when
		 * StrategyGetBuffer returned it, someone else could have pinned and
		 * (share-)exclusive-locked it by the time we get here. If we try to
		 * get the lock unconditionally, we'd block waiting for them; if they
		 * later block waiting for us, deadlock ensues. (This has been
		 * observed to happen when two backends are both trying to split btree
		 * index pages, and the second one just happens to be trying to split
		 * the page the first one got from StrategyGetBuffer.)
		 */
		if (!BufferLockConditional(buf, buf_hdr, BUFFER_LOCK_SHARE_EXCLUSIVE))
		{
			/*
			 * Someone else has locked the buffer, so give it up and loop back
			 * to get another one.
			 */
			UnpinBuffer(buf_hdr);
			goto again;
		}

		/*
		 * If using a nondefault strategy, and this victim came from the
		 * strategy ring, let the strategy decide whether to reject it when
		 * reusing it would require a WAL flush.  This only applies to
		 * permanent buffers; unlogged buffers can have fake LSNs, so
		 * XLogNeedsFlush() is not meaningful for them.
		 *
		 * We need to hold the content lock in at least share-exclusive mode
		 * to safely inspect the page LSN, so this couldn't have been done
		 * inside StrategyGetBuffer().
		 */
		if (strategy && from_ring &&
			buf_state & BM_PERMANENT &&
			XLogNeedsFlush(BufferGetLSN(buf_hdr)) &&
			StrategyRejectBuffer(strategy, buf_hdr, from_ring))
		{
			UnlockReleaseBuffer(buf);
			goto again;
		}

		/* OK, do the I/O */
		FlushBuffer(buf_hdr, NULL, IOOBJECT_RELATION, io_context);
		LockBuffer(buf, BUFFER_LOCK_UNLOCK);

		ScheduleBufferTagForWriteback(&BackendWritebackContext, io_context,
									  &buf_hdr->tag);
	}


	if (buf_state & BM_VALID)
	{
		/*
		 * When a BufferAccessStrategy is in use, blocks evicted from shared
		 * buffers are counted as IOOP_EVICT in the corresponding context
		 * (e.g. IOCONTEXT_BULKWRITE). Shared buffers are evicted by a
		 * strategy in two cases: 1) while initially claiming buffers for the
		 * strategy ring 2) to replace an existing strategy ring buffer
		 * because it is pinned or in use and cannot be reused.
		 *
		 * Blocks evicted from buffers already in the strategy ring are
		 * counted as IOOP_REUSE in the corresponding strategy context.
		 *
		 * At this point, we can accurately count evictions and reuses,
		 * because we have successfully claimed the valid buffer. Previously,
		 * we may have been forced to release the buffer due to concurrent
		 * pinners or erroring out.
		 */
		pgstat_count_io_op(IOOBJECT_RELATION, io_context,
						   from_ring ? IOOP_REUSE : IOOP_EVICT, 1, 0);
	}

	/*
	 * If the buffer has an entry in the buffer mapping table, delete it. This
	 * can fail because another backend could have pinned or dirtied the
	 * buffer.
	 */
	if ((buf_state & BM_TAG_VALID) && !InvalidateVictimBuffer(buf_hdr))
	{
		UnpinBuffer(buf_hdr);
		goto again;
	}

	/* a final set of sanity checks */
#ifdef USE_ASSERT_CHECKING
	buf_state = pg_atomic_read_u64(&buf_hdr->state);

	Assert(BUF_STATE_GET_REFCOUNT(buf_state) == 1);
	Assert(!(buf_state & (BM_TAG_VALID | BM_VALID | BM_DIRTY)));

	CheckBufferIsPinnedOnce(buf);
#endif

	return buf;
}

/*
 * Return the maximum number of buffers that a backend should try to pin once,
 * to avoid exceeding its fair share.  This is the highest value that
 * GetAdditionalPinLimit() could ever return.  Note that it may be zero on a
 * system with a very small buffer pool relative to max_connections.
 */
uint32
GetPinLimit(void)
{
	return MaxProportionalPins;
}

/*
 * Return the maximum number of additional buffers that this backend should
 * pin if it wants to stay under the per-backend limit, considering the number
 * of buffers it has already pinned.  Unlike LimitAdditionalPins(), the limit
 * return by this function can be zero.
 */
uint32
GetAdditionalPinLimit(void)
{
	uint32		estimated_pins_held;

	/*
	 * We get the number of "overflowed" pins for free, but don't know the
	 * number of pins in PrivateRefCountArray.  The cost of calculating that
	 * exactly doesn't seem worth it, so just assume the max.
	 */
	estimated_pins_held = PrivateRefCountOverflowed + REFCOUNT_ARRAY_ENTRIES;

	/* Is this backend already holding more than its fair share? */
	if (estimated_pins_held > MaxProportionalPins)
		return 0;

	return MaxProportionalPins - estimated_pins_held;
}

/*
 * Limit the number of pins a batch operation may additionally acquire, to
 * avoid running out of pinnable buffers.
 *
 * One additional pin is always allowed, on the assumption that the operation
 * requires at least one to make progress.
 */
void
LimitAdditionalPins(uint32 *additional_pins)
{
	uint32		limit;

	if (*additional_pins <= 1)
		return;

	limit = GetAdditionalPinLimit();
	limit = Max(limit, 1);
	if (limit < *additional_pins)
		*additional_pins = limit;
}

/*
 * Logic shared between ExtendBufferedRelBy(), ExtendBufferedRelTo(). Just to
 * avoid duplicating the tracing and relpersistence related logic.
 */
static BlockNumber
ExtendBufferedRelCommon(BufferManagerRelation bmr,
						ForkNumber fork,
						BufferAccessStrategy strategy,
						uint32 flags,
						uint32 extend_by,
						BlockNumber extend_upto,
						Buffer *buffers,
						uint32 *extended_by)
{
	BlockNumber first_block;

	TRACE_POSTGRESQL_BUFFER_EXTEND_START(fork,
										 BMR_GET_SMGR(bmr)->smgr_rlocator.locator.spcOid,
										 BMR_GET_SMGR(bmr)->smgr_rlocator.locator.dbOid,
										 BMR_GET_SMGR(bmr)->smgr_rlocator.locator.relNumber,
										 BMR_GET_SMGR(bmr)->smgr_rlocator.backend,
										 extend_by);

	if (bmr.relpersistence == RELPERSISTENCE_TEMP)
		first_block = ExtendBufferedRelLocal(bmr, fork, flags,
											 extend_by, extend_upto,
											 buffers, &extend_by);
	else
		first_block = ExtendBufferedRelShared(bmr, fork, strategy, flags,
											  extend_by, extend_upto,
											  buffers, &extend_by);
	*extended_by = extend_by;

	TRACE_POSTGRESQL_BUFFER_EXTEND_DONE(fork,
										BMR_GET_SMGR(bmr)->smgr_rlocator.locator.spcOid,
										BMR_GET_SMGR(bmr)->smgr_rlocator.locator.dbOid,
										BMR_GET_SMGR(bmr)->smgr_rlocator.locator.relNumber,
										BMR_GET_SMGR(bmr)->smgr_rlocator.backend,
										*extended_by,
										first_block);

	return first_block;
}

/*
 * Implementation of ExtendBufferedRelBy() and ExtendBufferedRelTo() for
 * shared buffers.
 */
static BlockNumber
ExtendBufferedRelShared(BufferManagerRelation bmr,
						ForkNumber fork,
						BufferAccessStrategy strategy,
						uint32 flags,
						uint32 extend_by,
						BlockNumber extend_upto,
						Buffer *buffers,
						uint32 *extended_by)
{
	BlockNumber first_block;
	IOContext	io_context = IOContextForStrategy(strategy);
	instr_time	io_start;

	LimitAdditionalPins(&extend_by);

	/*
	 * Acquire victim buffers for extension without holding extension lock.
	 * Writing out victim buffers is the most expensive part of extending the
	 * relation, particularly when doing so requires WAL flushes. Zeroing out
	 * the buffers is also quite expensive, so do that before holding the
	 * extension lock as well.
	 *
	 * These pages are pinned by us and not valid. While we hold the pin they
	 * can't be acquired as victim buffers by another backend.
	 */
	for (uint32 i = 0; i < extend_by; i++)
	{
		Block		buf_block;

		buffers[i] = GetVictimBuffer(strategy, io_context);
		buf_block = BufHdrGetBlock(GetBufferDescriptor(buffers[i] - 1));

		/* new buffers are zero-filled */
		MemSet(buf_block, 0, BLCKSZ);
	}

	/*
	 * Lock relation against concurrent extensions, unless requested not to.
	 *
	 * We use the same extension lock for all forks. That's unnecessarily
	 * restrictive, but currently extensions for forks don't happen often
	 * enough to make it worth locking more granularly.
	 *
	 * Note that another backend might have extended the relation by the time
	 * we get the lock.
	 */
	if (!(flags & EB_SKIP_EXTENSION_LOCK))
		LockRelationForExtension(bmr.rel, ExclusiveLock);

	/*
	 * If requested, invalidate size cache, so that smgrnblocks asks the
	 * kernel.
	 */
	if (flags & EB_CLEAR_SIZE_CACHE)
		BMR_GET_SMGR(bmr)->smgr_cached_nblocks[fork] = InvalidBlockNumber;

	first_block = smgrnblocks(BMR_GET_SMGR(bmr), fork);

	/*
	 * Now that we have the accurate relation size, check if the caller wants
	 * us to extend to only up to a specific size. If there were concurrent
	 * extensions, we might have acquired too many buffers and need to release
	 * them.
	 */
	if (extend_upto != InvalidBlockNumber)
	{
		uint32		orig_extend_by = extend_by;

		if (first_block > extend_upto)
			extend_by = 0;
		else if ((uint64) first_block + extend_by > extend_upto)
			extend_by = extend_upto - first_block;

		for (uint32 i = extend_by; i < orig_extend_by; i++)
		{
			BufferDesc *buf_hdr = GetBufferDescriptor(buffers[i] - 1);

			UnpinBuffer(buf_hdr);
		}

		if (extend_by == 0)
		{
			if (!(flags & EB_SKIP_EXTENSION_LOCK))
				UnlockRelationForExtension(bmr.rel, ExclusiveLock);
			*extended_by = extend_by;
			return first_block;
		}
	}

	/* Fail if relation is already at maximum possible length */
	if ((uint64) first_block + extend_by >= MaxBlockNumber)
		ereport(ERROR,
				(errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
				 errmsg("cannot extend relation %s beyond %u blocks",
						relpath(BMR_GET_SMGR(bmr)->smgr_rlocator, fork).str,
						MaxBlockNumber)));

	/*
	 * Insert buffers into buffer table, mark as IO_IN_PROGRESS.
	 *
	 * This needs to happen before we extend the relation, because as soon as
	 * we do, other backends can start to read in those pages.
	 */
	for (uint32 i = 0; i < extend_by; i++)
	{
		Buffer		victim_buf = buffers[i];
		BufferDesc *victim_buf_hdr = GetBufferDescriptor(victim_buf - 1);
		BufferTag	tag;
		uint32		hash;
		LWLock	   *partition_lock;
		int			existing_id;

		/* in case we need to pin an existing buffer below */
		ResourceOwnerEnlarge(CurrentResourceOwner);
		ReservePrivateRefCountEntry();

		InitBufferTag(&tag, &BMR_GET_SMGR(bmr)->smgr_rlocator.locator, fork,
					  first_block + i);
		hash = BufTableHashCode(&tag);
		partition_lock = BufMappingPartitionLock(hash);

		LWLockAcquire(partition_lock, LW_EXCLUSIVE);

		existing_id = BufTableInsert(&tag, hash, victim_buf_hdr->buf_id);

		/*
		 * We get here only in the corner case where we are trying to extend
		 * the relation but we found a pre-existing buffer. This can happen
		 * because a prior attempt at extending the relation failed, and
		 * because mdread doesn't complain about reads beyond EOF (when
		 * zero_damaged_pages is ON) and so a previous attempt to read a block
		 * beyond EOF could have left a "valid" zero-filled buffer.
		 *
		 * This has also been observed when relation was overwritten by
		 * external process. Since the legitimate cases should always have
		 * left a zero-filled buffer, complain if not PageIsNew.
		 */
		if (existing_id >= 0)
		{
			BufferDesc *existing_hdr = GetBufferDescriptor(existing_id);
			Block		buf_block;
			bool		valid;

			/*
			 * Pin the existing buffer before releasing the partition lock,
			 * preventing it from being evicted.
			 */
			valid = PinBuffer(existing_hdr, strategy, false);

			LWLockRelease(partition_lock);
			UnpinBuffer(victim_buf_hdr);

			buffers[i] = BufferDescriptorGetBuffer(existing_hdr);
			buf_block = BufHdrGetBlock(existing_hdr);

			if (valid && !PageIsNew((Page) buf_block))
				ereport(ERROR,
						(errmsg("unexpected data beyond EOF in block %u of relation \"%s\"",
								existing_hdr->tag.blockNum,
								relpath(BMR_GET_SMGR(bmr)->smgr_rlocator, fork).str)));

			/*
			 * We *must* do smgr[zero]extend before succeeding, else the page
			 * will not be reserved by the kernel, and the next P_NEW call
			 * will decide to return the same page.  Clear the BM_VALID bit,
			 * do StartSharedBufferIO() and proceed.
			 *
			 * Loop to handle the very small possibility that someone re-sets
			 * BM_VALID between our clearing it and StartSharedBufferIO
			 * inspecting it.
			 */
			while (true)
			{
				StartBufferIOResult sbres;

				pg_atomic_fetch_and_u64(&existing_hdr->state, ~BM_VALID);

				sbres = StartSharedBufferIO(existing_hdr, true, true, NULL);

				if (sbres != BUFFER_IO_ALREADY_DONE)
					break;
			}
		}
		else
		{
			uint64		buf_state;
			uint64		set_bits = 0;

			buf_state = LockBufHdr(victim_buf_hdr);

			/* some sanity checks while we hold the buffer header lock */
			Assert(!(buf_state & (BM_VALID | BM_TAG_VALID | BM_DIRTY)));
			Assert(BUF_STATE_GET_REFCOUNT(buf_state) == 1);

			victim_buf_hdr->tag = tag;

			set_bits |= BM_TAG_VALID | BUF_USAGECOUNT_ONE;
			if (bmr.relpersistence == RELPERSISTENCE_PERMANENT || fork == INIT_FORKNUM)
				set_bits |= BM_PERMANENT;

			UnlockBufHdrExt(victim_buf_hdr, buf_state,
							set_bits, 0,
							0);

			LWLockRelease(partition_lock);

			/* XXX: could combine the locked operations in it with the above */
			StartSharedBufferIO(victim_buf_hdr, true, true, NULL);
		}
	}

	io_start = pgstat_prepare_io_time(track_io_timing);

	/*
	 * Note: if smgrzeroextend fails, we will end up with buffers that are
	 * allocated but not marked BM_VALID.  The next relation extension will
	 * still select the same block number (because the relation didn't get any
	 * longer on disk) and so future attempts to extend the relation will find
	 * the same buffers (if they have not been recycled) but come right back
	 * here to try smgrzeroextend again.
	 *
	 * We don't need to set checksum for all-zero pages.
	 */
	smgrzeroextend(BMR_GET_SMGR(bmr), fork, first_block, extend_by, false);

	/*
	 * Release the file-extension lock; it's now OK for someone else to extend
	 * the relation some more.
	 *
	 * We remove IO_IN_PROGRESS after this, as waking up waiting backends can
	 * take noticeable time.
	 */
	if (!(flags & EB_SKIP_EXTENSION_LOCK))
		UnlockRelationForExtension(bmr.rel, ExclusiveLock);

	pgstat_count_io_op_time(IOOBJECT_RELATION, io_context, IOOP_EXTEND,
							io_start, 1, extend_by * BLCKSZ);

	/* Set BM_VALID, terminate IO, and wake up any waiters */
	for (uint32 i = 0; i < extend_by; i++)
	{
		Buffer		buf = buffers[i];
		BufferDesc *buf_hdr = GetBufferDescriptor(buf - 1);
		bool		lock = false;

		if (flags & EB_LOCK_FIRST && i == 0)
			lock = true;
		else if (flags & EB_LOCK_TARGET)
		{
			Assert(extend_upto != InvalidBlockNumber);
			if (first_block + i + 1 == extend_upto)
				lock = true;
		}

		if (lock)
			LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);

		TerminateBufferIO(buf_hdr, false, BM_VALID, true, false);
	}

	pgBufferUsage.shared_blks_written += extend_by;

	*extended_by = extend_by;

	return first_block;
}

/*
 * BufferIsLockedByMe
 *
 *      Checks if this backend has the buffer locked in any mode.
 *
 * Buffer must be pinned.
 */
bool
BufferIsLockedByMe(Buffer buffer)
{
	BufferDesc *bufHdr;

	Assert(BufferIsPinned(buffer));

	if (BufferIsLocal(buffer))
	{
		/* Content locks are not maintained for local buffers. */
		return true;
	}
	else
	{
		bufHdr = GetBufferDescriptor(buffer - 1);
		return BufferLockHeldByMe(bufHdr);
	}
}

/*
 * BufferIsLockedByMeInMode
 *
 *      Checks if this backend has the buffer locked in the specified mode.
 *
 * Buffer must be pinned.
 */
bool
BufferIsLockedByMeInMode(Buffer buffer, BufferLockMode mode)
{
	BufferDesc *bufHdr;

	Assert(BufferIsPinned(buffer));

	if (BufferIsLocal(buffer))
	{
		/* Content locks are not maintained for local buffers. */
		return true;
	}
	else
	{
		bufHdr = GetBufferDescriptor(buffer - 1);
		return BufferLockHeldByMeInMode(bufHdr, mode);
	}
}

/*
 * BufferIsDirty
 *
 *		Checks if buffer is already dirty.
 *
 * Buffer must be pinned and [share-]exclusive-locked.  (Without such a lock,
 * the result may be stale before it's returned.)
 */
bool
BufferIsDirty(Buffer buffer)
{
	BufferDesc *bufHdr;

	Assert(BufferIsPinned(buffer));

	if (BufferIsLocal(buffer))
	{
		int			bufid = -buffer - 1;

		bufHdr = GetLocalBufferDescriptor(bufid);
		/* Content locks are not maintained for local buffers. */
	}
	else
	{
		bufHdr = GetBufferDescriptor(buffer - 1);
		Assert(BufferIsLockedByMeInMode(buffer, BUFFER_LOCK_SHARE_EXCLUSIVE) ||
			   BufferIsLockedByMeInMode(buffer, BUFFER_LOCK_EXCLUSIVE));
	}

	return pg_atomic_read_u64(&bufHdr->state) & BM_DIRTY;
}

/*
 * MarkBufferDirty
 *
 *		Marks buffer contents as dirty (actual write happens later).
 *
 * Buffer must be pinned and exclusive-locked.  (If caller does not hold
 * exclusive lock, then somebody could be in process of writing the buffer,
 * leading to risk of bad data written to disk.)
 */
void
MarkBufferDirty(Buffer buffer)
{
	BufferDesc *bufHdr;
	uint64		buf_state;
	uint64		old_buf_state;

	if (!BufferIsValid(buffer))
		elog(ERROR, "bad buffer ID: %d", buffer);

	if (BufferIsLocal(buffer))
	{
		MarkLocalBufferDirty(buffer);
		return;
	}

	bufHdr = GetBufferDescriptor(buffer - 1);

	Assert(BufferIsPinned(buffer));
	Assert(BufferIsLockedByMeInMode(buffer, BUFFER_LOCK_EXCLUSIVE));

	/*
	 * NB: We have to wait for the buffer header spinlock to be not held, as
	 * TerminateBufferIO() relies on the spinlock.
	 */
	old_buf_state = pg_atomic_read_u64(&bufHdr->state);
	for (;;)
	{
		if (old_buf_state & BM_LOCKED)
			old_buf_state = WaitBufHdrUnlocked(bufHdr);

		buf_state = old_buf_state;

		Assert(BUF_STATE_GET_REFCOUNT(buf_state) > 0);
		buf_state |= BM_DIRTY;

		if (pg_atomic_compare_exchange_u64(&bufHdr->state, &old_buf_state,
										   buf_state))
			break;
	}

	/*
	 * If the buffer was not dirty already, do vacuum accounting.
	 */
	if (!(old_buf_state & BM_DIRTY))
	{
		pgBufferUsage.shared_blks_dirtied++;
		if (VacuumCostActive)
			VacuumCostBalance += VacuumCostPageDirty;
	}
}

/*
 * ReleaseAndReadBuffer -- combine ReleaseBuffer() and ReadBuffer()
 *
 * Formerly, this saved one cycle of acquiring/releasing the BufMgrLock
 * compared to calling the two routines separately.  Now it's mainly just
 * a convenience function.  However, if the passed buffer is valid and
 * already contains the desired block, we just return it as-is; and that
 * does save considerable work compared to a full release and reacquire.
 *
 * Note: it is OK to pass buffer == InvalidBuffer, indicating that no old
 * buffer actually needs to be released.  This case is the same as ReadBuffer,
 * but can save some tests in the caller.
 */
Buffer
ReleaseAndReadBuffer(Buffer buffer,
					 Relation relation,
					 BlockNumber blockNum)
{
	ForkNumber	forkNum = MAIN_FORKNUM;
	BufferDesc *bufHdr;

	if (BufferIsValid(buffer))
	{
		Assert(BufferIsPinned(buffer));
		if (BufferIsLocal(buffer))
		{
			bufHdr = GetLocalBufferDescriptor(-buffer - 1);
			if (bufHdr->tag.blockNum == blockNum &&
				BufTagMatchesRelFileLocator(&bufHdr->tag, &relation->rd_locator) &&
				BufTagGetForkNum(&bufHdr->tag) == forkNum)
				return buffer;
			UnpinLocalBuffer(buffer);
		}
		else
		{
			bufHdr = GetBufferDescriptor(buffer - 1);
			/* we have pin, so it's ok to examine tag without spinlock */
			if (bufHdr->tag.blockNum == blockNum &&
				BufTagMatchesRelFileLocator(&bufHdr->tag, &relation->rd_locator) &&
				BufTagGetForkNum(&bufHdr->tag) == forkNum)
				return buffer;
			UnpinBuffer(bufHdr);
		}
	}

	return ReadBuffer(relation, blockNum);
}

/*
 * PinBuffer -- make buffer unavailable for replacement.
 *
 * For the default access strategy, the buffer's usage_count is incremented
 * when we first pin it; for other strategies we just make sure the usage_count
 * isn't zero.  (The idea of the latter is that we don't want synchronized
 * heap scans to inflate the count, but we need it to not be zero to discourage
 * other backends from stealing buffers from our ring.  As long as we cycle
 * through the ring faster than the global clock-sweep cycles, buffers in
 * our ring won't be chosen as victims for replacement by other backends.)
 *
 * This should be applied only to shared buffers, never local ones.
 *
 * Since buffers are pinned/unpinned very frequently, pin buffers without
 * taking the buffer header lock; instead update the state variable in loop of
 * CAS operations. Hopefully it's just a single CAS.
 *
 * Note that ResourceOwnerEnlarge() and ReservePrivateRefCountEntry()
 * must have been done already.
 *
 * Returns true if buffer is BM_VALID, else false.  This provision allows
 * some callers to avoid an extra spinlock cycle.  If skip_if_not_valid is
 * true, then a false return value also indicates that the buffer was
 * (recently) invalid and has not been pinned.
 */
static bool
PinBuffer(BufferDesc *buf, BufferAccessStrategy strategy,
		  bool skip_if_not_valid)
{
	Buffer		b = BufferDescriptorGetBuffer(buf);
	bool		result;
	PrivateRefCountEntry *ref;

	Assert(!BufferIsLocal(b));
	Assert(ReservedRefCountSlot != -1);

	ref = GetPrivateRefCountEntry(b, true);

	if (ref == NULL)
	{
		uint64		buf_state;
		uint64		old_buf_state;

		old_buf_state = pg_atomic_read_u64(&buf->state);
		for (;;)
		{
			if (unlikely(skip_if_not_valid && !(old_buf_state & BM_VALID)))
				return false;

			/*
			 * We're not allowed to increase the refcount while the buffer
			 * header spinlock is held. Wait for the lock to be released.
			 */
			if (unlikely(old_buf_state & BM_LOCKED))
			{
				old_buf_state = WaitBufHdrUnlocked(buf);

				/* perform checks at the top of the loop again */
				continue;
			}

			buf_state = old_buf_state;

			/* increase refcount */
			buf_state += BUF_REFCOUNT_ONE;

			if (strategy == NULL)
			{
				/* Default case: increase usagecount unless already max. */
				if (BUF_STATE_GET_USAGECOUNT(buf_state) < BM_MAX_USAGE_COUNT)
					buf_state += BUF_USAGECOUNT_ONE;
			}
			else
			{
				/*
				 * Ring buffers shouldn't evict others from pool.  Thus we
				 * don't make usagecount more than 1.
				 */
				if (BUF_STATE_GET_USAGECOUNT(buf_state) == 0)
					buf_state += BUF_USAGECOUNT_ONE;
			}

			if (pg_atomic_compare_exchange_u64(&buf->state, &old_buf_state,
											   buf_state))
			{
				result = (buf_state & BM_VALID) != 0;

				TrackNewBufferPin(b);
				break;
			}
		}
	}
	else
	{
		/*
		 * If we previously pinned the buffer, it is likely to be valid, but
		 * it may not be if StartReadBuffers() was called and
		 * WaitReadBuffers() hasn't been called yet.  We'll check by loading
		 * the flags without locking.  This is racy, but it's OK to return
		 * false spuriously: when WaitReadBuffers() calls StartBufferIO(),
		 * it'll see that it's now valid.
		 *
		 * Note: We deliberately avoid a Valgrind client request here.
		 * Individual access methods can optionally superimpose buffer page
		 * client requests on top of our client requests to enforce that
		 * buffers are only accessed while locked (and pinned).  It's possible
		 * that the buffer page is legitimately non-accessible here.  We
		 * cannot meddle with that.
		 */
		result = (pg_atomic_read_u64(&buf->state) & BM_VALID) != 0;

		Assert(ref->data.refcount > 0);
		ref->data.refcount++;
		ResourceOwnerRememberBuffer(CurrentResourceOwner, b);
	}

	return result;
}

/*
 * PinBuffer_Locked -- as above, but caller already locked the buffer header.
 * The spinlock is released before return.
 *
 * As this function is called with the spinlock held, the caller has to
 * previously call ReservePrivateRefCountEntry() and
 * ResourceOwnerEnlarge(CurrentResourceOwner);
 *
 * Currently, no callers of this function want to modify the buffer's
 * usage_count at all, so there's no need for a strategy parameter.
 * Also we don't bother with a BM_VALID test (the caller could check that for
 * itself).
 *
 * Also all callers only ever use this function when it's known that the
 * buffer can't have a preexisting pin by this backend. That allows us to skip
 * searching the private refcount array & hash, which is a boon, because the
 * spinlock is still held.
 *
 * Note: use of this routine is frequently mandatory, not just an optimization
 * to save a spin lock/unlock cycle, because we need to pin a buffer before
 * its state can change under us.
 */
static void
PinBuffer_Locked(BufferDesc *buf)
{
	uint64		old_buf_state;

	/*
	 * As explained, We don't expect any preexisting pins. That allows us to
	 * manipulate the PrivateRefCount after releasing the spinlock
	 */
	Assert(GetPrivateRefCountEntry(BufferDescriptorGetBuffer(buf), false) == NULL);

	/*
	 * Since we hold the buffer spinlock, we can update the buffer state and
	 * release the lock in one operation.
	 */
	old_buf_state = pg_atomic_read_u64(&buf->state);

	UnlockBufHdrExt(buf, old_buf_state,
					0, 0, 1);

	TrackNewBufferPin(BufferDescriptorGetBuffer(buf));
}

/*
 * Support for waking up another backend that is waiting for the cleanup lock
 * to be released using BM_PIN_COUNT_WAITER.
 *
 * See LockBufferForCleanup().
 *
 * Expected to be called just after releasing a buffer pin (in a BufferDesc,
 * not just reducing the backend-local pincount for the buffer).
 */
static void
WakePinCountWaiter(BufferDesc *buf)
{
	/*
	 * Acquire the buffer header lock, re-check that there's a waiter. Another
	 * backend could have unpinned this buffer, and already woken up the
	 * waiter.
	 *
	 * There's no danger of the buffer being replaced after we unpinned it
	 * above, as it's pinned by the waiter. The waiter removes
	 * BM_PIN_COUNT_WAITER if it stops waiting for a reason other than this
	 * backend waking it up.
	 */
	uint64		buf_state = LockBufHdr(buf);

	if ((buf_state & BM_PIN_COUNT_WAITER) &&
		BUF_STATE_GET_REFCOUNT(buf_state) == 1)
	{
		/* we just released the last pin other than the waiter's */
		int			wait_backend_pgprocno = buf->wait_backend_pgprocno;

		UnlockBufHdrExt(buf, buf_state,
						0, BM_PIN_COUNT_WAITER,
						0);
		ProcSendSignal(wait_backend_pgprocno);
	}
	else
		UnlockBufHdr(buf);
}

/*
 * UnpinBuffer -- make buffer available for replacement.
 *
 * This should be applied only to shared buffers, never local ones.  This
 * always adjusts CurrentResourceOwner.
 */
static void
UnpinBuffer(BufferDesc *buf)
{
	Buffer		b = BufferDescriptorGetBuffer(buf);

	ResourceOwnerForgetBuffer(CurrentResourceOwner, b);
	UnpinBufferNoOwner(buf);
}

static void
UnpinBufferNoOwner(BufferDesc *buf)
{
	PrivateRefCountEntry *ref;
	Buffer		b = BufferDescriptorGetBuffer(buf);

	Assert(!BufferIsLocal(b));

	/* not moving as we're likely deleting it soon anyway */
	ref = GetPrivateRefCountEntry(b, false);
	Assert(ref != NULL);
	Assert(ref->data.refcount > 0);
	ref->data.refcount--;
	if (ref->data.refcount == 0)
	{
		uint64		old_buf_state;

		/*
		 * Mark buffer non-accessible to Valgrind.
		 *
		 * Note that the buffer may have already been marked non-accessible
		 * within access method code that enforces that buffers are only
		 * accessed while a buffer lock is held.
		 */
		VALGRIND_MAKE_MEM_NOACCESS(BufHdrGetBlock(buf), BLCKSZ);

		/*
		 * I'd better not still hold the buffer content lock. Can't use
		 * BufferIsLockedByMe(), as that asserts the buffer is pinned.
		 */
		Assert(!BufferLockHeldByMe(buf));

		/* decrement the shared reference count */
		old_buf_state = pg_atomic_fetch_sub_u64(&buf->state, BUF_REFCOUNT_ONE);

		/* Support LockBufferForCleanup() */
		if (old_buf_state & BM_PIN_COUNT_WAITER)
			WakePinCountWaiter(buf);

		ForgetPrivateRefCountEntry(ref);
	}
}

/*
 * Set up backend-local tracking of a buffer pinned the first time by this
 * backend.
 */
inline void
TrackNewBufferPin(Buffer buf)
{
	PrivateRefCountEntry *ref;

	ref = NewPrivateRefCountEntry(buf);
	ref->data.refcount++;

	ResourceOwnerRememberBuffer(CurrentResourceOwner, buf);

	/*
	 * This is the first pin for this page by this backend, mark its page as
	 * defined to valgrind. While the page contents might not actually be
	 * valid yet, we don't currently guarantee that such pages are marked
	 * undefined or non-accessible.
	 *
	 * It's not necessarily the prettiest to do this here, but otherwise we'd
	 * need this block of code in multiple places.
	 */
	VALGRIND_MAKE_MEM_DEFINED(BufHdrGetBlock(GetBufferDescriptor(buf - 1)),
							  BLCKSZ);
}

#define ST_SORT sort_checkpoint_bufferids
#define ST_ELEMENT_TYPE CkptSortItem
#define ST_COMPARE(a, b) ckpt_buforder_comparator(a, b)
#define ST_SCOPE static
#define ST_DEFINE
#include "lib/sort_template.h"

/*
 * BufferSync -- Write out all dirty buffers in the pool.
 *
 * This is called at checkpoint time to write out all dirty shared buffers.
 * The checkpoint request flags should be passed in.  If CHECKPOINT_FAST is
 * set, we disable delays between writes; if CHECKPOINT_IS_SHUTDOWN,
 * CHECKPOINT_END_OF_RECOVERY or CHECKPOINT_FLUSH_UNLOGGED is set, we write
 * even unlogged buffers, which are otherwise skipped.  The remaining flags
 * currently have no effect here.
 */
static void
BufferSync(int flags)
{
	uint64		buf_state;
	int			buf_id;
	int			num_to_scan;
	int			num_spaces;
	int			num_processed;
	int			num_written;
	CkptTsStatus *per_ts_stat = NULL;
	Oid			last_tsid;
	binaryheap *ts_heap;
	int			i;
	uint64		mask = BM_DIRTY;
	WritebackContext wb_context;

	/*
	 * Unless this is a shutdown checkpoint or we have been explicitly told,
	 * we write only permanent, dirty buffers.  But at shutdown or end of
	 * recovery, we write all dirty buffers.
	 */
	if (!((flags & (CHECKPOINT_IS_SHUTDOWN | CHECKPOINT_END_OF_RECOVERY |
					CHECKPOINT_FLUSH_UNLOGGED))))
		mask |= BM_PERMANENT;

	/*
	 * Loop over all buffers, and mark the ones that need to be written with
	 * BM_CHECKPOINT_NEEDED.  Count them as we go (num_to_scan), so that we
	 * can estimate how much work needs to be done.
	 *
	 * This allows us to write only those pages that were dirty when the
	 * checkpoint began, and not those that get dirtied while it proceeds.
	 * Whenever a page with BM_CHECKPOINT_NEEDED is written out, either by us
	 * later in this function, or by normal backends or the bgwriter cleaning
	 * scan, the flag is cleared.  Any buffer dirtied after this point won't
	 * have the flag set.
	 *
	 * Note that if we fail to write some buffer, we may leave buffers with
	 * BM_CHECKPOINT_NEEDED still set.  This is OK since any such buffer would
	 * certainly need to be written for the next checkpoint attempt, too.
	 */
	num_to_scan = 0;
	for (buf_id = 0; buf_id < NBuffers; buf_id++)
	{
		BufferDesc *bufHdr = GetBufferDescriptor(buf_id);
		uint64		set_bits = 0;

		/*
		 * Header spinlock is enough to examine BM_DIRTY, see comment in
		 * SyncOneBuffer.
		 */
		buf_state = LockBufHdr(bufHdr);

		if ((buf_state & mask) == mask)
		{
			CkptSortItem *item;

			set_bits = BM_CHECKPOINT_NEEDED;

			item = &CkptBufferIds[num_to_scan++];
			item->buf_id = buf_id;
			item->tsId = bufHdr->tag.spcOid;
			item->relNumber = BufTagGetRelNumber(&bufHdr->tag);
			item->forkNum = BufTagGetForkNum(&bufHdr->tag);
			item->blockNum = bufHdr->tag.blockNum;
		}

		UnlockBufHdrExt(bufHdr, buf_state,
						set_bits, 0,
						0);

		/* Check for barrier events in case NBuffers is large. */
		if (ProcSignalBarrierPending)
			ProcessProcSignalBarrier();
	}

	if (num_to_scan == 0)
		return;					/* nothing to do */

	WritebackContextInit(&wb_context, &checkpoint_flush_after);

	TRACE_POSTGRESQL_BUFFER_SYNC_START(NBuffers, num_to_scan);

	/*
	 * Sort buffers that need to be written to reduce the likelihood of random
	 * IO. The sorting is also important for the implementation of balancing
	 * writes between tablespaces. Without balancing writes we'd potentially
	 * end up writing to the tablespaces one-by-one; possibly overloading the
	 * underlying system.
	 */
	sort_checkpoint_bufferids(CkptBufferIds, num_to_scan);

	num_spaces = 0;

	/*
	 * Allocate progress status for each tablespace with buffers that need to
	 * be flushed. This requires the to-be-flushed array to be sorted.
	 */
	last_tsid = InvalidOid;
	for (i = 0; i < num_to_scan; i++)
	{
		CkptTsStatus *s;
		Oid			cur_tsid;

		cur_tsid = CkptBufferIds[i].tsId;

		/*
		 * Grow array of per-tablespace status structs, every time a new
		 * tablespace is found.
		 */
		if (last_tsid == InvalidOid || last_tsid != cur_tsid)
		{
			Size		sz;

			num_spaces++;

			/*
			 * Not worth adding grow-by-power-of-2 logic here - even with a
			 * few hundred tablespaces this should be fine.
			 */
			sz = sizeof(CkptTsStatus) * num_spaces;

			if (per_ts_stat == NULL)
				per_ts_stat = (CkptTsStatus *) palloc(sz);
			else
				per_ts_stat = (CkptTsStatus *) repalloc(per_ts_stat, sz);

			s = &per_ts_stat[num_spaces - 1];
			memset(s, 0, sizeof(*s));
			s->tsId = cur_tsid;

			/*
			 * The first buffer in this tablespace. As CkptBufferIds is sorted
			 * by tablespace all (s->num_to_scan) buffers in this tablespace
			 * will follow afterwards.
			 */
			s->index = i;

			/*
			 * progress_slice will be determined once we know how many buffers
			 * are in each tablespace, i.e. after this loop.
			 */

			last_tsid = cur_tsid;
		}
		else
		{
			s = &per_ts_stat[num_spaces - 1];
		}

		s->num_to_scan++;

		/* Check for barrier events. */
		if (ProcSignalBarrierPending)
			ProcessProcSignalBarrier();
	}

	Assert(num_spaces > 0);

	/*
	 * Build a min-heap over the write-progress in the individual tablespaces,
	 * and compute how large a portion of the total progress a single
	 * processed buffer is.
	 */
	ts_heap = binaryheap_allocate(num_spaces,
								  ts_ckpt_progress_comparator,
								  NULL);

	for (i = 0; i < num_spaces; i++)
	{
		CkptTsStatus *ts_stat = &per_ts_stat[i];

		ts_stat->progress_slice = (float8) num_to_scan / ts_stat->num_to_scan;

		binaryheap_add_unordered(ts_heap, PointerGetDatum(ts_stat));
	}

	binaryheap_build(ts_heap);

	/*
	 * Iterate through to-be-checkpointed buffers and write the ones (still)
	 * marked with BM_CHECKPOINT_NEEDED. The writes are balanced between
	 * tablespaces; otherwise the sorting would lead to only one tablespace
	 * receiving writes at a time, making inefficient use of the hardware.
	 */
	num_processed = 0;
	num_written = 0;
	while (!binaryheap_empty(ts_heap))
	{
		BufferDesc *bufHdr = NULL;
		CkptTsStatus *ts_stat = (CkptTsStatus *)
			DatumGetPointer(binaryheap_first(ts_heap));

		buf_id = CkptBufferIds[ts_stat->index].buf_id;
		Assert(buf_id != -1);

		bufHdr = GetBufferDescriptor(buf_id);

		num_processed++;

		/*
		 * We don't need to acquire the lock here, because we're only looking
		 * at a single bit. It's possible that someone else writes the buffer
		 * and clears the flag right after we check, but that doesn't matter
		 * since SyncOneBuffer will then do nothing.  However, there is a
		 * further race condition: it's conceivable that between the time we
		 * examine the bit here and the time SyncOneBuffer acquires the lock,
		 * someone else not only wrote the buffer but replaced it with another
		 * page and dirtied it.  In that improbable case, SyncOneBuffer will
		 * write the buffer though we didn't need to.  It doesn't seem worth
		 * guarding against this, though.
		 */
		if (pg_atomic_read_u64(&bufHdr->state) & BM_CHECKPOINT_NEEDED)
		{
			if (SyncOneBuffer(buf_id, false, &wb_context) & BUF_WRITTEN)
			{
				TRACE_POSTGRESQL_BUFFER_SYNC_WRITTEN(buf_id);
				PendingCheckpointerStats.buffers_written++;
				num_written++;
			}
		}

		/*
		 * Measure progress independent of actually having to flush the buffer
		 * - otherwise writing become unbalanced.
		 */
		ts_stat->progress += ts_stat->progress_slice;
		ts_stat->num_scanned++;
		ts_stat->index++;

		/* Have all the buffers from the tablespace been processed? */
		if (ts_stat->num_scanned == ts_stat->num_to_scan)
		{
			binaryheap_remove_first(ts_heap);
		}
		else
		{
			/* update heap with the new progress */
			binaryheap_replace_first(ts_heap, PointerGetDatum(ts_stat));
		}

		/*
		 * Sleep to throttle our I/O rate.
		 *
		 * (This will check for barrier events even if it doesn't sleep.)
		 */
		CheckpointWriteDelay(flags, (double) num_processed / num_to_scan);
	}

	/*
	 * Issue all pending flushes. Only checkpointer calls BufferSync(), so
	 * IOContext will always be IOCONTEXT_NORMAL.
	 */
	IssuePendingWritebacks(&wb_context, IOCONTEXT_NORMAL);

	pfree(per_ts_stat);
	per_ts_stat = NULL;
	binaryheap_free(ts_heap);

	/*
	 * Update checkpoint statistics. As noted above, this doesn't include
	 * buffers written by other backends or bgwriter scan.
	 */
	CheckpointStats.ckpt_bufs_written += num_written;

	TRACE_POSTGRESQL_BUFFER_SYNC_DONE(NBuffers, num_written, num_to_scan);
}

/*
 * BgBufferSync -- Write out some dirty buffers in the pool.
 *
 * This is called periodically by the background writer process.
 *
 * Returns true if it's appropriate for the bgwriter process to go into
 * low-power hibernation mode.  (This happens if the strategy clock-sweep
 * has been "lapped" and no buffer allocations have occurred recently,
 * or if the bgwriter has been effectively disabled by setting
 * bgwriter_lru_maxpages to 0.)
 */
bool
BgBufferSync(WritebackContext *wb_context)
{
	/* info obtained from freelist.c */
	int			strategy_buf_id;
	uint32		strategy_passes;
	uint32		recent_alloc;

	/*
	 * Information saved between calls so we can determine the strategy
	 * point's advance rate and avoid scanning already-cleaned buffers.
	 */
	static bool saved_info_valid = false;
	static int	prev_strategy_buf_id;
	static uint32 prev_strategy_passes;
	static int	next_to_clean;
	static uint32 next_passes;

	/* Moving averages of allocation rate and clean-buffer density */
	static float smoothed_alloc = 0;
	static float smoothed_density = 10.0;

	/* Potentially these could be tunables, but for now, not */
	float		smoothing_samples = 16;
	float		scan_whole_pool_milliseconds = 120000.0;

	/* Used to compute how far we scan ahead */
	long		strategy_delta;
	int			bufs_to_lap;
	int			bufs_ahead;
	float		scans_per_alloc;
	int			reusable_buffers_est;
	int			upcoming_alloc_est;
	int			min_scan_buffers;

	/* Variables for the scanning loop proper */
	int			num_to_scan;
	int			num_written;
	int			reusable_buffers;

	/* Variables for final smoothed_density update */
	long		new_strategy_delta;
	uint32		new_recent_alloc;

	/*
	 * Find out where the clock-sweep currently is, and how many buffer
	 * allocations have happened since our last call.
	 */
	strategy_buf_id = StrategySyncStart(&strategy_passes, &recent_alloc);

	/* Report buffer alloc counts to pgstat */
	PendingBgWriterStats.buf_alloc += recent_alloc;

	/*
	 * If we're not running the LRU scan, just stop after doing the stats
	 * stuff.  We mark the saved state invalid so that we can recover sanely
	 * if LRU scan is turned back on later.
	 */
	if (bgwriter_lru_maxpages <= 0)
	{
		saved_info_valid = false;
		return true;
	}

	/*
	 * Compute strategy_delta = how many buffers have been scanned by the
	 * clock-sweep since last time.  If first time through, assume none. Then
	 * see if we are still ahead of the clock-sweep, and if so, how many
	 * buffers we could scan before we'd catch up with it and "lap" it. Note:
	 * weird-looking coding of xxx_passes comparisons are to avoid bogus
	 * behavior when the passes counts wrap around.
	 */
	if (saved_info_valid)
	{
		int32		passes_delta = strategy_passes - prev_strategy_passes;

		strategy_delta = strategy_buf_id - prev_strategy_buf_id;
		strategy_delta += (long) passes_delta * NBuffers;

		Assert(strategy_delta >= 0);

		if ((int32) (next_passes - strategy_passes) > 0)
		{
			/* we're one pass ahead of the strategy point */
			bufs_to_lap = strategy_buf_id - next_to_clean;
#ifdef BGW_DEBUG
			elog(DEBUG2, "bgwriter ahead: bgw %u-%u strategy %u-%u delta=%ld lap=%d",
				 next_passes, next_to_clean,
				 strategy_passes, strategy_buf_id,
				 strategy_delta, bufs_to_lap);
#endif
		}
		else if (next_passes == strategy_passes &&
				 next_to_clean >= strategy_buf_id)
		{
			/* on same pass, but ahead or at least not behind */
			bufs_to_lap = NBuffers - (next_to_clean - strategy_buf_id);
#ifdef BGW_DEBUG
			elog(DEBUG2, "bgwriter ahead: bgw %u-%u strategy %u-%u delta=%ld lap=%d",
				 next_passes, next_to_clean,
				 strategy_passes, strategy_buf_id,
				 strategy_delta, bufs_to_lap);
#endif
		}
		else
		{
			/*
			 * We're behind, so skip forward to the strategy point and start
			 * cleaning from there.
			 */
#ifdef BGW_DEBUG
			elog(DEBUG2, "bgwriter behind: bgw %u-%u strategy %u-%u delta=%ld",
				 next_passes, next_to_clean,
				 strategy_passes, strategy_buf_id,
				 strategy_delta);
#endif
			next_to_clean = strategy_buf_id;
			next_passes = strategy_passes;
			bufs_to_lap = NBuffers;
		}
	}
	else
	{
		/*
		 * Initializing at startup or after LRU scanning had been off. Always
		 * start at the strategy point.
		 */
#ifdef BGW_DEBUG
		elog(DEBUG2, "bgwriter initializing: strategy %u-%u",
			 strategy_passes, strategy_buf_id);
#endif
		strategy_delta = 0;
		next_to_clean = strategy_buf_id;
		next_passes = strategy_passes;
		bufs_to_lap = NBuffers;
	}

	/* Update saved info for next time */
	prev_strategy_buf_id = strategy_buf_id;
	prev_strategy_passes = strategy_passes;
	saved_info_valid = true;

	/*
	 * Compute how many buffers had to be scanned for each new allocation, ie,
	 * 1/density of reusable buffers, and track a moving average of that.
	 *
	 * If the strategy point didn't move, we don't update the density estimate
	 */
	if (strategy_delta > 0 && recent_alloc > 0)
	{
		scans_per_alloc = (float) strategy_delta / (float) recent_alloc;
		smoothed_density += (scans_per_alloc - smoothed_density) /
			smoothing_samples;
	}

	/*
	 * Estimate how many reusable buffers there are between the current
	 * strategy point and where we've scanned ahead to, based on the smoothed
	 * density estimate.
	 */
	bufs_ahead = NBuffers - bufs_to_lap;
	reusable_buffers_est = (float) bufs_ahead / smoothed_density;

	/*
	 * Track a moving average of recent buffer allocations.  Here, rather than
	 * a true average we want a fast-attack, slow-decline behavior: we
	 * immediately follow any increase.
	 */
	if (smoothed_alloc <= (float) recent_alloc)
		smoothed_alloc = recent_alloc;
	else
		smoothed_alloc += ((float) recent_alloc - smoothed_alloc) /
			smoothing_samples;

	/* Scale the estimate by a GUC to allow more aggressive tuning. */
	upcoming_alloc_est = (int) (smoothed_alloc * bgwriter_lru_multiplier);

	/*
	 * If recent_alloc remains at zero for many cycles, smoothed_alloc will
	 * eventually underflow to zero, and the underflows produce annoying
	 * kernel warnings on some platforms.  Once upcoming_alloc_est has gone to
	 * zero, there's no point in tracking smaller and smaller values of
	 * smoothed_alloc, so just reset it to exactly zero to avoid this
	 * syndrome.  It will pop back up as soon as recent_alloc increases.
	 */
	if (upcoming_alloc_est == 0)
		smoothed_alloc = 0;

	/*
	 * Even in cases where there's been little or no buffer allocation
	 * activity, we want to make a small amount of progress through the buffer
	 * cache so that as many reusable buffers as possible are clean after an
	 * idle period.
	 *
	 * (scan_whole_pool_milliseconds / BgWriterDelay) computes how many times
	 * the BGW will be called during the scan_whole_pool time; slice the
	 * buffer pool into that many sections.
	 */
	min_scan_buffers = (int) (NBuffers / (scan_whole_pool_milliseconds / BgWriterDelay));

	if (upcoming_alloc_est < (min_scan_buffers + reusable_buffers_est))
	{
#ifdef BGW_DEBUG
		elog(DEBUG2, "bgwriter: alloc_est=%d too small, using min=%d + reusable_est=%d",
			 upcoming_alloc_est, min_scan_buffers, reusable_buffers_est);
#endif
		upcoming_alloc_est = min_scan_buffers + reusable_buffers_est;
	}

	/*
	 * Now write out dirty reusable buffers, working forward from the
	 * next_to_clean point, until we have lapped the strategy scan, or cleaned
	 * enough buffers to match our estimate of the next cycle's allocation
	 * requirements, or hit the bgwriter_lru_maxpages limit.
	 */

	num_to_scan = bufs_to_lap;
	num_written = 0;
	reusable_buffers = reusable_buffers_est;

	/* Execute the LRU scan */
	while (num_to_scan > 0 && reusable_buffers < upcoming_alloc_est)
	{
		int			sync_state = SyncOneBuffer(next_to_clean, true,
											   wb_context);

		if (++next_to_clean >= NBuffers)
		{
			next_to_clean = 0;
			next_passes++;
		}
		num_to_scan--;

		if (sync_state & BUF_WRITTEN)
		{
			reusable_buffers++;
			if (++num_written >= bgwriter_lru_maxpages)
			{
				PendingBgWriterStats.maxwritten_clean++;
				break;
			}
		}
		else if (sync_state & BUF_REUSABLE)
			reusable_buffers++;
	}

	PendingBgWriterStats.buf_written_clean += num_written;

#ifdef BGW_DEBUG
	elog(DEBUG1, "bgwriter: recent_alloc=%u smoothed=%.2f delta=%ld ahead=%d density=%.2f reusable_est=%d upcoming_est=%d scanned=%d wrote=%d reusable=%d",
		 recent_alloc, smoothed_alloc, strategy_delta, bufs_ahead,
		 smoothed_density, reusable_buffers_est, upcoming_alloc_est,
		 bufs_to_lap - num_to_scan,
		 num_written,
		 reusable_buffers - reusable_buffers_est);
#endif

	/*
	 * Consider the above scan as being like a new allocation scan.
	 * Characterize its density and update the smoothed one based on it. This
	 * effectively halves the moving average period in cases where both the
	 * strategy and the background writer are doing some useful scanning,
	 * which is helpful because a long memory isn't as desirable on the
	 * density estimates.
	 */
	new_strategy_delta = bufs_to_lap - num_to_scan;
	new_recent_alloc = reusable_buffers - reusable_buffers_est;
	if (new_strategy_delta > 0 && new_recent_alloc > 0)
	{
		scans_per_alloc = (float) new_strategy_delta / (float) new_recent_alloc;
		smoothed_density += (scans_per_alloc - smoothed_density) /
			smoothing_samples;

#ifdef BGW_DEBUG
		elog(DEBUG2, "bgwriter: cleaner density alloc=%u scan=%ld density=%.2f new smoothed=%.2f",
			 new_recent_alloc, new_strategy_delta,
			 scans_per_alloc, smoothed_density);
#endif
	}

	/* Return true if OK to hibernate */
	return (bufs_to_lap == 0 && recent_alloc == 0);
}

/*
 * SyncOneBuffer -- process a single buffer during syncing.
 *
 * If skip_recently_used is true, we don't write currently-pinned buffers, nor
 * buffers marked recently used, as these are not replacement candidates.
 *
 * Returns a bitmask containing the following flag bits:
 *	BUF_WRITTEN: we wrote the buffer.
 *	BUF_REUSABLE: buffer is available for replacement, ie, it has
 *		pin count 0 and usage count 0.
 *
 * (BUF_WRITTEN could be set in error if FlushBuffer finds the buffer clean
 * after locking it, but we don't care all that much.)
 */
static int
SyncOneBuffer(int buf_id, bool skip_recently_used, WritebackContext *wb_context)
{
	BufferDesc *bufHdr = GetBufferDescriptor(buf_id);
	int			result = 0;
	uint64		buf_state;
	BufferTag	tag;

	/* Make sure we can handle the pin */
	ReservePrivateRefCountEntry();
	ResourceOwnerEnlarge(CurrentResourceOwner);

	/*
	 * Check whether buffer needs writing.
	 *
	 * We can make this check without taking the buffer content lock so long
	 * as we mark pages dirty in access methods *before* logging changes with
	 * XLogInsert(): if someone marks the buffer dirty just after our check we
	 * don't worry because our checkpoint.redo points before log record for
	 * upcoming changes and so we are not required to write such dirty buffer.
	 */
	buf_state = LockBufHdr(bufHdr);

	if (BUF_STATE_GET_REFCOUNT(buf_state) == 0 &&
		BUF_STATE_GET_USAGECOUNT(buf_state) == 0)
	{
		result |= BUF_REUSABLE;
	}
	else if (skip_recently_used)
	{
		/* Caller told us not to write recently-used buffers */
		UnlockBufHdr(bufHdr);
		return result;
	}

	if (!(buf_state & BM_VALID) || !(buf_state & BM_DIRTY))
	{
		/* It's clean, so nothing to do */
		UnlockBufHdr(bufHdr);
		return result;
	}

	/*
	 * Pin it, share-exclusive-lock it, write it.  (FlushBuffer will do
	 * nothing if the buffer is clean by the time we've locked it.)
	 */
	PinBuffer_Locked(bufHdr);

	FlushUnlockedBuffer(bufHdr, NULL, IOOBJECT_RELATION, IOCONTEXT_NORMAL);

	tag = bufHdr->tag;

	UnpinBuffer(bufHdr);

	/*
	 * SyncOneBuffer() is only called by checkpointer and bgwriter, so
	 * IOContext will always be IOCONTEXT_NORMAL.
	 */
	ScheduleBufferTagForWriteback(wb_context, IOCONTEXT_NORMAL, &tag);

	return result | BUF_WRITTEN;
}

/*
 *		AtEOXact_Buffers - clean up at end of transaction.
 *
 *		As of PostgreSQL 8.0, buffer pins should get released by the
 *		ResourceOwner mechanism.  This routine is just a debugging
 *		cross-check that no pins remain.
 */
void
AtEOXact_Buffers(bool isCommit)
{
	CheckForBufferLeaks();

	AtEOXact_LocalBuffers(isCommit);

	Assert(PrivateRefCountOverflowed == 0);
}

/*
 * Initialize access to shared buffer pool
 *
 * This is called during backend startup (whether standalone or under the
 * postmaster).  It sets up for this backend's access to the already-existing
 * buffer pool.
 */
void
InitBufferManagerAccess(void)
{
	/*
	 * An advisory limit on the number of pins each backend should hold, based
	 * on shared_buffers and the maximum number of connections possible.
	 * That's very pessimistic, but outside toy-sized shared_buffers it should
	 * allow plenty of pins.  LimitAdditionalPins() and
	 * GetAdditionalPinLimit() can be used to check the remaining balance.
	 */
	MaxProportionalPins = NBuffers / (MaxBackends + NUM_AUXILIARY_PROCS);

	memset(&PrivateRefCountArray, 0, sizeof(PrivateRefCountArray));
	memset(&PrivateRefCountArrayKeys, 0, sizeof(PrivateRefCountArrayKeys));

	PrivateRefCountHash = refcount_create(CurrentMemoryContext, 100, NULL);

	/*
	 * AtProcExit_Buffers needs LWLock access, and thereby has to be called at
	 * the corresponding phase of backend shutdown.
	 */
	Assert(MyProc != NULL);
	on_shmem_exit(AtProcExit_Buffers, 0);
}

/*
 * During backend exit, ensure that we released all shared-buffer locks and
 * assert that we have no remaining pins.
 */
static void
AtProcExit_Buffers(int code, Datum arg)
{
	UnlockBuffers();

	CheckForBufferLeaks();

	/* localbuf.c needs a chance too */
	AtProcExit_LocalBuffers();
}

/*
 *		CheckForBufferLeaks - ensure this backend holds no buffer pins
 *
 *		As of PostgreSQL 8.0, buffer pins should get released by the
 *		ResourceOwner mechanism.  This routine is just a debugging
 *		cross-check that no pins remain.
 */
static void
CheckForBufferLeaks(void)
{
#ifdef USE_ASSERT_CHECKING
	int			RefCountErrors = 0;
	PrivateRefCountEntry *res;
	int			i;
	char	   *s;

	/* check the array */
	for (i = 0; i < REFCOUNT_ARRAY_ENTRIES; i++)
	{
		if (PrivateRefCountArrayKeys[i] != InvalidBuffer)
		{
			res = &PrivateRefCountArray[i];

			s = DebugPrintBufferRefcount(res->buffer);
			elog(WARNING, "buffer refcount leak: %s", s);
			pfree(s);

			RefCountErrors++;
		}
	}

	/* if necessary search the hash */
	if (PrivateRefCountOverflowed)
	{
		refcount_iterator iter;

		refcount_start_iterate(PrivateRefCountHash, &iter);
		while ((res = refcount_iterate(PrivateRefCountHash, &iter)) != NULL)
		{
			s = DebugPrintBufferRefcount(res->buffer);
			elog(WARNING, "buffer refcount leak: %s", s);
			pfree(s);
			RefCountErrors++;
		}
	}

	Assert(RefCountErrors == 0);
#endif
}

#ifdef USE_ASSERT_CHECKING
/*
 * Check for exclusive-locked catalog buffers.  This is the core of
 * AssertCouldGetRelation().
 *
 * A backend would self-deadlock on the content lock if the catalog scan read
 * the exclusive-locked buffer.  The main threat is exclusive-locked buffers
 * of catalogs used in relcache, because a catcache search on any catalog may
 * build that catalog's relcache entry.  We don't have an inventory of
 * catalogs relcache uses, so just check buffers of most catalogs.
 *
 * It's better to minimize waits while holding an exclusive buffer lock, so it
 * would be nice to broaden this check not to be catalog-specific.  However,
 * bttextcmp() accesses pg_collation, and non-core opclasses might similarly
 * read tables.  That is deadlock-free as long as there's no loop in the
 * dependency graph: modifying table A may cause an opclass to read table B,
 * but it must not cause a read of table A.
 */
void
AssertBufferLocksPermitCatalogRead(void)
{
	PrivateRefCountEntry *res;

	/* check the array */
	for (int i = 0; i < REFCOUNT_ARRAY_ENTRIES; i++)
	{
		if (PrivateRefCountArrayKeys[i] != InvalidBuffer)
		{
			res = &PrivateRefCountArray[i];

			if (res->buffer == InvalidBuffer)
				continue;

			AssertNotCatalogBufferLock(res->buffer, res->data.lockmode);
		}
	}

	/* if necessary search the hash */
	if (PrivateRefCountOverflowed)
	{
		refcount_iterator iter;

		refcount_start_iterate(PrivateRefCountHash, &iter);
		while ((res = refcount_iterate(PrivateRefCountHash, &iter)) != NULL)
		{
			AssertNotCatalogBufferLock(res->buffer, res->data.lockmode);
		}
	}
}

static void
AssertNotCatalogBufferLock(Buffer buffer, BufferLockMode mode)
{
	BufferDesc *bufHdr = GetBufferDescriptor(buffer - 1);
	BufferTag	tag;
	Oid			relid;

	if (mode != BUFFER_LOCK_EXCLUSIVE)
		return;

	tag = bufHdr->tag;

	/*
	 * This relNumber==relid assumption holds until a catalog experiences
	 * VACUUM FULL or similar.  After a command like that, relNumber will be
	 * in the normal (non-catalog) range, and we lose the ability to detect
	 * hazardous access to that catalog.  Calling RelidByRelfilenumber() would
	 * close that gap, but RelidByRelfilenumber() might then deadlock with a
	 * held lock.
	 */
	relid = tag.relNumber;

	if (IsCatalogTextUniqueIndexOid(relid)) /* see comments at the callee */
		return;

	Assert(!IsCatalogRelationOid(relid));
}
#endif


/*
 * Helper routine to issue warnings when a buffer is unexpectedly pinned
 */
char *
DebugPrintBufferRefcount(Buffer buffer)
{
	BufferDesc *buf;
	int32		loccount;
	char	   *result;
	ProcNumber	backend;
	uint64		buf_state;

	Assert(BufferIsValid(buffer));
	if (BufferIsLocal(buffer))
	{
		buf = GetLocalBufferDescriptor(-buffer - 1);
		loccount = LocalRefCount[-buffer - 1];
		backend = MyProcNumber;
	}
	else
	{
		buf = GetBufferDescriptor(buffer - 1);
		loccount = GetPrivateRefCount(buffer);
		backend = INVALID_PROC_NUMBER;
	}

	/* theoretically we should lock the bufHdr here */
	buf_state = pg_atomic_read_u64(&buf->state);

	result = psprintf("[%03d] (rel=%s, blockNum=%u, flags=0x%" PRIx64 ", refcount=%u %d)",
					  buffer,
					  relpathbackend(BufTagGetRelFileLocator(&buf->tag), backend,
									 BufTagGetForkNum(&buf->tag)).str,
					  buf->tag.blockNum, buf_state & BUF_FLAG_MASK,
					  BUF_STATE_GET_REFCOUNT(buf_state), loccount);
	return result;
}

/*
 * CheckPointBuffers
 *
 * Flush all dirty blocks in buffer pool to disk at checkpoint time.
 *
 * Note: temporary relations do not participate in checkpoints, so they don't
 * need to be flushed.
 */
void
CheckPointBuffers(int flags)
{
	BufferSync(flags);
}

/*
 * BufferGetBlockNumber
 *		Returns the block number associated with a buffer.
 *
 * Note:
 *		Assumes that the buffer is valid and pinned, else the
 *		value may be obsolete immediately...
 */
BlockNumber
BufferGetBlockNumber(Buffer buffer)
{
	BufferDesc *bufHdr;

	Assert(BufferIsPinned(buffer));

	if (BufferIsLocal(buffer))
		bufHdr = GetLocalBufferDescriptor(-buffer - 1);
	else
		bufHdr = GetBufferDescriptor(buffer - 1);

	/* pinned, so OK to read tag without spinlock */
	return bufHdr->tag.blockNum;
}

/*
 * BufferGetTag
 *		Returns the relfilelocator, fork number and block number associated with
 *		a buffer.
 */
void
BufferGetTag(Buffer buffer, RelFileLocator *rlocator, ForkNumber *forknum,
			 BlockNumber *blknum)
{
	BufferDesc *bufHdr;

	/* Do the same checks as BufferGetBlockNumber. */
	Assert(BufferIsPinned(buffer));

	if (BufferIsLocal(buffer))
		bufHdr = GetLocalBufferDescriptor(-buffer - 1);
	else
		bufHdr = GetBufferDescriptor(buffer - 1);

	/* pinned, so OK to read tag without spinlock */
	*rlocator = BufTagGetRelFileLocator(&bufHdr->tag);
	*forknum = BufTagGetForkNum(&bufHdr->tag);
	*blknum = bufHdr->tag.blockNum;
}

/*
 * FlushBuffer
 *		Physically write out a shared buffer.
 *
 * NOTE: this actually just passes the buffer contents to the kernel; the
 * real write to disk won't happen until the kernel feels like it.  This
 * is okay from our point of view since we can redo the changes from WAL.
 * However, we will need to force the changes to disk via fsync before
 * we can checkpoint WAL.
 *
 * The caller must hold a pin on the buffer and have
 * (share-)exclusively-locked the buffer contents.
 *
 * If the caller has an smgr reference for the buffer's relation, pass it
 * as the second parameter.  If not, pass NULL.
 */
static void
FlushBuffer(BufferDesc *buf, SMgrRelation reln, IOObject io_object,
			IOContext io_context)
{
	XLogRecPtr	recptr;
	ErrorContextCallback errcallback;
	instr_time	io_start;
	Block		bufBlock;

	Assert(BufferLockHeldByMeInMode(buf, BUFFER_LOCK_EXCLUSIVE) ||
		   BufferLockHeldByMeInMode(buf, BUFFER_LOCK_SHARE_EXCLUSIVE));

	/*
	 * Try to start an I/O operation.  If StartBufferIO returns false, then
	 * someone else flushed the buffer before we could, so we need not do
	 * anything.
	 */
	if (StartSharedBufferIO(buf, false, true, NULL) == BUFFER_IO_ALREADY_DONE)
		return;

	/* Setup error traceback support for ereport() */
	errcallback.callback = shared_buffer_write_error_callback;
	errcallback.arg = buf;
	errcallback.previous = error_context_stack;
	error_context_stack = &errcallback;

	/* Find smgr relation for buffer */
	if (reln == NULL)
		reln = smgropen(BufTagGetRelFileLocator(&buf->tag), INVALID_PROC_NUMBER);

	TRACE_POSTGRESQL_BUFFER_FLUSH_START(BufTagGetForkNum(&buf->tag),
										buf->tag.blockNum,
										reln->smgr_rlocator.locator.spcOid,
										reln->smgr_rlocator.locator.dbOid,
										reln->smgr_rlocator.locator.relNumber);

	/*
	 * As we hold at least a share-exclusive lock on the buffer, the LSN
	 * cannot change during the flush (and thus can't be torn).
	 */
	recptr = BufferGetLSN(buf);

	/*
	 * Force XLOG flush up to buffer's LSN.  This implements the basic WAL
	 * rule that log updates must hit disk before any of the data-file changes
	 * they describe do.
	 *
	 * However, this rule does not apply to unlogged relations, which will be
	 * lost after a crash anyway.  Most unlogged relation pages do not bear
	 * LSNs since we never emit WAL records for them, and therefore flushing
	 * up through the buffer LSN would be useless, but harmless.  However,
	 * some index AMs use LSNs internally to detect concurrent page
	 * modifications, and therefore unlogged index pages bear "fake" LSNs
	 * generated by XLogGetFakeLSN.  It is unlikely but possible that the fake
	 * LSN counter could advance past the WAL insertion point; and if it did
	 * happen, attempting to flush WAL through that location would fail, with
	 * disastrous system-wide consequences.  To make sure that can't happen,
	 * skip the flush if the buffer isn't permanent.
	 */
	if (pg_atomic_read_u64(&buf->state) & BM_PERMANENT)
		XLogFlush(recptr);

	/*
	 * Now it's safe to write the buffer to disk. Note that no one else should
	 * have been able to write it, while we were busy with log flushing,
	 * because we got the exclusive right to perform I/O by setting the
	 * BM_IO_IN_PROGRESS bit.
	 */
	bufBlock = BufHdrGetBlock(buf);

	/* Update page checksum if desired. */
	PageSetChecksum((Page) bufBlock, buf->tag.blockNum);

	io_start = pgstat_prepare_io_time(track_io_timing);

	smgrwrite(reln,
			  BufTagGetForkNum(&buf->tag),
			  buf->tag.blockNum,
			  bufBlock,
			  false);

	/*
	 * When a strategy is in use, only flushes of dirty buffers already in the
	 * strategy ring are counted as strategy writes (IOCONTEXT
	 * [BULKREAD|BULKWRITE|VACUUM] IOOP_WRITE) for the purpose of IO
	 * statistics tracking.
	 *
	 * If a shared buffer initially added to the ring must be flushed before
	 * being used, this is counted as an IOCONTEXT_NORMAL IOOP_WRITE.
	 *
	 * If a shared buffer which was added to the ring later because the
	 * current strategy buffer is pinned or in use or because all strategy
	 * buffers were dirty and rejected (for BAS_BULKREAD operations only)
	 * requires flushing, this is counted as an IOCONTEXT_NORMAL IOOP_WRITE
	 * (from_ring will be false).
	 *
	 * When a strategy is not in use, the write can only be a "regular" write
	 * of a dirty shared buffer (IOCONTEXT_NORMAL IOOP_WRITE).
	 */
	pgstat_count_io_op_time(io_object, io_context,
							IOOP_WRITE, io_start, 1, BLCKSZ);

	pgBufferUsage.shared_blks_written++;

	/*
	 * Mark the buffer as clean and end the BM_IO_IN_PROGRESS state.
	 */
	TerminateBufferIO(buf, true, 0, true, false);

	TRACE_POSTGRESQL_BUFFER_FLUSH_DONE(BufTagGetForkNum(&buf->tag),
									   buf->tag.blockNum,
									   reln->smgr_rlocator.locator.spcOid,
									   reln->smgr_rlocator.locator.dbOid,
									   reln->smgr_rlocator.locator.relNumber);

	/* Pop the error context stack */
	error_context_stack = errcallback.previous;
}

/*
 * Convenience wrapper around FlushBuffer() that locks/unlocks the buffer
 * before/after calling FlushBuffer().
 */
static void
FlushUnlockedBuffer(BufferDesc *buf, SMgrRelation reln,
					IOObject io_object, IOContext io_context)
{
	Buffer		buffer = BufferDescriptorGetBuffer(buf);

	BufferLockAcquire(buffer, buf, BUFFER_LOCK_SHARE_EXCLUSIVE);
	FlushBuffer(buf, reln, io_object, io_context);
	BufferLockUnlock(buffer, buf);
}

/*
 * RelationGetNumberOfBlocksInFork
 *		Determines the current number of pages in the specified relation fork.
 *
 * Note that the accuracy of the result will depend on the details of the
 * relation's storage. For builtin AMs it'll be accurate, but for external AMs
 * it might not be.
 */
BlockNumber
RelationGetNumberOfBlocksInFork(Relation relation, ForkNumber forkNum)
{
	if (RELKIND_HAS_TABLE_AM(relation->rd_rel->relkind))
	{
		/*
		 * Not every table AM uses BLCKSZ wide fixed size blocks. Therefore
		 * tableam returns the size in bytes - but for the purpose of this
		 * routine, we want the number of blocks. Therefore divide, rounding
		 * up.
		 */
		uint64		szbytes;

		szbytes = table_relation_size(relation, forkNum);

		return (szbytes + (BLCKSZ - 1)) / BLCKSZ;
	}
	else if (RELKIND_HAS_STORAGE(relation->rd_rel->relkind))
	{
		return smgrnblocks(RelationGetSmgr(relation), forkNum);
	}
	else
		Assert(false);

	return 0;					/* keep compiler quiet */
}

/*
 * BufferIsPermanent
 *		Determines whether a buffer will potentially still be around after
 *		a crash.  Caller must hold a buffer pin.
 */
bool
BufferIsPermanent(Buffer buffer)
{
	BufferDesc *bufHdr;

	/* Local buffers are used only for temp relations. */
	if (BufferIsLocal(buffer))
		return false;

	/* Make sure we've got a real buffer, and that we hold a pin on it. */
	Assert(BufferIsValid(buffer));
	Assert(BufferIsPinned(buffer));

	/*
	 * BM_PERMANENT can't be changed while we hold a pin on the buffer, so we
	 * need not bother with the buffer header spinlock.  Even if someone else
	 * changes the buffer header state while we're doing this, the state is
	 * changed atomically, so we'll read the old value or the new value, but
	 * not random garbage.
	 */
	bufHdr = GetBufferDescriptor(buffer - 1);
	return (pg_atomic_read_u64(&bufHdr->state) & BM_PERMANENT) != 0;
}

/*
 * BufferGetLSNAtomic
 *		Retrieves the LSN of the buffer atomically.
 *
 * This is necessary for some callers who may only hold a share lock on
 * the buffer. A share lock allows a concurrent backend to set hint bits
 * on the page, which in turn may require a WAL record to be emitted.
 *
 * On platforms with 8 byte atomic reads/writes, we don't need to do any
 * additional locking. On platforms not supporting such 8 byte atomic
 * reads/writes, we need to actually take the header lock.
 */
XLogRecPtr
BufferGetLSNAtomic(Buffer buffer)
{
	/* Make sure we've got a real buffer, and that we hold a pin on it. */
	Assert(BufferIsValid(buffer));
	Assert(BufferIsPinned(buffer));

#ifdef PG_HAVE_8BYTE_SINGLE_COPY_ATOMICITY
	return PageGetLSN(BufferGetPage(buffer));
#else
	{
		char	   *page = BufferGetPage(buffer);
		BufferDesc *bufHdr;
		XLogRecPtr	lsn;

		/*
		 * If we don't need locking for correctness, fastpath out.
		 */
		if (!XLogHintBitIsNeeded() || BufferIsLocal(buffer))
			return PageGetLSN(page);

		bufHdr = GetBufferDescriptor(buffer - 1);
		LockBufHdr(bufHdr);
		lsn = PageGetLSN(page);
		UnlockBufHdr(bufHdr);

		return lsn;
	}
#endif
}

/* ---------------------------------------------------------------------
 *		DropRelationBuffers
 *
 *		This function removes from the buffer pool all the pages of the
 *		specified relation forks that have block numbers >= firstDelBlock.
 *		(In particular, with firstDelBlock = 0, all pages are removed.)
 *		Dirty pages are simply dropped, without bothering to write them
 *		out first.  Therefore, this is NOT rollback-able, and so should be
 *		used only with extreme caution!
 *
 *		Currently, this is called only from smgr.c when the underlying file
 *		is about to be deleted or truncated (firstDelBlock is needed for
 *		the truncation case).  The data in the affected pages would therefore
 *		be deleted momentarily anyway, and there is no point in writing it.
 *		It is the responsibility of higher-level code to ensure that the
 *		deletion or truncation does not lose any data that could be needed
 *		later.  It is also the responsibility of higher-level code to ensure
 *		that no other process could be trying to load more pages of the
 *		relation into buffers.
 * --------------------------------------------------------------------
 */
void
DropRelationBuffers(SMgrRelation smgr_reln, ForkNumber *forkNum,
					int nforks, BlockNumber *firstDelBlock)
{
	int			i;
	int			j;
	RelFileLocatorBackend rlocator;
	BlockNumber nForkBlock[MAX_FORKNUM];
	uint64		nBlocksToInvalidate = 0;

	rlocator = smgr_reln->smgr_rlocator;

	/* If it's a local relation, it's localbuf.c's problem. */
	if (RelFileLocatorBackendIsTemp(rlocator))
	{
		if (rlocator.backend == MyProcNumber)
			DropRelationLocalBuffers(rlocator.locator, forkNum, nforks,
									 firstDelBlock);

		return;
	}

	/*
	 * To remove all the pages of the specified relation forks from the buffer
	 * pool, we need to scan the entire buffer pool but we can optimize it by
	 * finding the buffers from BufMapping table provided we know the exact
	 * size of each fork of the relation. The exact size is required to ensure
	 * that we don't leave any buffer for the relation being dropped as
	 * otherwise the background writer or checkpointer can lead to a PANIC
	 * error while flushing buffers corresponding to files that don't exist.
	 *
	 * To know the exact size, we rely on the size cached for each fork by us
	 * during recovery which limits the optimization to recovery and on
	 * standbys but we can easily extend it once we have shared cache for
	 * relation size.
	 *
	 * In recovery, we cache the value returned by the first lseek(SEEK_END)
	 * and the future writes keeps the cached value up-to-date. See
	 * smgrextend. It is possible that the value of the first lseek is smaller
	 * than the actual number of existing blocks in the file due to buggy
	 * Linux kernels that might not have accounted for the recent write. But
	 * that should be fine because there must not be any buffers after that
	 * file size.
	 */
	for (i = 0; i < nforks; i++)
	{
		/* Get the number of blocks for a relation's fork */
		nForkBlock[i] = smgrnblocks_cached(smgr_reln, forkNum[i]);

		if (nForkBlock[i] == InvalidBlockNumber)
		{
			nBlocksToInvalidate = InvalidBlockNumber;
			break;
		}

		/* calculate the number of blocks to be invalidated */
		nBlocksToInvalidate += (nForkBlock[i] - firstDelBlock[i]);
	}

	/*
	 * We apply the optimization iff the total number of blocks to invalidate
	 * is below the BUF_DROP_FULL_SCAN_THRESHOLD.
	 */
	if (BlockNumberIsValid(nBlocksToInvalidate) &&
		nBlocksToInvalidate < BUF_DROP_FULL_SCAN_THRESHOLD)
	{
		for (j = 0; j < nforks; j++)
			FindAndDropRelationBuffers(rlocator.locator, forkNum[j],
									   nForkBlock[j], firstDelBlock[j]);
		return;
	}

	for (i = 0; i < NBuffers; i++)
	{
		BufferDesc *bufHdr = GetBufferDescriptor(i);

		/*
		 * We can make this a tad faster by prechecking the buffer tag before
		 * we attempt to lock the buffer; this saves a lot of lock
		 * acquisitions in typical cases.  It should be safe because the
		 * caller must have AccessExclusiveLock on the relation, or some other
		 * reason to be certain that no one is loading new pages of the rel
		 * into the buffer pool.  (Otherwise we might well miss such pages
		 * entirely.)  Therefore, while the tag might be changing while we
		 * look at it, it can't be changing *to* a value we care about, only
		 * *away* from such a value.  So false negatives are impossible, and
		 * false positives are safe because we'll recheck after getting the
		 * buffer lock.
		 *
		 * We could check forkNum and blockNum as well as the rlocator, but
		 * the incremental win from doing so seems small.
		 */
		if (!BufTagMatchesRelFileLocator(&bufHdr->tag, &rlocator.locator))
			continue;

		LockBufHdr(bufHdr);

		for (j = 0; j < nforks; j++)
		{
			if (BufTagMatchesRelFileLocator(&bufHdr->tag, &rlocator.locator) &&
				BufTagGetForkNum(&bufHdr->tag) == forkNum[j] &&
				bufHdr->tag.blockNum >= firstDelBlock[j])
			{
				InvalidateBuffer(bufHdr);	/* releases spinlock */
				break;
			}
		}
		if (j >= nforks)
			UnlockBufHdr(bufHdr);
	}
}

/* ---------------------------------------------------------------------
 *		DropRelationsAllBuffers
 *
 *		This function removes from the buffer pool all the pages of all
 *		forks of the specified relations.  It's equivalent to calling
 *		DropRelationBuffers once per fork per relation with firstDelBlock = 0.
 *		--------------------------------------------------------------------
 */
void
DropRelationsAllBuffers(SMgrRelation *smgr_reln, int nlocators)
{
	int			i;
	int			n = 0;
	SMgrRelation *rels;
	BlockNumber (*block)[MAX_FORKNUM + 1];
	uint64		nBlocksToInvalidate = 0;
	RelFileLocator *locators;
	bool		cached = true;
	bool		use_bsearch;

	if (nlocators == 0)
		return;

	rels = palloc_array(SMgrRelation, nlocators);	/* non-local relations */

	/* If it's a local relation, it's localbuf.c's problem. */
	for (i = 0; i < nlocators; i++)
	{
		if (RelFileLocatorBackendIsTemp(smgr_reln[i]->smgr_rlocator))
		{
			if (smgr_reln[i]->smgr_rlocator.backend == MyProcNumber)
				DropRelationAllLocalBuffers(smgr_reln[i]->smgr_rlocator.locator);
		}
		else
			rels[n++] = smgr_reln[i];
	}

	/*
	 * If there are no non-local relations, then we're done. Release the
	 * memory and return.
	 */
	if (n == 0)
	{
		pfree(rels);
		return;
	}

	/*
	 * This is used to remember the number of blocks for all the relations
	 * forks.
	 */
	block = (BlockNumber (*)[MAX_FORKNUM + 1])
		palloc(sizeof(BlockNumber) * n * (MAX_FORKNUM + 1));

	/*
	 * We can avoid scanning the entire buffer pool if we know the exact size
	 * of each of the given relation forks. See DropRelationBuffers.
	 */
	for (i = 0; i < n && cached; i++)
	{
		for (int j = 0; j <= MAX_FORKNUM; j++)
		{
			/* Get the number of blocks for a relation's fork. */
			block[i][j] = smgrnblocks_cached(rels[i], j);

			/* We need to only consider the relation forks that exists. */
			if (block[i][j] == InvalidBlockNumber)
			{
				if (!smgrexists(rels[i], j))
					continue;
				cached = false;
				break;
			}

			/* calculate the total number of blocks to be invalidated */
			nBlocksToInvalidate += block[i][j];
		}
	}

	/*
	 * We apply the optimization iff the total number of blocks to invalidate
	 * is below the BUF_DROP_FULL_SCAN_THRESHOLD.
	 */
	if (cached && nBlocksToInvalidate < BUF_DROP_FULL_SCAN_THRESHOLD)
	{
		for (i = 0; i < n; i++)
		{
			for (int j = 0; j <= MAX_FORKNUM; j++)
			{
				/* ignore relation forks that doesn't exist */
				if (!BlockNumberIsValid(block[i][j]))
					continue;

				/* drop all the buffers for a particular relation fork */
				FindAndDropRelationBuffers(rels[i]->smgr_rlocator.locator,
										   j, block[i][j], 0);
			}
		}

		pfree(block);
		pfree(rels);
		return;
	}

	pfree(block);
	locators = palloc_array(RelFileLocator, n); /* non-local relations */
	for (i = 0; i < n; i++)
		locators[i] = rels[i]->smgr_rlocator.locator;

	/*
	 * For low number of relations to drop just use a simple walk through, to
	 * save the bsearch overhead. The threshold to use is rather a guess than
	 * an exactly determined value, as it depends on many factors (CPU and RAM
	 * speeds, amount of shared buffers etc.).
	 */
	use_bsearch = n > RELS_BSEARCH_THRESHOLD;

	/* sort the list of rlocators if necessary */
	if (use_bsearch)
		qsort(locators, n, sizeof(RelFileLocator), rlocator_comparator);

	for (i = 0; i < NBuffers; i++)
	{
		RelFileLocator *rlocator = NULL;
		BufferDesc *bufHdr = GetBufferDescriptor(i);

		/*
		 * As in DropRelationBuffers, an unlocked precheck should be safe and
		 * saves some cycles.
		 */

		if (!use_bsearch)
		{
			int			j;

			for (j = 0; j < n; j++)
			{
				if (BufTagMatchesRelFileLocator(&bufHdr->tag, &locators[j]))
				{
					rlocator = &locators[j];
					break;
				}
			}
		}
		else
		{
			RelFileLocator locator;

			locator = BufTagGetRelFileLocator(&bufHdr->tag);
			rlocator = bsearch(&locator,
							   locators, n, sizeof(RelFileLocator),
							   rlocator_comparator);
		}

		/* buffer doesn't belong to any of the given relfilelocators; skip it */
		if (rlocator == NULL)
			continue;

		LockBufHdr(bufHdr);
		if (BufTagMatchesRelFileLocator(&bufHdr->tag, rlocator))
			InvalidateBuffer(bufHdr);	/* releases spinlock */
		else
			UnlockBufHdr(bufHdr);
	}

	pfree(locators);
	pfree(rels);
}

/* ---------------------------------------------------------------------
 *		FindAndDropRelationBuffers
 *
 *		This function performs look up in BufMapping table and removes from the
 *		buffer pool all the pages of the specified relation fork that has block
 *		number >= firstDelBlock. (In particular, with firstDelBlock = 0, all
 *		pages are removed.)
 * --------------------------------------------------------------------
 */
static void
FindAndDropRelationBuffers(RelFileLocator rlocator, ForkNumber forkNum,
						   BlockNumber nForkBlock,
						   BlockNumber firstDelBlock)
{
	BlockNumber curBlock;

	for (curBlock = firstDelBlock; curBlock < nForkBlock; curBlock++)
	{
		uint32		bufHash;	/* hash value for tag */
		BufferTag	bufTag;		/* identity of requested block */
		LWLock	   *bufPartitionLock;	/* buffer partition lock for it */
		int			buf_id;
		BufferDesc *bufHdr;

		/* create a tag so we can lookup the buffer */
		InitBufferTag(&bufTag, &rlocator, forkNum, curBlock);

		/* determine its hash code and partition lock ID */
		bufHash = BufTableHashCode(&bufTag);
		bufPartitionLock = BufMappingPartitionLock(bufHash);

		/* Check that it is in the buffer pool. If not, do nothing. */
		LWLockAcquire(bufPartitionLock, LW_SHARED);
		buf_id = BufTableLookup(&bufTag, bufHash);
		LWLockRelease(bufPartitionLock);

		if (buf_id < 0)
			continue;

		bufHdr = GetBufferDescriptor(buf_id);

		/*
		 * We need to lock the buffer header and recheck if the buffer is
		 * still associated with the same block because the buffer could be
		 * evicted by some other backend loading blocks for a different
		 * relation after we release lock on the BufMapping table.
		 */
		LockBufHdr(bufHdr);

		if (BufTagMatchesRelFileLocator(&bufHdr->tag, &rlocator) &&
			BufTagGetForkNum(&bufHdr->tag) == forkNum &&
			bufHdr->tag.blockNum >= firstDelBlock)
			InvalidateBuffer(bufHdr);	/* releases spinlock */
		else
			UnlockBufHdr(bufHdr);
	}
}

/* ---------------------------------------------------------------------
 *		DropDatabaseBuffers
 *
 *		This function removes all the buffers in the buffer cache for a
 *		particular database.  Dirty pages are simply dropped, without
 *		bothering to write them out first.  This is used when we destroy a
 *		database, to avoid trying to flush data to disk when the directory
 *		tree no longer exists.  Implementation is pretty similar to
 *		DropRelationBuffers() which is for destroying just one relation.
 * --------------------------------------------------------------------
 */
void
DropDatabaseBuffers(Oid dbid)
{
	int			i;

	/*
	 * We needn't consider local buffers, since by assumption the target
	 * database isn't our own.
	 */

	for (i = 0; i < NBuffers; i++)
	{
		BufferDesc *bufHdr = GetBufferDescriptor(i);

		/*
		 * As in DropRelationBuffers, an unlocked precheck should be safe and
		 * saves some cycles.
		 */
		if (bufHdr->tag.dbOid != dbid)
			continue;

		LockBufHdr(bufHdr);
		if (bufHdr->tag.dbOid == dbid)
			InvalidateBuffer(bufHdr);	/* releases spinlock */
		else
			UnlockBufHdr(bufHdr);
	}
}

/* ---------------------------------------------------------------------
 *		FlushRelationBuffers
 *
 *		This function writes all dirty pages of a relation out to disk
 *		(or more accurately, out to kernel disk buffers), ensuring that the
 *		kernel has an up-to-date view of the relation.
 *
 *		Generally, the caller should be holding AccessExclusiveLock on the
 *		target relation to ensure that no other backend is busy dirtying
 *		more blocks of the relation; the effects can't be expected to last
 *		after the lock is released.
 *
 *		XXX currently it sequentially searches the buffer pool, should be
 *		changed to more clever ways of searching.  This routine is not
 *		used in any performance-critical code paths, so it's not worth
 *		adding additional overhead to normal paths to make it go faster.
 * --------------------------------------------------------------------
 */
void
FlushRelationBuffers(Relation rel)
{
	int			i;
	BufferDesc *bufHdr;
	SMgrRelation srel = RelationGetSmgr(rel);

	if (RelationUsesLocalBuffers(rel))
	{
		for (i = 0; i < NLocBuffer; i++)
		{
			uint64		buf_state;

			bufHdr = GetLocalBufferDescriptor(i);
			if (BufTagMatchesRelFileLocator(&bufHdr->tag, &rel->rd_locator) &&
				((buf_state = pg_atomic_read_u64(&bufHdr->state)) &
				 (BM_VALID | BM_DIRTY)) == (BM_VALID | BM_DIRTY))
			{
				ErrorContextCallback errcallback;

				/* Setup error traceback support for ereport() */
				errcallback.callback = local_buffer_write_error_callback;
				errcallback.arg = bufHdr;
				errcallback.previous = error_context_stack;
				error_context_stack = &errcallback;

				/* Make sure we can handle the pin */
				ReservePrivateRefCountEntry();
				ResourceOwnerEnlarge(CurrentResourceOwner);

				/*
				 * Pin/unpin mostly to make valgrind work, but it also seems
				 * like the right thing to do.
				 */
				PinLocalBuffer(bufHdr, false);


				FlushLocalBuffer(bufHdr, srel);

				UnpinLocalBuffer(BufferDescriptorGetBuffer(bufHdr));

				/* Pop the error context stack */
				error_context_stack = errcallback.previous;
			}
		}

		return;
	}

	for (i = 0; i < NBuffers; i++)
	{
		uint64		buf_state;

		bufHdr = GetBufferDescriptor(i);

		/*
		 * As in DropRelationBuffers, an unlocked precheck should be safe and
		 * saves some cycles.
		 */
		if (!BufTagMatchesRelFileLocator(&bufHdr->tag, &rel->rd_locator))
			continue;

		/* Make sure we can handle the pin */
		ReservePrivateRefCountEntry();
		ResourceOwnerEnlarge(CurrentResourceOwner);

		buf_state = LockBufHdr(bufHdr);
		if (BufTagMatchesRelFileLocator(&bufHdr->tag, &rel->rd_locator) &&
			(buf_state & (BM_VALID | BM_DIRTY)) == (BM_VALID | BM_DIRTY))
		{
			PinBuffer_Locked(bufHdr);
			FlushUnlockedBuffer(bufHdr, srel, IOOBJECT_RELATION, IOCONTEXT_NORMAL);
			UnpinBuffer(bufHdr);
		}
		else
			UnlockBufHdr(bufHdr);
	}
}

/* ---------------------------------------------------------------------
 *		FlushRelationsAllBuffers
 *
 *		This function flushes out of the buffer pool all the pages of all
 *		forks of the specified smgr relations.  It's equivalent to calling
 *		FlushRelationBuffers once per relation.  The relations are assumed not
 *		to use local buffers.
 * --------------------------------------------------------------------
 */
void
FlushRelationsAllBuffers(SMgrRelation *smgrs, int nrels)
{
	int			i;
	SMgrSortArray *srels;
	bool		use_bsearch;

	if (nrels == 0)
		return;

	/* fill-in array for qsort */
	srels = palloc_array(SMgrSortArray, nrels);

	for (i = 0; i < nrels; i++)
	{
		Assert(!RelFileLocatorBackendIsTemp(smgrs[i]->smgr_rlocator));

		srels[i].rlocator = smgrs[i]->smgr_rlocator.locator;
		srels[i].srel = smgrs[i];
	}

	/*
	 * Save the bsearch overhead for low number of relations to sync. See
	 * DropRelationsAllBuffers for details.
	 */
	use_bsearch = nrels > RELS_BSEARCH_THRESHOLD;

	/* sort the list of SMgrRelations if necessary */
	if (use_bsearch)
		qsort(srels, nrels, sizeof(SMgrSortArray), rlocator_comparator);

	for (i = 0; i < NBuffers; i++)
	{
		SMgrSortArray *srelent = NULL;
		BufferDesc *bufHdr = GetBufferDescriptor(i);
		uint64		buf_state;

		/*
		 * As in DropRelationBuffers, an unlocked precheck should be safe and
		 * saves some cycles.
		 */

		if (!use_bsearch)
		{
			int			j;

			for (j = 0; j < nrels; j++)
			{
				if (BufTagMatchesRelFileLocator(&bufHdr->tag, &srels[j].rlocator))
				{
					srelent = &srels[j];
					break;
				}
			}
		}
		else
		{
			RelFileLocator rlocator;

			rlocator = BufTagGetRelFileLocator(&bufHdr->tag);
			srelent = bsearch(&rlocator,
							  srels, nrels, sizeof(SMgrSortArray),
							  rlocator_comparator);
		}

		/* buffer doesn't belong to any of the given relfilelocators; skip it */
		if (srelent == NULL)
			continue;

		/* Make sure we can handle the pin */
		ReservePrivateRefCountEntry();
		ResourceOwnerEnlarge(CurrentResourceOwner);

		buf_state = LockBufHdr(bufHdr);
		if (BufTagMatchesRelFileLocator(&bufHdr->tag, &srelent->rlocator) &&
			(buf_state & (BM_VALID | BM_DIRTY)) == (BM_VALID | BM_DIRTY))
		{
			PinBuffer_Locked(bufHdr);
			FlushUnlockedBuffer(bufHdr, srelent->srel, IOOBJECT_RELATION, IOCONTEXT_NORMAL);
			UnpinBuffer(bufHdr);
		}
		else
			UnlockBufHdr(bufHdr);
	}

	pfree(srels);
}

/* ---------------------------------------------------------------------
 *		RelationCopyStorageUsingBuffer
 *
 *		Copy fork's data using bufmgr.  Same as RelationCopyStorage but instead
 *		of using smgrread and smgrextend this will copy using bufmgr APIs.
 *
 *		Refer comments atop CreateAndCopyRelationData() for details about
 *		'permanent' parameter.
 * --------------------------------------------------------------------
 */
static void
RelationCopyStorageUsingBuffer(RelFileLocator srclocator,
							   RelFileLocator dstlocator,
							   ForkNumber forkNum, bool permanent)
{
	Buffer		srcBuf;
	Buffer		dstBuf;
	Page		srcPage;
	Page		dstPage;
	bool		use_wal;
	BlockNumber nblocks;
	BlockNumber blkno;
	PGIOAlignedBlock buf;
	BufferAccessStrategy bstrategy_src;
	BufferAccessStrategy bstrategy_dst;
	BlockRangeReadStreamPrivate p;
	ReadStream *src_stream;
	SMgrRelation src_smgr;

	/*
	 * In general, we want to write WAL whenever wal_level > 'minimal', but we
	 * can skip it when copying any fork of an unlogged relation other than
	 * the init fork.
	 */
	use_wal = XLogIsNeeded() && (permanent || forkNum == INIT_FORKNUM);

	/* Get number of blocks in the source relation. */
	nblocks = smgrnblocks(smgropen(srclocator, INVALID_PROC_NUMBER),
						  forkNum);

	/* Nothing to copy; just return. */
	if (nblocks == 0)
		return;

	/*
	 * Bulk extend the destination relation of the same size as the source
	 * relation before starting to copy block by block.
	 */
	memset(buf.data, 0, BLCKSZ);
	smgrextend(smgropen(dstlocator, INVALID_PROC_NUMBER), forkNum, nblocks - 1,
			   buf.data, true);

	/* This is a bulk operation, so use buffer access strategies. */
	bstrategy_src = GetAccessStrategy(BAS_BULKREAD);
	bstrategy_dst = GetAccessStrategy(BAS_BULKWRITE);

	/* Initialize streaming read */
	p.current_blocknum = 0;
	p.last_exclusive = nblocks;
	src_smgr = smgropen(srclocator, INVALID_PROC_NUMBER);

	/*
	 * It is safe to use batchmode as block_range_read_stream_cb takes no
	 * locks.
	 */
	src_stream = read_stream_begin_smgr_relation(READ_STREAM_FULL |
												 READ_STREAM_USE_BATCHING,
												 bstrategy_src,
												 src_smgr,
												 permanent ? RELPERSISTENCE_PERMANENT : RELPERSISTENCE_UNLOGGED,
												 forkNum,
												 block_range_read_stream_cb,
												 &p,
												 0);

	/* Iterate over each block of the source relation file. */
	for (blkno = 0; blkno < nblocks; blkno++)
	{
		CHECK_FOR_INTERRUPTS();

		/* Read block from source relation. */
		srcBuf = read_stream_next_buffer(src_stream, NULL);
		LockBuffer(srcBuf, BUFFER_LOCK_SHARE);
		srcPage = BufferGetPage(srcBuf);

		dstBuf = ReadBufferWithoutRelcache(dstlocator, forkNum,
										   BufferGetBlockNumber(srcBuf),
										   RBM_ZERO_AND_LOCK, bstrategy_dst,
										   permanent);
		dstPage = BufferGetPage(dstBuf);

		START_CRIT_SECTION();

		/* Copy page data from the source to the destination. */
		memcpy(dstPage, srcPage, BLCKSZ);
		MarkBufferDirty(dstBuf);

		/* WAL-log the copied page. */
		if (use_wal)
			log_newpage_buffer(dstBuf, true);

		END_CRIT_SECTION();

		UnlockReleaseBuffer(dstBuf);
		UnlockReleaseBuffer(srcBuf);
	}
	Assert(read_stream_next_buffer(src_stream, NULL) == InvalidBuffer);
	read_stream_end(src_stream);

	FreeAccessStrategy(bstrategy_src);
	FreeAccessStrategy(bstrategy_dst);
}

/* ---------------------------------------------------------------------
 *		CreateAndCopyRelationData
 *
 *		Create destination relation storage and copy all forks from the
 *		source relation to the destination.
 *
 *		Pass permanent as true for permanent relations and false for
 *		unlogged relations.  Currently this API is not supported for
 *		temporary relations.
 * --------------------------------------------------------------------
 */
void
CreateAndCopyRelationData(RelFileLocator src_rlocator,
						  RelFileLocator dst_rlocator, bool permanent)
{
	char		relpersistence;
	SMgrRelation src_rel;
	SMgrRelation dst_rel;

	/* Set the relpersistence. */
	relpersistence = permanent ?
		RELPERSISTENCE_PERMANENT : RELPERSISTENCE_UNLOGGED;

	src_rel = smgropen(src_rlocator, INVALID_PROC_NUMBER);
	dst_rel = smgropen(dst_rlocator, INVALID_PROC_NUMBER);

	/*
	 * Create and copy all forks of the relation.  During create database we
	 * have a separate cleanup mechanism which deletes complete database
	 * directory.  Therefore, each individual relation doesn't need to be
	 * registered for cleanup.
	 */
	RelationCreateStorage(dst_rlocator, relpersistence, false);

	/* copy main fork. */
	RelationCopyStorageUsingBuffer(src_rlocator, dst_rlocator, MAIN_FORKNUM,
								   permanent);

	/* copy those extra forks that exist */
	for (ForkNumber forkNum = MAIN_FORKNUM + 1;
		 forkNum <= MAX_FORKNUM; forkNum++)
	{
		if (smgrexists(src_rel, forkNum))
		{
			smgrcreate(dst_rel, forkNum, false);

			/*
			 * WAL log creation if the relation is persistent, or this is the
			 * init fork of an unlogged relation.
			 */
			if (permanent || forkNum == INIT_FORKNUM)
				log_smgrcreate(&dst_rlocator, forkNum);

			/* Copy a fork's data, block by block. */
			RelationCopyStorageUsingBuffer(src_rlocator, dst_rlocator, forkNum,
										   permanent);
		}
	}
}

/* ---------------------------------------------------------------------
 *		FlushDatabaseBuffers
 *
 *		This function writes all dirty pages of a database out to disk
 *		(or more accurately, out to kernel disk buffers), ensuring that the
 *		kernel has an up-to-date view of the database.
 *
 *		Generally, the caller should be holding an appropriate lock to ensure
 *		no other backend is active in the target database; otherwise more
 *		pages could get dirtied.
 *
 *		Note we don't worry about flushing any pages of temporary relations.
 *		It's assumed these wouldn't be interesting.
 * --------------------------------------------------------------------
 */
void
FlushDatabaseBuffers(Oid dbid)
{
	int			i;
	BufferDesc *bufHdr;

	for (i = 0; i < NBuffers; i++)
	{
		uint64		buf_state;

		bufHdr = GetBufferDescriptor(i);

		/*
		 * As in DropRelationBuffers, an unlocked precheck should be safe and
		 * saves some cycles.
		 */
		if (bufHdr->tag.dbOid != dbid)
			continue;

		/* Make sure we can handle the pin */
		ReservePrivateRefCountEntry();
		ResourceOwnerEnlarge(CurrentResourceOwner);

		buf_state = LockBufHdr(bufHdr);
		if (bufHdr->tag.dbOid == dbid &&
			(buf_state & (BM_VALID | BM_DIRTY)) == (BM_VALID | BM_DIRTY))
		{
			PinBuffer_Locked(bufHdr);
			FlushUnlockedBuffer(bufHdr, NULL, IOOBJECT_RELATION, IOCONTEXT_NORMAL);
			UnpinBuffer(bufHdr);
		}
		else
			UnlockBufHdr(bufHdr);
	}
}

/*
 * Flush a previously, share-exclusively or exclusively, locked and pinned
 * buffer to the OS.
 */
void
FlushOneBuffer(Buffer buffer)
{
	BufferDesc *bufHdr;

	/* currently not needed, but no fundamental reason not to support */
	Assert(!BufferIsLocal(buffer));

	Assert(BufferIsPinned(buffer));

	bufHdr = GetBufferDescriptor(buffer - 1);

	Assert(BufferIsLockedByMe(buffer));

	FlushBuffer(bufHdr, NULL, IOOBJECT_RELATION, IOCONTEXT_NORMAL);
}

/*
 * ReleaseBuffer -- release the pin on a buffer
 */
void
ReleaseBuffer(Buffer buffer)
{
	if (!BufferIsValid(buffer))
		elog(ERROR, "bad buffer ID: %d", buffer);

	if (BufferIsLocal(buffer))
		UnpinLocalBuffer(buffer);
	else
		UnpinBuffer(GetBufferDescriptor(buffer - 1));
}

/*
 * UnlockReleaseBuffer -- release the content lock and pin on a buffer
 *
 * This is just a, more efficient, shorthand for a common combination.
 */
void
UnlockReleaseBuffer(Buffer buffer)
{
	int			mode;
	BufferDesc *buf;
	PrivateRefCountEntry *ref;
	uint64		sub;
	uint64		lockstate;

	Assert(BufferIsPinned(buffer));

	if (BufferIsLocal(buffer))
	{
		UnpinLocalBuffer(buffer);
		return;
	}

	ResourceOwnerForgetBuffer(CurrentResourceOwner, buffer);

	buf = GetBufferDescriptor(buffer - 1);

	mode = BufferLockDisownInternal(buffer, buf);

	/* compute state modification for lock release */
	sub = BufferLockReleaseSub(mode);

	/* compute state modification for pin release */
	ref = GetPrivateRefCountEntry(buffer, false);
	Assert(ref != NULL);
	Assert(ref->data.refcount > 0);
	ref->data.refcount--;

	/* no more backend local pins, reduce shared pin count */
	if (likely(ref->data.refcount == 0))
	{
		/* See comment in UnpinBufferNoOwner() */
		VALGRIND_MAKE_MEM_NOACCESS(BufHdrGetBlock(buf), BLCKSZ);

		sub |= BUF_REFCOUNT_ONE;
		ForgetPrivateRefCountEntry(ref);
	}

	/* perform the lock and pin release in one atomic op */
	lockstate = pg_atomic_sub_fetch_u64(&buf->state, sub);

	/* wake up waiters for the lock */
	BufferLockProcessRelease(buf, mode, lockstate);

	/* wake up waiter for the pin release */
	if (lockstate & BM_PIN_COUNT_WAITER)
		WakePinCountWaiter(buf);

	/*
	 * Now okay to allow cancel/die interrupts again, which were held when the
	 * lock was acquired.
	 */
	RESUME_INTERRUPTS();
}

/*
 * IncrBufferRefCount
 *		Increment the pin count on a buffer that we have *already* pinned
 *		at least once.
 *
 *		This function cannot be used on a buffer we do not have pinned,
 *		because it doesn't change the shared buffer state.
 */
void
IncrBufferRefCount(Buffer buffer)
{
	Assert(BufferIsPinned(buffer));
	ResourceOwnerEnlarge(CurrentResourceOwner);
	if (BufferIsLocal(buffer))
		LocalRefCount[-buffer - 1]++;
	else
	{
		PrivateRefCountEntry *ref;

		ref = GetPrivateRefCountEntry(buffer, true);
		Assert(ref != NULL);
		ref->data.refcount++;
	}
	ResourceOwnerRememberBuffer(CurrentResourceOwner, buffer);
}

/*
 * Shared-buffer only helper for MarkBufferDirtyHint() and
 * BufferSetHintBits16().
 *
 * This is separated out because it turns out that the repeated checks for
 * local buffers, repeated GetBufferDescriptor() and repeated reading of the
 * buffer's state sufficiently hurts the performance of BufferSetHintBits16().
 */
static inline void
MarkSharedBufferDirtyHint(Buffer buffer, BufferDesc *bufHdr, uint64 lockstate,
						  bool buffer_std)
{
	Page		page = BufferGetPage(buffer);

	Assert(GetPrivateRefCount(buffer) > 0);

	/* here, either share-exclusive or exclusive lock is OK */
	Assert(BufferLockHeldByMeInMode(bufHdr, BUFFER_LOCK_EXCLUSIVE) ||
		   BufferLockHeldByMeInMode(bufHdr, BUFFER_LOCK_SHARE_EXCLUSIVE));

	/*
	 * This routine might get called many times on the same page, if we are
	 * making the first scan after commit of an xact that added/deleted many
	 * tuples. So, be as quick as we can if the buffer is already dirty.
	 *
	 * As we are holding (at least) a share-exclusive lock, nobody could have
	 * cleaned or dirtied the page concurrently, so we can just rely on the
	 * previously fetched value here without any danger of races.
	 */
	if (unlikely(!(lockstate & BM_DIRTY)))
	{
		XLogRecPtr	lsn = InvalidXLogRecPtr;
		bool		wal_log = false;
		uint64		buf_state;

		/*
		 * If we need to protect hint bit updates from torn writes, WAL-log a
		 * full page image of the page. This full page image is only necessary
		 * if the hint bit update is the first change to the page since the
		 * last checkpoint.
		 *
		 * We don't check full_page_writes here because that logic is included
		 * when we call XLogInsert() since the value changes dynamically.
		 */
		if (XLogHintBitIsNeeded() && (lockstate & BM_PERMANENT))
		{
			/*
			 * If we must not write WAL, due to a relfilelocator-specific
			 * condition or being in recovery, don't dirty the page.  We can
			 * set the hint, just not dirty the page as a result so the hint
			 * is lost when we evict the page or shutdown.
			 *
			 * See src/backend/storage/page/README for longer discussion.
			 */
			if (RecoveryInProgress() ||
				RelFileLocatorSkippingWAL(BufTagGetRelFileLocator(&bufHdr->tag)))
				return;

			wal_log = true;
		}

		/*
		 * We must mark the page dirty before we emit the WAL record, as per
		 * the usual rules, to ensure that BufferSync()/SyncOneBuffer() try to
		 * flush the buffer, even if we haven't inserted the WAL record yet.
		 * As we hold at least a share-exclusive lock, checkpoints will wait
		 * for this backend to be done with the buffer before continuing. If
		 * we did it the other way round, a checkpoint could start between
		 * writing the WAL record and marking the buffer dirty.
		 */
		buf_state = LockBufHdr(bufHdr);

		/*
		 * It should not be possible for the buffer to already be dirty, see
		 * comment above.
		 */
		Assert(!(buf_state & BM_DIRTY));
		Assert(BUF_STATE_GET_REFCOUNT(buf_state) > 0);
		UnlockBufHdrExt(bufHdr, buf_state,
						BM_DIRTY,
						0, 0);

		/*
		 * If the block is already dirty because we either made a change or
		 * set a hint already, then we don't need to write a full page image.
		 * Note that aggressive cleaning of blocks dirtied by hint bit setting
		 * would increase the call rate. Bulk setting of hint bits would
		 * reduce the call rate...
		 */
		if (wal_log)
			lsn = XLogSaveBufferForHint(buffer, buffer_std);

		if (XLogRecPtrIsValid(lsn))
		{
			/*
			 * Set the page LSN if we wrote a backup block. To allow backends
			 * that only hold a share lock on the buffer to read the LSN in a
			 * tear-free manner, we set the page LSN while holding the buffer
			 * header lock. This allows any reader of an LSN who holds only a
			 * share lock to also obtain a buffer header lock before using
			 * PageGetLSN() to read the LSN in a tear free way. This is done
			 * in BufferGetLSNAtomic().
			 *
			 * If checksums are enabled, you might think we should reset the
			 * checksum here. That will happen when the page is written
			 * sometime later in this checkpoint cycle.
			 */
			buf_state = LockBufHdr(bufHdr);
			PageSetLSN(page, lsn);
			UnlockBufHdr(bufHdr);
		}

		pgBufferUsage.shared_blks_dirtied++;
		if (VacuumCostActive)
			VacuumCostBalance += VacuumCostPageDirty;
	}
}

/*
 * MarkBufferDirtyHint
 *
 *	Mark a buffer dirty for non-critical changes.
 *
 * This is essentially the same as MarkBufferDirty, except:
 *
 * 1. The caller does not write WAL; so if checksums are enabled, we may need
 *	  to write an XLOG_FPI_FOR_HINT WAL record to protect against torn pages.
 * 2. The caller might have only a share-exclusive-lock instead of an
 *	  exclusive-lock on the buffer's content lock.
 * 3. This function does not guarantee that the buffer is always marked dirty
 *	  (it e.g. can't always on a hot standby), so it cannot be used for
 *	  important changes.
 */
inline void
MarkBufferDirtyHint(Buffer buffer, bool buffer_std)
{
	BufferDesc *bufHdr;

	if (!BufferIsValid(buffer))
		elog(ERROR, "bad buffer ID: %d", buffer);

	if (BufferIsLocal(buffer))
	{
		MarkLocalBufferDirty(buffer);
		return;
	}

	bufHdr = GetBufferDescriptor(buffer - 1);

	MarkSharedBufferDirtyHint(buffer, bufHdr,
							  pg_atomic_read_u64(&bufHdr->state),
							  buffer_std);
}

/*
 * Release buffer content locks for shared buffers.
 *
 * Used to clean up after errors.
 *
 * Currently, we can expect that resource owner cleanup, via
 * ResOwnerReleaseBuffer(), took care of releasing buffer content locks per
 * se; the only thing we need to deal with here is clearing any PIN_COUNT
 * request that was in progress.
 */
void
UnlockBuffers(void)
{
	BufferDesc *buf = PinCountWaitBuf;

	if (buf)
	{
		uint64		buf_state;
		uint64		unset_bits = 0;

		buf_state = LockBufHdr(buf);

		/*
		 * Don't complain if flag bit not set; it could have been reset but we
		 * got a cancel/die interrupt before getting the signal.
		 */
		if ((buf_state & BM_PIN_COUNT_WAITER) != 0 &&
			buf->wait_backend_pgprocno == MyProcNumber)
			unset_bits = BM_PIN_COUNT_WAITER;

		UnlockBufHdrExt(buf, buf_state,
						0, unset_bits,
						0);

		PinCountWaitBuf = NULL;
	}
}

/*
 * Acquire the buffer content lock in the specified mode
 *
 * If the lock is not available, sleep until it is.
 *
 * Side effect: cancel/die interrupts are held off until lock release.
 *
 * This uses almost the same locking approach as lwlock.c's
 * LWLockAcquire(). See documentation at the top of lwlock.c for a more
 * detailed discussion.
 *
 * The reason that this, and most of the other BufferLock* functions, get both
 * the Buffer and BufferDesc* as parameters, is that looking up one from the
 * other repeatedly shows up noticeably in profiles.
 *
 * Callers should provide a constant for mode, for more efficient code
 * generation.
 */
static inline void
BufferLockAcquire(Buffer buffer, BufferDesc *buf_hdr, BufferLockMode mode)
{
	PrivateRefCountEntry *entry;
	int			extraWaits = 0;

	/*
	 * Get reference to the refcount entry before we hold the lock, it seems
	 * better to do before holding the lock.
	 */
	entry = GetPrivateRefCountEntry(buffer, true);

	/*
	 * We better not already hold a lock on the buffer.
	 */
	Assert(entry->data.lockmode == BUFFER_LOCK_UNLOCK);

	/*
	 * Lock out cancel/die interrupts until we exit the code section protected
	 * by the content lock.  This ensures that interrupts will not interfere
	 * with manipulations of data structures in shared memory.
	 */
	HOLD_INTERRUPTS();

	for (;;)
	{
		uint32		wait_event = 0; /* initialized to avoid compiler warning */
		bool		mustwait;

		/*
		 * Try to grab the lock the first time, we're not in the waitqueue
		 * yet/anymore.
		 */
		mustwait = BufferLockAttempt(buf_hdr, mode);

		if (likely(!mustwait))
		{
			break;
		}

		/*
		 * Ok, at this point we couldn't grab the lock on the first try. We
		 * cannot simply queue ourselves to the end of the list and wait to be
		 * woken up because by now the lock could long have been released.
		 * Instead add us to the queue and try to grab the lock again. If we
		 * succeed we need to revert the queuing and be happy, otherwise we
		 * recheck the lock. If we still couldn't grab it, we know that the
		 * other locker will see our queue entries when releasing since they
		 * existed before we checked for the lock.
		 */

		/* add to the queue */
		BufferLockQueueSelf(buf_hdr, mode);

		/* we're now guaranteed to be woken up if necessary */
		mustwait = BufferLockAttempt(buf_hdr, mode);

		/* ok, grabbed the lock the second time round, need to undo queueing */
		if (!mustwait)
		{
			BufferLockDequeueSelf(buf_hdr);
			break;
		}

		switch (mode)
		{
			case BUFFER_LOCK_EXCLUSIVE:
				wait_event = WAIT_EVENT_BUFFER_EXCLUSIVE;
				break;
			case BUFFER_LOCK_SHARE_EXCLUSIVE:
				wait_event = WAIT_EVENT_BUFFER_SHARE_EXCLUSIVE;
				break;
			case BUFFER_LOCK_SHARE:
				wait_event = WAIT_EVENT_BUFFER_SHARED;
				break;
			case BUFFER_LOCK_UNLOCK:
				pg_unreachable();

		}
		pgstat_report_wait_start(wait_event);

		/*
		 * Wait until awakened.
		 *
		 * It is possible that we get awakened for a reason other than being
		 * signaled by BufferLockWakeup().  If so, loop back and wait again.
		 * Once we've gotten the lock, re-increment the sema by the number of
		 * additional signals received.
		 */
		for (;;)
		{
			PGSemaphoreLock(MyProc->sem);
			if (MyProc->lwWaiting == LW_WS_NOT_WAITING)
				break;
			extraWaits++;
		}

		pgstat_report_wait_end();

		/* Retrying, allow BufferLockReleaseSub to release waiters again. */
		pg_atomic_fetch_and_u64(&buf_hdr->state, ~BM_LOCK_WAKE_IN_PROGRESS);
	}

	/* Remember that we now hold this lock */
	entry->data.lockmode = mode;

	/*
	 * Fix the process wait semaphore's count for any absorbed wakeups.
	 */
	while (unlikely(extraWaits-- > 0))
		PGSemaphoreUnlock(MyProc->sem);
}

/*
 * Release a previously acquired buffer content lock.
 */
static void
BufferLockUnlock(Buffer buffer, BufferDesc *buf_hdr)
{
	BufferLockMode mode;
	uint64		oldstate;
	uint64		sub;

	mode = BufferLockDisownInternal(buffer, buf_hdr);

	/*
	 * Release my hold on lock, after that it can immediately be acquired by
	 * others, even if we still have to wakeup other waiters.
	 */
	sub = BufferLockReleaseSub(mode);

	oldstate = pg_atomic_sub_fetch_u64(&buf_hdr->state, sub);

	BufferLockProcessRelease(buf_hdr, mode, oldstate);

	/*
	 * Now okay to allow cancel/die interrupts.
	 */
	RESUME_INTERRUPTS();
}


/*
 * Acquire the content lock for the buffer, but only if we don't have to wait.
 *
 * It is allowed to try to conditionally acquire a lock on a buffer that this
 * backend has already locked, but the lock acquisition will always fail, even
 * if the new lock acquisition does not conflict with an already held lock
 * (e.g. two share locks). This is because we currently do not have space to
 * track multiple lock ownerships of the same buffer within one backend.  That
 * is ok for the current uses of BufferLockConditional().
 */
static bool
BufferLockConditional(Buffer buffer, BufferDesc *buf_hdr, BufferLockMode mode)
{
	PrivateRefCountEntry *entry = GetPrivateRefCountEntry(buffer, true);
	bool		mustwait;

	/*
	 * As described above, if we're trying to lock a buffer this backend
	 * already has locked, return false, independent of the existing and
	 * desired lock level.
	 */
	if (entry->data.lockmode != BUFFER_LOCK_UNLOCK)
		return false;

	/*
	 * Lock out cancel/die interrupts until we exit the code section protected
	 * by the content lock.  This ensures that interrupts will not interfere
	 * with manipulations of data structures in shared memory.
	 */
	HOLD_INTERRUPTS();

	/* Check for the lock */
	mustwait = BufferLockAttempt(buf_hdr, mode);

	if (mustwait)
	{
		/* Failed to get lock, so release interrupt holdoff */
		RESUME_INTERRUPTS();
	}
	else
	{
		entry->data.lockmode = mode;
	}

	return !mustwait;
}

/*
 * Internal function that tries to atomically acquire the content lock in the
 * passed in mode.
 *
 * This function will not block waiting for a lock to become free - that's the
 * caller's job.
 *
 * Similar to LWLockAttemptLock().
 */
static inline bool
BufferLockAttempt(BufferDesc *buf_hdr, BufferLockMode mode)
{
	uint64		old_state;

	/*
	 * Read once outside the loop, later iterations will get the newer value
	 * via compare & exchange.
	 */
	old_state = pg_atomic_read_u64(&buf_hdr->state);

	/* loop until we've determined whether we could acquire the lock or not */
	while (true)
	{
		uint64		desired_state;
		bool		lock_free;

		desired_state = old_state;

		if (mode == BUFFER_LOCK_EXCLUSIVE)
		{
			lock_free = (old_state & BM_LOCK_MASK) == 0;
			if (lock_free)
				desired_state += BM_LOCK_VAL_EXCLUSIVE;
		}
		else if (mode == BUFFER_LOCK_SHARE_EXCLUSIVE)
		{
			lock_free = (old_state & (BM_LOCK_VAL_EXCLUSIVE | BM_LOCK_VAL_SHARE_EXCLUSIVE)) == 0;
			if (lock_free)
				desired_state += BM_LOCK_VAL_SHARE_EXCLUSIVE;
		}
		else
		{
			lock_free = (old_state & BM_LOCK_VAL_EXCLUSIVE) == 0;
			if (lock_free)
				desired_state += BM_LOCK_VAL_SHARED;
		}

		/*
		 * Attempt to swap in the state we are expecting. If we didn't see
		 * lock to be free, that's just the old value. If we saw it as free,
		 * we'll attempt to mark it acquired. The reason that we always swap
		 * in the value is that this doubles as a memory barrier. We could try
		 * to be smarter and only swap in values if we saw the lock as free,
		 * but benchmark haven't shown it as beneficial so far.
		 *
		 * Retry if the value changed since we last looked at it.
		 */
		if (likely(pg_atomic_compare_exchange_u64(&buf_hdr->state,
												  &old_state, desired_state)))
		{
			if (lock_free)
			{
				/* Great! Got the lock. */
				return false;
			}
			else
				return true;	/* somebody else has the lock */
		}
	}

	pg_unreachable();
}

/*
 * Add ourselves to the end of the content lock's wait queue.
 */
static void
BufferLockQueueSelf(BufferDesc *buf_hdr, BufferLockMode mode)
{
	/*
	 * If we don't have a PGPROC structure, there's no way to wait. This
	 * should never occur, since MyProc should only be null during shared
	 * memory initialization.
	 */
	if (MyProc == NULL)
		elog(PANIC, "cannot wait without a PGPROC structure");

	if (MyProc->lwWaiting != LW_WS_NOT_WAITING)
		elog(PANIC, "queueing for lock while waiting on another one");

	LockBufHdr(buf_hdr);

	/* setting the flag is protected by the spinlock */
	pg_atomic_fetch_or_u64(&buf_hdr->state, BM_LOCK_HAS_WAITERS);

	/*
	 * These are currently used both for lwlocks and buffer content locks,
	 * which is acceptable, although not pretty, because a backend can't wait
	 * for both types of locks at the same time.
	 */
	MyProc->lwWaiting = LW_WS_WAITING;
	MyProc->lwWaitMode = mode;

	proclist_push_tail(&buf_hdr->lock_waiters, MyProcNumber, lwWaitLink);

	/* Can release the mutex now */
	UnlockBufHdr(buf_hdr);
}

/*
 * Remove ourselves from the waitlist.
 *
 * This is used if we queued ourselves because we thought we needed to sleep
 * but, after further checking, we discovered that we don't actually need to
 * do so.
 */
static void
BufferLockDequeueSelf(BufferDesc *buf_hdr)
{
	bool		on_waitlist;

	LockBufHdr(buf_hdr);

	on_waitlist = MyProc->lwWaiting == LW_WS_WAITING;
	if (on_waitlist)
		proclist_delete(&buf_hdr->lock_waiters, MyProcNumber, lwWaitLink);

	if (proclist_is_empty(&buf_hdr->lock_waiters) &&
		(pg_atomic_read_u64(&buf_hdr->state) & BM_LOCK_HAS_WAITERS) != 0)
	{
		pg_atomic_fetch_and_u64(&buf_hdr->state, ~BM_LOCK_HAS_WAITERS);
	}

	/* XXX: combine with fetch_and above? */
	UnlockBufHdr(buf_hdr);

	/* clear waiting state again, nice for debugging */
	if (on_waitlist)
		MyProc->lwWaiting = LW_WS_NOT_WAITING;
	else
	{
		int			extraWaits = 0;


		/*
		 * Somebody else dequeued us and has or will wake us up. Deal with the
		 * superfluous absorption of a wakeup.
		 */

		/*
		 * Clear BM_LOCK_WAKE_IN_PROGRESS if somebody woke us before we
		 * removed ourselves - they'll have set it.
		 */
		pg_atomic_fetch_and_u64(&buf_hdr->state, ~BM_LOCK_WAKE_IN_PROGRESS);

		/*
		 * Now wait for the scheduled wakeup, otherwise our ->lwWaiting would
		 * get reset at some inconvenient point later. Most of the time this
		 * will immediately return.
		 */
		for (;;)
		{
			PGSemaphoreLock(MyProc->sem);
			if (MyProc->lwWaiting == LW_WS_NOT_WAITING)
				break;
			extraWaits++;
		}

		/*
		 * Fix the process wait semaphore's count for any absorbed wakeups.
		 */
		while (extraWaits-- > 0)
			PGSemaphoreUnlock(MyProc->sem);
	}
}

/*
 * Stop treating lock as held by current backend.
 *
 * After calling this function it's the callers responsibility to ensure that
 * the lock gets released, even in case of an error. This only is desirable if
 * the lock is going to be released in a different process than the process
 * that acquired it.
 */
static inline void
BufferLockDisown(Buffer buffer, BufferDesc *buf_hdr)
{
	BufferLockDisownInternal(buffer, buf_hdr);
	RESUME_INTERRUPTS();
}

/*
 * Stop treating lock as held by current backend.
 *
 * This is the code that can be shared between actually releasing a lock
 * (BufferLockUnlock()) and just not tracking ownership of the lock anymore
 * without releasing the lock (BufferLockDisown()).
 */
static inline int
BufferLockDisownInternal(Buffer buffer, BufferDesc *buf_hdr)
{
	BufferLockMode mode;
	PrivateRefCountEntry *ref;

	ref = GetPrivateRefCountEntry(buffer, false);
	if (ref == NULL)
		elog(ERROR, "lock %d is not held", buffer);
	mode = ref->data.lockmode;
	ref->data.lockmode = BUFFER_LOCK_UNLOCK;

	return mode;
}

/*
 * Wakeup all the lockers that currently have a chance to acquire the lock.
 *
 * wake_exclusive indicates whether exclusive lock waiters should be woken up.
 */
static void
BufferLockWakeup(BufferDesc *buf_hdr, bool wake_exclusive)
{
	bool		new_wake_in_progress = false;
	bool		wake_share_exclusive = true;
	proclist_head wakeup;
	proclist_mutable_iter iter;

	proclist_init(&wakeup);

	/* lock wait list while collecting backends to wake up */
	LockBufHdr(buf_hdr);

	proclist_foreach_modify(iter, &buf_hdr->lock_waiters, lwWaitLink)
	{
		PGPROC	   *waiter = GetPGProcByNumber(iter.cur);

		/*
		 * Already woke up a conflicting lock, so skip over this wait list
		 * entry.
		 */
		if (!wake_exclusive && waiter->lwWaitMode == BUFFER_LOCK_EXCLUSIVE)
			continue;
		if (!wake_share_exclusive && waiter->lwWaitMode == BUFFER_LOCK_SHARE_EXCLUSIVE)
			continue;

		proclist_delete(&buf_hdr->lock_waiters, iter.cur, lwWaitLink);
		proclist_push_tail(&wakeup, iter.cur, lwWaitLink);

		/*
		 * Prevent additional wakeups until retryer gets to run. Backends that
		 * are just waiting for the lock to become free don't retry
		 * automatically.
		 */
		new_wake_in_progress = true;

		/*
		 * Signal that the process isn't on the wait list anymore. This allows
		 * BufferLockDequeueSelf() to remove itself from the waitlist with a
		 * proclist_delete(), rather than having to check if it has been
		 * removed from the list.
		 */
		Assert(waiter->lwWaiting == LW_WS_WAITING);
		waiter->lwWaiting = LW_WS_PENDING_WAKEUP;

		/*
		 * Don't wakeup further waiters after waking a conflicting waiter.
		 */
		if (waiter->lwWaitMode == BUFFER_LOCK_SHARE)
		{
			/*
			 * Share locks conflict with exclusive locks.
			 */
			wake_exclusive = false;
		}
		else if (waiter->lwWaitMode == BUFFER_LOCK_SHARE_EXCLUSIVE)
		{
			/*
			 * Share-exclusive locks conflict with share-exclusive and
			 * exclusive locks.
			 */
			wake_exclusive = false;
			wake_share_exclusive = false;
		}
		else if (waiter->lwWaitMode == BUFFER_LOCK_EXCLUSIVE)
		{
			/*
			 * Exclusive locks conflict with all other locks, there's no point
			 * in waking up anybody else.
			 */
			break;
		}
	}

	Assert(proclist_is_empty(&wakeup) || pg_atomic_read_u64(&buf_hdr->state) & BM_LOCK_HAS_WAITERS);

	/* unset required flags, and release lock, in one fell swoop */
	{
		uint64		old_state;
		uint64		desired_state;

		old_state = pg_atomic_read_u64(&buf_hdr->state);
		while (true)
		{
			desired_state = old_state;

			/* compute desired flags */

			if (new_wake_in_progress)
				desired_state |= BM_LOCK_WAKE_IN_PROGRESS;
			else
				desired_state &= ~BM_LOCK_WAKE_IN_PROGRESS;

			if (proclist_is_empty(&buf_hdr->lock_waiters))
				desired_state &= ~BM_LOCK_HAS_WAITERS;

			desired_state &= ~BM_LOCKED;	/* release lock */

			if (pg_atomic_compare_exchange_u64(&buf_hdr->state, &old_state,
											   desired_state))
				break;
		}
	}

	/* Awaken any waiters I removed from the queue. */
	proclist_foreach_modify(iter, &wakeup, lwWaitLink)
	{
		PGPROC	   *waiter = GetPGProcByNumber(iter.cur);

		proclist_delete(&wakeup, iter.cur, lwWaitLink);

		/*
		 * Guarantee that lwWaiting being unset only becomes visible once the
		 * unlink from the link has completed. Otherwise the target backend
		 * could be woken up for other reason and enqueue for a new lock - if
		 * that happens before the list unlink happens, the list would end up
		 * being corrupted.
		 *
		 * The barrier pairs with the LockBufHdr() when enqueuing for another
		 * lock.
		 */
		pg_write_barrier();
		waiter->lwWaiting = LW_WS_NOT_WAITING;
		PGSemaphoreUnlock(waiter->sem);
	}
}

/*
 * Compute subtraction from buffer state for a release of a held lock in
 * `mode`.
 *
 * This is separated from BufferLockUnlock() as we want to combine the lock
 * release with other atomic operations when possible, leading to the lock
 * release being done in multiple places, each needing to compute what to
 * subtract from the lock state.
 */
static inline uint64
BufferLockReleaseSub(BufferLockMode mode)
{
	/*
	 * Turns out that a switch() leads gcc to generate sufficiently worse code
	 * for this to show up in profiles...
	 */
	if (mode == BUFFER_LOCK_EXCLUSIVE)
		return BM_LOCK_VAL_EXCLUSIVE;
	else if (mode == BUFFER_LOCK_SHARE_EXCLUSIVE)
		return BM_LOCK_VAL_SHARE_EXCLUSIVE;
	else
	{
		Assert(mode == BUFFER_LOCK_SHARE);
		return BM_LOCK_VAL_SHARED;
	}

	return 0;					/* keep compiler quiet */
}

/*
 * Handle work that needs to be done after releasing a lock that was held in
 * `mode`, where `lockstate` is the result of the atomic operation modifying
 * the state variable.
 *
 * This is separated from BufferLockUnlock() as we want to combine the lock
 * release with other atomic operations when possible, leading to the lock
 * release being done in multiple places.
 */
static void
BufferLockProcessRelease(BufferDesc *buf_hdr, BufferLockMode mode, uint64 lockstate)
{
	bool		check_waiters = false;
	bool		wake_exclusive = false;

	/* nobody else can have that kind of lock */
	Assert(!(lockstate & BM_LOCK_VAL_EXCLUSIVE));

	/*
	 * If we're still waiting for backends to get scheduled, don't wake them
	 * up again. Otherwise check if we need to look through the waitqueue to
	 * wake other backends.
	 */
	if ((lockstate & BM_LOCK_HAS_WAITERS) &&
		!(lockstate & BM_LOCK_WAKE_IN_PROGRESS))
	{
		if ((lockstate & BM_LOCK_MASK) == 0)
		{
			/*
			 * We released a lock and the lock was, in that moment, free. We
			 * therefore can wake waiters for any kind of lock.
			 */
			check_waiters = true;
			wake_exclusive = true;
		}
		else if (mode == BUFFER_LOCK_SHARE_EXCLUSIVE)
		{
			/*
			 * We released the lock, but another backend still holds a lock.
			 * We can't have released an exclusive lock, as there couldn't
			 * have been other lock holders. If we released a share lock, no
			 * waiters need to be woken up, as there must be other share
			 * lockers. However, if we held a share-exclusive lock, another
			 * backend now could acquire a share-exclusive lock.
			 */
			check_waiters = true;
			wake_exclusive = false;
		}
	}

	/*
	 * As waking up waiters requires the spinlock to be acquired, only do so
	 * if necessary.
	 */
	if (check_waiters)
		BufferLockWakeup(buf_hdr, wake_exclusive);
}

/*
 * BufferLockHeldByMeInMode - test whether my process holds the content lock
 * in the specified mode
 *
 * This is meant as debug support only.
 */
static bool
BufferLockHeldByMeInMode(BufferDesc *buf_hdr, BufferLockMode mode)
{
	PrivateRefCountEntry *entry =
		GetPrivateRefCountEntry(BufferDescriptorGetBuffer(buf_hdr), false);

	if (!entry)
		return false;
	else
		return entry->data.lockmode == mode;
}

/*
 * BufferLockHeldByMe - test whether my process holds the content lock in any
 * mode
 *
 * This is meant as debug support only.
 */
static bool
BufferLockHeldByMe(BufferDesc *buf_hdr)
{
	PrivateRefCountEntry *entry =
		GetPrivateRefCountEntry(BufferDescriptorGetBuffer(buf_hdr), false);

	if (!entry)
		return false;
	else
		return entry->data.lockmode != BUFFER_LOCK_UNLOCK;
}

/*
 * Release the content lock for the buffer.
 */
void
UnlockBuffer(Buffer buffer)
{
	BufferDesc *buf_hdr;

	Assert(BufferIsPinned(buffer));
	if (BufferIsLocal(buffer))
		return;					/* local buffers need no lock */

	buf_hdr = GetBufferDescriptor(buffer - 1);
	BufferLockUnlock(buffer, buf_hdr);
}

/*
 * Acquire the content_lock for the buffer.
 */
void
LockBufferInternal(Buffer buffer, BufferLockMode mode)
{
	BufferDesc *buf_hdr;

	/*
	 * We can't wait if we haven't got a PGPROC.  This should only occur
	 * during bootstrap or shared memory initialization.  Put an Assert here
	 * to catch unsafe coding practices.
	 */
	Assert(!(MyProc == NULL && IsUnderPostmaster));

	/* handled in LockBuffer() wrapper */
	Assert(mode != BUFFER_LOCK_UNLOCK);

	Assert(BufferIsPinned(buffer));
	if (BufferIsLocal(buffer))
		return;					/* local buffers need no lock */

	buf_hdr = GetBufferDescriptor(buffer - 1);

	/*
	 * Test the most frequent lock modes first. While a switch (mode) would be
	 * nice, at least gcc generates considerably worse code for it.
	 *
	 * Call BufferLockAcquire() with a constant argument for mode, to generate
	 * more efficient code for the different lock modes.
	 */
	if (mode == BUFFER_LOCK_SHARE)
		BufferLockAcquire(buffer, buf_hdr, BUFFER_LOCK_SHARE);
	else if (mode == BUFFER_LOCK_EXCLUSIVE)
		BufferLockAcquire(buffer, buf_hdr, BUFFER_LOCK_EXCLUSIVE);
	else if (mode == BUFFER_LOCK_SHARE_EXCLUSIVE)
		BufferLockAcquire(buffer, buf_hdr, BUFFER_LOCK_SHARE_EXCLUSIVE);
	else
		elog(ERROR, "unrecognized buffer lock mode: %d", mode);
}

/*
 * Acquire the content_lock for the buffer, but only if we don't have to wait.
 *
 * This assumes the caller wants BUFFER_LOCK_EXCLUSIVE mode.
 */
bool
ConditionalLockBuffer(Buffer buffer)
{
	BufferDesc *buf;

	Assert(BufferIsPinned(buffer));
	if (BufferIsLocal(buffer))
		return true;			/* act as though we got it */

	buf = GetBufferDescriptor(buffer - 1);

	return BufferLockConditional(buffer, buf, BUFFER_LOCK_EXCLUSIVE);
}

/*
 * Verify that this backend is pinning the buffer exactly once.
 *
 * NOTE: Like in BufferIsPinned(), what we check here is that *this* backend
 * holds a pin on the buffer.  We do not care whether some other backend does.
 */
void
CheckBufferIsPinnedOnce(Buffer buffer)
{
	if (BufferIsLocal(buffer))
	{
		if (LocalRefCount[-buffer - 1] != 1)
			elog(ERROR, "incorrect local pin count: %d",
				 LocalRefCount[-buffer - 1]);
	}
	else
	{
		if (GetPrivateRefCount(buffer) != 1)
			elog(ERROR, "incorrect local pin count: %d",
				 GetPrivateRefCount(buffer));
	}
}

/*
 * LockBufferForCleanup - lock a buffer in preparation for deleting items
 *
 * Items may be deleted from a disk page only when the caller (a) holds an
 * exclusive lock on the buffer and (b) has observed that no other backend
 * holds a pin on the buffer.  If there is a pin, then the other backend
 * might have a pointer into the buffer (for example, a heapscan reference
 * to an item --- see README for more details).  It's OK if a pin is added
 * after the cleanup starts, however; the newly-arrived backend will be
 * unable to look at the page until we release the exclusive lock.
 *
 * To implement this protocol, a would-be deleter must pin the buffer and
 * then call LockBufferForCleanup().  LockBufferForCleanup() is similar to
 * LockBuffer(buffer, BUFFER_LOCK_EXCLUSIVE), except that it loops until
 * it has successfully observed pin count = 1.
 */
void
LockBufferForCleanup(Buffer buffer)
{
	BufferDesc *bufHdr;
	TimestampTz waitStart = 0;
	bool		waiting = false;
	bool		logged_recovery_conflict = false;

	Assert(BufferIsPinned(buffer));
	Assert(PinCountWaitBuf == NULL);

	CheckBufferIsPinnedOnce(buffer);

	/*
	 * We do not yet need to be worried about in-progress AIOs holding a pin,
	 * as we, so far, only support doing reads via AIO and this function can
	 * only be called once the buffer is valid (i.e. no read can be in
	 * flight).
	 */

	/* Nobody else to wait for */
	if (BufferIsLocal(buffer))
		return;

	bufHdr = GetBufferDescriptor(buffer - 1);

	for (;;)
	{
		uint64		buf_state;
		uint64		unset_bits = 0;

		/* Try to acquire lock */
		LockBuffer(buffer, BUFFER_LOCK_EXCLUSIVE);
		buf_state = LockBufHdr(bufHdr);

		Assert(BUF_STATE_GET_REFCOUNT(buf_state) > 0);
		if (BUF_STATE_GET_REFCOUNT(buf_state) == 1)
		{
			/* Successfully acquired exclusive lock with pincount 1 */
			UnlockBufHdr(bufHdr);

			/*
			 * Emit the log message if recovery conflict on buffer pin was
			 * resolved but the startup process waited longer than
			 * deadlock_timeout for it.
			 */
			if (logged_recovery_conflict)
				LogRecoveryConflict(RECOVERY_CONFLICT_BUFFERPIN,
									waitStart, GetCurrentTimestamp(),
									NULL, false);

			if (waiting)
			{
				/* reset ps display to remove the suffix if we added one */
				set_ps_display_remove_suffix();
				waiting = false;
			}
			return;
		}
		/* Failed, so mark myself as waiting for pincount 1 */
		if (buf_state & BM_PIN_COUNT_WAITER)
		{
			UnlockBufHdr(bufHdr);
			LockBuffer(buffer, BUFFER_LOCK_UNLOCK);
			elog(ERROR, "multiple backends attempting to wait for pincount 1");
		}
		bufHdr->wait_backend_pgprocno = MyProcNumber;
		PinCountWaitBuf = bufHdr;
		UnlockBufHdrExt(bufHdr, buf_state,
						BM_PIN_COUNT_WAITER, 0,
						0);
		LockBuffer(buffer, BUFFER_LOCK_UNLOCK);

		/* Wait to be signaled by UnpinBuffer() */
		if (InHotStandby)
		{
			if (!waiting)
			{
				/* adjust the process title to indicate that it's waiting */
				set_ps_display_suffix("waiting");
				waiting = true;
			}

			/*
			 * Emit the log message if the startup process is waiting longer
			 * than deadlock_timeout for recovery conflict on buffer pin.
			 *
			 * Skip this if first time through because the startup process has
			 * not started waiting yet in this case. So, the wait start
			 * timestamp is set after this logic.
			 */
			if (waitStart != 0 && !logged_recovery_conflict)
			{
				TimestampTz now = GetCurrentTimestamp();

				if (TimestampDifferenceExceeds(waitStart, now,
											   DeadlockTimeout))
				{
					LogRecoveryConflict(RECOVERY_CONFLICT_BUFFERPIN,
										waitStart, now, NULL, true);
					logged_recovery_conflict = true;
				}
			}

			/*
			 * Set the wait start timestamp if logging is enabled and first
			 * time through.
			 */
			if (log_recovery_conflict_waits && waitStart == 0)
				waitStart = GetCurrentTimestamp();

			/* Publish the bufid that Startup process waits on */
			SetStartupBufferPinWaitBufId(buffer - 1);
			/* Set alarm and then wait to be signaled by UnpinBuffer() */
			ResolveRecoveryConflictWithBufferPin();
			/* Reset the published bufid */
			SetStartupBufferPinWaitBufId(-1);
		}
		else
			ProcWaitForSignal(WAIT_EVENT_BUFFER_CLEANUP);

		/*
		 * Remove flag marking us as waiter. Normally this will not be set
		 * anymore, but ProcWaitForSignal() can return for other signals as
		 * well.  We take care to only reset the flag if we're the waiter, as
		 * theoretically another backend could have started waiting. That's
		 * impossible with the current usages due to table level locking, but
		 * better be safe.
		 */
		buf_state = LockBufHdr(bufHdr);
		if ((buf_state & BM_PIN_COUNT_WAITER) != 0 &&
			bufHdr->wait_backend_pgprocno == MyProcNumber)
			unset_bits |= BM_PIN_COUNT_WAITER;

		UnlockBufHdrExt(bufHdr, buf_state,
						0, unset_bits,
						0);

		PinCountWaitBuf = NULL;
		/* Loop back and try again */
	}
}

/*
 * Check called from ProcessRecoveryConflictInterrupts() when Startup process
 * requests cancellation of all pin holders that are blocking it.
 */
bool
HoldingBufferPinThatDelaysRecovery(void)
{
	int			bufid = GetStartupBufferPinWaitBufId();

	/*
	 * If we get woken slowly then it's possible that the Startup process was
	 * already woken by other backends before we got here. Also possible that
	 * we get here by multiple interrupts or interrupts at inappropriate
	 * times, so make sure we do nothing if the bufid is not set.
	 */
	if (bufid < 0)
		return false;

	if (GetPrivateRefCount(bufid + 1) > 0)
		return true;

	return false;
}

/*
 * ConditionalLockBufferForCleanup - as above, but don't wait to get the lock
 *
 * We won't loop, but just check once to see if the pin count is OK.  If
 * not, return false with no lock held.
 */
bool
ConditionalLockBufferForCleanup(Buffer buffer)
{
	BufferDesc *bufHdr;
	uint64		buf_state,
				refcount;

	Assert(BufferIsValid(buffer));

	/* see AIO related comment in LockBufferForCleanup() */

	if (BufferIsLocal(buffer))
	{
		refcount = LocalRefCount[-buffer - 1];
		/* There should be exactly one pin */
		Assert(refcount > 0);
		if (refcount != 1)
			return false;
		/* Nobody else to wait for */
		return true;
	}

	/* There should be exactly one local pin */
	refcount = GetPrivateRefCount(buffer);
	Assert(refcount);
	if (refcount != 1)
		return false;

	/* Try to acquire lock */
	if (!ConditionalLockBuffer(buffer))
		return false;

	bufHdr = GetBufferDescriptor(buffer - 1);
	buf_state = LockBufHdr(bufHdr);
	refcount = BUF_STATE_GET_REFCOUNT(buf_state);

	Assert(refcount > 0);
	if (refcount == 1)
	{
		/* Successfully acquired exclusive lock with pincount 1 */
		UnlockBufHdr(bufHdr);
		return true;
	}

	/* Failed, so release the lock */
	UnlockBufHdr(bufHdr);
	LockBuffer(buffer, BUFFER_LOCK_UNLOCK);
	return false;
}

/*
 * IsBufferCleanupOK - as above, but we already have the lock
 *
 * Check whether it's OK to perform cleanup on a buffer we've already
 * locked.  If we observe that the pin count is 1, our exclusive lock
 * happens to be a cleanup lock, and we can proceed with anything that
 * would have been allowable had we sought a cleanup lock originally.
 */
bool
IsBufferCleanupOK(Buffer buffer)
{
	BufferDesc *bufHdr;
	uint64		buf_state;

	Assert(BufferIsValid(buffer));

	/* see AIO related comment in LockBufferForCleanup() */

	if (BufferIsLocal(buffer))
	{
		/* There should be exactly one pin */
		if (LocalRefCount[-buffer - 1] != 1)
			return false;
		/* Nobody else to wait for */
		return true;
	}

	/* There should be exactly one local pin */
	if (GetPrivateRefCount(buffer) != 1)
		return false;

	bufHdr = GetBufferDescriptor(buffer - 1);

	/* caller must hold exclusive lock on buffer */
	Assert(BufferIsLockedByMeInMode(buffer, BUFFER_LOCK_EXCLUSIVE));

	buf_state = LockBufHdr(bufHdr);

	Assert(BUF_STATE_GET_REFCOUNT(buf_state) > 0);
	if (BUF_STATE_GET_REFCOUNT(buf_state) == 1)
	{
		/* pincount is OK. */
		UnlockBufHdr(bufHdr);
		return true;
	}

	UnlockBufHdr(bufHdr);
	return false;
}

/*
 * Helper for BufferBeginSetHintBits() and BufferSetHintBits16().
 *
 * This checks if the current lock mode already suffices to allow hint bits
 * being set and, if not, whether the current lock can be upgraded.
 *
 * Updates *lockstate when returning true.
 */
static inline bool
SharedBufferBeginSetHintBits(Buffer buffer, BufferDesc *buf_hdr, uint64 *lockstate)
{
	uint64		old_state;
	PrivateRefCountEntry *ref;
	BufferLockMode mode;

	ref = GetPrivateRefCountEntry(buffer, true);

	if (ref == NULL)
		elog(ERROR, "buffer is not pinned");

	mode = ref->data.lockmode;
	if (mode == BUFFER_LOCK_UNLOCK)
		elog(ERROR, "buffer is not locked");

	/* we're done if we are already holding a sufficient lock level */
	if (mode == BUFFER_LOCK_EXCLUSIVE || mode == BUFFER_LOCK_SHARE_EXCLUSIVE)
	{
		*lockstate = pg_atomic_read_u64(&buf_hdr->state);
		return true;
	}

	/*
	 * We are only holding a share lock right now, try to upgrade it to
	 * SHARE_EXCLUSIVE.
	 */
	Assert(mode == BUFFER_LOCK_SHARE);

	old_state = pg_atomic_read_u64(&buf_hdr->state);
	while (true)
	{
		uint64		desired_state;

		desired_state = old_state;

		/*
		 * Can't upgrade if somebody else holds the lock in exclusive or
		 * share-exclusive mode.
		 */
		if (unlikely((old_state & (BM_LOCK_VAL_EXCLUSIVE | BM_LOCK_VAL_SHARE_EXCLUSIVE)) != 0))
		{
			return false;
		}

		/* currently held lock state */
		desired_state -= BM_LOCK_VAL_SHARED;

		/* new lock level */
		desired_state += BM_LOCK_VAL_SHARE_EXCLUSIVE;

		if (likely(pg_atomic_compare_exchange_u64(&buf_hdr->state,
												  &old_state, desired_state)))
		{
			ref->data.lockmode = BUFFER_LOCK_SHARE_EXCLUSIVE;
			*lockstate = desired_state;

			return true;
		}
	}
}

/*
 * Try to acquire the right to set hint bits on the buffer.
 *
 * To be allowed to set hint bits, this backend needs to hold either a
 * share-exclusive or an exclusive lock. In case this backend only holds a
 * share lock, this function will try to upgrade the lock to
 * share-exclusive. The caller is only allowed to set hint bits if true is
 * returned.
 *
 * Once BufferBeginSetHintBits() has returned true, hint bits may be set
 * without further calls to BufferBeginSetHintBits(), until the buffer is
 * unlocked.
 *
 *
 * Requiring a share-exclusive lock to set hint bits prevents setting hint
 * bits on buffers that are currently being written out, which could corrupt
 * the checksum on the page. Flushing buffers also requires a share-exclusive
 * lock.
 *
 * Due to a lock >= share-exclusive being required to set hint bits, only one
 * backend can set hint bits at a time. Allowing multiple backends to set hint
 * bits would require more complicated locking: For setting hint bits we'd
 * need to store the count of backends currently setting hint bits, for I/O we
 * would need another lock-level conflicting with the hint-setting
 * lock-level. Given that the share-exclusive lock for setting hint bits is
 * only held for a short time, that backends often would just set the same
 * hint bits and that the cost of occasionally not setting hint bits in hotly
 * accessed pages is fairly low, this seems like an acceptable tradeoff.
 */
bool
BufferBeginSetHintBits(Buffer buffer)
{
	BufferDesc *buf_hdr;
	uint64		lockstate;

	if (BufferIsLocal(buffer))
	{
		/*
		 * NB: Will need to check if there is a write in progress, once it is
		 * possible for writes to be done asynchronously.
		 */
		return true;
	}

	buf_hdr = GetBufferDescriptor(buffer - 1);

	return SharedBufferBeginSetHintBits(buffer, buf_hdr, &lockstate);
}

/*
 * End a phase of setting hint bits on this buffer, started with
 * BufferBeginSetHintBits().
 *
 * This would strictly speaking not be required (i.e. the caller could do
 * MarkBufferDirtyHint() if so desired), but allows us to perform some sanity
 * checks.
 */
void
BufferFinishSetHintBits(Buffer buffer, bool mark_dirty, bool buffer_std)
{
	if (!BufferIsLocal(buffer))
		Assert(BufferIsLockedByMeInMode(buffer, BUFFER_LOCK_SHARE_EXCLUSIVE) ||
			   BufferIsLockedByMeInMode(buffer, BUFFER_LOCK_EXCLUSIVE));

	if (mark_dirty)
		MarkBufferDirtyHint(buffer, buffer_std);
}

/*
 * Try to set hint bits on a single 16bit value in a buffer.
 *
 * If hint bits are allowed to be set, set *ptr = val, try to mark the buffer
 * dirty and return true. Otherwise false is returned.
 *
 * *ptr needs to be a pointer to memory within the buffer.
 *
 * This is a bit faster than BufferBeginSetHintBits() /
 * BufferFinishSetHintBits() when setting hints once in a buffer, but slower
 * than the former when setting hint bits multiple times in the same buffer.
 */
bool
BufferSetHintBits16(uint16 *ptr, uint16 val, Buffer buffer)
{
	BufferDesc *buf_hdr;
	uint64		lockstate;
#ifdef USE_ASSERT_CHECKING
	char	   *page;

	/* verify that the address is on the page */
	page = BufferGetPage(buffer);
	Assert((char *) ptr >= page && (char *) ptr < (page + BLCKSZ));
#endif

	if (BufferIsLocal(buffer))
	{
		*ptr = val;

		MarkLocalBufferDirty(buffer);

		return true;
	}

	buf_hdr = GetBufferDescriptor(buffer - 1);

	if (SharedBufferBeginSetHintBits(buffer, buf_hdr, &lockstate))
	{
		*ptr = val;

		MarkSharedBufferDirtyHint(buffer, buf_hdr, lockstate, true);

		return true;
	}

	return false;
}


/*
 *	Functions for buffer I/O handling
 *
 *	Also note that these are used only for shared buffers, not local ones.
 */

/*
 * WaitIO -- Block until the IO_IN_PROGRESS flag on 'buf' is cleared.
 */
static void
WaitIO(BufferDesc *buf)
{
	ConditionVariable *cv = BufferDescriptorGetIOCV(buf);

	/*
	 * Should never end up here with unsubmitted IO, as no AIO unaware code
	 * may be used while in batch mode and AIO aware code needs to have
	 * submitted all staged IO to avoid deadlocks & slowness.
	 */
	Assert(!pgaio_have_staged());

	ConditionVariablePrepareToSleep(cv);
	for (;;)
	{
		uint64		buf_state;
		PgAioWaitRef iow;

		/*
		 * It may not be necessary to acquire the spinlock to check the flag
		 * here, but since this test is essential for correctness, we'd better
		 * play it safe.
		 */
		buf_state = LockBufHdr(buf);

		/*
		 * Copy the wait reference while holding the spinlock. This protects
		 * against a concurrent TerminateBufferIO() in another backend from
		 * clearing the wref while it's being read.
		 */
		iow = buf->io_wref;
		UnlockBufHdr(buf);

		/* no IO in progress, we don't need to wait */
		if (!(buf_state & BM_IO_IN_PROGRESS))
			break;

		/*
		 * The buffer has asynchronous IO in progress, wait for it to
		 * complete.
		 */
		if (pgaio_wref_valid(&iow))
		{
			pgaio_wref_wait(&iow);

			/*
			 * The AIO subsystem internally uses condition variables and thus
			 * might remove this backend from the BufferDesc's CV. While that
			 * wouldn't cause a correctness issue (the first CV sleep just
			 * immediately returns if not already registered), it seems worth
			 * avoiding unnecessary loop iterations, given that we take care
			 * to do so at the start of the function.
			 */
			ConditionVariablePrepareToSleep(cv);
			continue;
		}

		/* wait on BufferDesc->cv, e.g. for concurrent synchronous IO */
		ConditionVariableSleep(cv, WAIT_EVENT_BUFFER_IO);
	}
	ConditionVariableCancelSleep();
}

/*
 * StartSharedBufferIO: begin I/O on this buffer
 *	(Assumptions)
 *	The buffer is Pinned
 *
 * In several scenarios the buffer may already be undergoing I/O in this or
 * another backend. How to best handle that depends on the caller's
 * situation. It might be appropriate to wait synchronously (e.g., because the
 * buffer is about to be invalidated); wait asynchronously, using the buffer's
 * IO wait reference (e.g., because the caller is doing readahead and doesn't
 * need the buffer to be ready immediately); or to not wait at all (e.g.,
 * because the caller is trying to combine IO for this buffer with another
 * buffer).
 *
 * How and whether to wait is controlled by the wait and io_wref
 * parameters. In detail:
 *
 * - If the caller passes a non-NULL io_wref and the buffer has an I/O wait
 *   reference, the *io_wref is set to the buffer's io_wref and
 *   BUFFER_IO_IN_PROGRESS is returned. This is done regardless of the wait
 *   parameter.
 *
 * - If the caller passes a NULL io_wref (i.e. the caller does not want to
 *   asynchronously wait for the completion of the IO), wait = false and the
 *   buffer is undergoing IO, BUFFER_IO_IN_PROGRESS is returned.
 *
 * - If wait = true and either the buffer does not have a wait reference,
 *   or the caller passes io_wref = NULL, WaitIO() is used to wait for the IO
 *   to complete.  To avoid the potential of deadlocks and unnecessary delays,
 *   all staged I/O is submitted before waiting.
 *
 * Input operations are only attempted on buffers that are not BM_VALID, and
 * output operations only on buffers that are BM_VALID and BM_DIRTY, so we can
 * always tell if the work is already done.  If no I/O is necessary,
 * BUFFER_IO_ALREADY_DONE is returned.
 *
 * If we successfully marked the buffer as BM_IO_IN_PROGRESS,
 * BUFFER_IO_READY_FOR_IO is returned.
 */
StartBufferIOResult
StartSharedBufferIO(BufferDesc *buf, bool forInput, bool wait, PgAioWaitRef *io_wref)
{
	uint64		buf_state;

	ResourceOwnerEnlarge(CurrentResourceOwner);

	for (;;)
	{
		buf_state = LockBufHdr(buf);

		if (!(buf_state & BM_IO_IN_PROGRESS))
			break;

		/* Join the existing IO */
		if (io_wref != NULL && pgaio_wref_valid(&buf->io_wref))
		{
			*io_wref = buf->io_wref;
			UnlockBufHdr(buf);

			return BUFFER_IO_IN_PROGRESS;
		}
		else if (!wait)
		{
			UnlockBufHdr(buf);
			return BUFFER_IO_IN_PROGRESS;
		}
		else
		{
			/*
			 * With wait = true, we always have to wait if the caller has
			 * passed io_wref = NULL.
			 *
			 * Even with io_wref != NULL, we have to wait if the buffer's wait
			 * ref is not valid but the IO is in progress, someone else
			 * started IO but hasn't set the wait ref yet. We have no choice
			 * but to wait until the IO completes.
			 */
			UnlockBufHdr(buf);

			/*
			 * If this backend currently has staged IO, submit it before
			 * waiting for in-progress IO, to avoid potential deadlocks and
			 * unnecessary delays.
			 */
			pgaio_submit_staged();

			WaitIO(buf);
		}
	}

	/* Once we get here, there is definitely no I/O active on this buffer */

	/* Check if someone else already did the I/O */
	if (forInput ? (buf_state & BM_VALID) : !(buf_state & BM_DIRTY))
	{
		UnlockBufHdr(buf);
		return BUFFER_IO_ALREADY_DONE;
	}

	/*
	 * No IO in progress and not already done; we will start IO. It's possible
	 * that the IO was in progress but we're not done, because the IO errored
	 * out. We'll do the IO ourselves.
	 */
	UnlockBufHdrExt(buf, buf_state,
					BM_IO_IN_PROGRESS, 0,
					0);

	ResourceOwnerRememberBufferIO(CurrentResourceOwner,
								  BufferDescriptorGetBuffer(buf));

	return BUFFER_IO_READY_FOR_IO;
}

/*
 * Wrapper around StartSharedBufferIO / StartLocalBufferIO. Only to be used
 * when the caller doesn't otherwise need to care about local vs shared. See
 * StartSharedBufferIO() for details.
 */
StartBufferIOResult
StartBufferIO(Buffer buffer, bool forInput, bool wait, PgAioWaitRef *io_wref)
{
	BufferDesc *buf_hdr;

	if (BufferIsLocal(buffer))
	{
		buf_hdr = GetLocalBufferDescriptor(-buffer - 1);

		return StartLocalBufferIO(buf_hdr, forInput, wait, io_wref);
	}
	else
	{
		buf_hdr = GetBufferDescriptor(buffer - 1);

		return StartSharedBufferIO(buf_hdr, forInput, wait, io_wref);
	}
}

/*
 * TerminateBufferIO: release a buffer we were doing I/O on
 *	(Assumptions)
 *	My process is executing IO for the buffer
 *	BM_IO_IN_PROGRESS bit is set for the buffer
 *	The buffer is Pinned
 *
 * If clear_dirty is true, we clear the buffer's BM_DIRTY flag.  This is
 * appropriate when terminating a successful write.
 *
 * set_flag_bits gets ORed into the buffer's flags.  It must include
 * BM_IO_ERROR in a failure case.  For successful completion it could
 * be 0, or BM_VALID if we just finished reading in the page.
 *
 * If forget_owner is true, we release the buffer I/O from the current
 * resource owner. (forget_owner=false is used when the resource owner itself
 * is being released)
 */
void
TerminateBufferIO(BufferDesc *buf, bool clear_dirty, uint64 set_flag_bits,
				  bool forget_owner, bool release_aio)
{
	uint64		buf_state;
	uint64		unset_flag_bits = 0;
	int			refcount_change = 0;

	buf_state = LockBufHdr(buf);

	Assert(buf_state & BM_IO_IN_PROGRESS);
	unset_flag_bits |= BM_IO_IN_PROGRESS;

	/* Clear earlier errors, if this IO failed, it'll be marked again */
	unset_flag_bits |= BM_IO_ERROR;

	if (clear_dirty)
		unset_flag_bits |= BM_DIRTY | BM_CHECKPOINT_NEEDED;

	if (release_aio)
	{
		/* release ownership by the AIO subsystem */
		Assert(BUF_STATE_GET_REFCOUNT(buf_state) > 0);
		refcount_change = -1;
		pgaio_wref_clear(&buf->io_wref);
	}

	buf_state = UnlockBufHdrExt(buf, buf_state,
								set_flag_bits, unset_flag_bits,
								refcount_change);

	if (forget_owner)
		ResourceOwnerForgetBufferIO(CurrentResourceOwner,
									BufferDescriptorGetBuffer(buf));

	ConditionVariableBroadcast(BufferDescriptorGetIOCV(buf));

	/*
	 * Support LockBufferForCleanup()
	 *
	 * We may have just released the last pin other than the waiter's. In most
	 * cases, this backend holds another pin on the buffer. But, if, for
	 * example, this backend is completing an IO issued by another backend, it
	 * may be time to wake the waiter.
	 */
	if (release_aio && (buf_state & BM_PIN_COUNT_WAITER))
		WakePinCountWaiter(buf);
}

/*
 * AbortBufferIO: Clean up active buffer I/O after an error.
 *
 *	All LWLocks & content locks we might have held have been released, but we
 *	haven't yet released buffer pins, so the buffer is still pinned.
 *
 *	If I/O was in progress, we always set BM_IO_ERROR, even though it's
 *	possible the error condition wasn't related to the I/O.
 *
 *  Note: this does not remove the buffer I/O from the resource owner.
 *  That's correct when we're releasing the whole resource owner, but
 *  beware if you use this in other contexts.
 */
static void
AbortBufferIO(Buffer buffer)
{
	BufferDesc *buf_hdr = GetBufferDescriptor(buffer - 1);
	uint64		buf_state;

	buf_state = LockBufHdr(buf_hdr);
	Assert(buf_state & (BM_IO_IN_PROGRESS | BM_TAG_VALID));

	if (!(buf_state & BM_VALID))
	{
		Assert(!(buf_state & BM_DIRTY));
		UnlockBufHdr(buf_hdr);
	}
	else
	{
		Assert(buf_state & BM_DIRTY);
		UnlockBufHdr(buf_hdr);

		/* Issue notice if this is not the first failure... */
		if (buf_state & BM_IO_ERROR)
		{
			/* Buffer is pinned, so we can read tag without spinlock */
			ereport(WARNING,
					(errcode(ERRCODE_IO_ERROR),
					 errmsg("could not write block %u of %s",
							buf_hdr->tag.blockNum,
							relpathperm(BufTagGetRelFileLocator(&buf_hdr->tag),
										BufTagGetForkNum(&buf_hdr->tag)).str),
					 errdetail("Multiple failures --- write error might be permanent.")));
		}
	}

	TerminateBufferIO(buf_hdr, false, BM_IO_ERROR, false, false);
}

/*
 * Error context callback for errors occurring during shared buffer writes.
 */
static void
shared_buffer_write_error_callback(void *arg)
{
	BufferDesc *bufHdr = (BufferDesc *) arg;

	/* Buffer is pinned, so we can read the tag without locking the spinlock */
	if (bufHdr != NULL)
		errcontext("writing block %u of relation \"%s\"",
				   bufHdr->tag.blockNum,
				   relpathperm(BufTagGetRelFileLocator(&bufHdr->tag),
							   BufTagGetForkNum(&bufHdr->tag)).str);
}

/*
 * Error context callback for errors occurring during local buffer writes.
 */
static void
local_buffer_write_error_callback(void *arg)
{
	BufferDesc *bufHdr = (BufferDesc *) arg;

	if (bufHdr != NULL)
		errcontext("writing block %u of relation \"%s\"",
				   bufHdr->tag.blockNum,
				   relpathbackend(BufTagGetRelFileLocator(&bufHdr->tag),
								  MyProcNumber,
								  BufTagGetForkNum(&bufHdr->tag)).str);
}

/*
 * RelFileLocator qsort/bsearch comparator; see RelFileLocatorEquals.
 */
static int
rlocator_comparator(const void *p1, const void *p2)
{
	RelFileLocator n1 = *(const RelFileLocator *) p1;
	RelFileLocator n2 = *(const RelFileLocator *) p2;

	if (n1.relNumber < n2.relNumber)
		return -1;
	else if (n1.relNumber > n2.relNumber)
		return 1;

	if (n1.dbOid < n2.dbOid)
		return -1;
	else if (n1.dbOid > n2.dbOid)
		return 1;

	if (n1.spcOid < n2.spcOid)
		return -1;
	else if (n1.spcOid > n2.spcOid)
		return 1;
	else
		return 0;
}

/*
 * Lock buffer header - set BM_LOCKED in buffer state.
 */
uint64
LockBufHdr(BufferDesc *desc)
{
	uint64		old_buf_state;

	Assert(!BufferIsLocal(BufferDescriptorGetBuffer(desc)));

	while (true)
	{
		/*
		 * Always try once to acquire the lock directly, without setting up
		 * the spin-delay infrastructure. The work necessary for that shows up
		 * in profiles and is rarely necessary.
		 */
		old_buf_state = pg_atomic_fetch_or_u64(&desc->state, BM_LOCKED);
		if (likely(!(old_buf_state & BM_LOCKED)))
			break;				/* got lock */

		/* and then spin without atomic operations until lock is released */
		{
			SpinDelayStatus delayStatus;

			init_local_spin_delay(&delayStatus);

			while (old_buf_state & BM_LOCKED)
			{
				perform_spin_delay(&delayStatus);
				old_buf_state = pg_atomic_read_u64(&desc->state);
			}
			finish_spin_delay(&delayStatus);
		}

		/*
		 * Retry. The lock might obviously already be re-acquired by the time
		 * we're attempting to get it again.
		 */
	}

	return old_buf_state | BM_LOCKED;
}

/*
 * Wait until the BM_LOCKED flag isn't set anymore and return the buffer's
 * state at that point.
 *
 * Obviously the buffer could be locked by the time the value is returned, so
 * this is primarily useful in CAS style loops.
 */
pg_noinline uint64
WaitBufHdrUnlocked(BufferDesc *buf)
{
	SpinDelayStatus delayStatus;
	uint64		buf_state;

	init_local_spin_delay(&delayStatus);

	buf_state = pg_atomic_read_u64(&buf->state);

	while (buf_state & BM_LOCKED)
	{
		perform_spin_delay(&delayStatus);
		buf_state = pg_atomic_read_u64(&buf->state);
	}

	finish_spin_delay(&delayStatus);

	return buf_state;
}

/*
 * BufferTag comparator.
 */
static inline int
buffertag_comparator(const BufferTag *ba, const BufferTag *bb)
{
	int			ret;
	RelFileLocator rlocatora;
	RelFileLocator rlocatorb;

	rlocatora = BufTagGetRelFileLocator(ba);
	rlocatorb = BufTagGetRelFileLocator(bb);

	ret = rlocator_comparator(&rlocatora, &rlocatorb);

	if (ret != 0)
		return ret;

	if (BufTagGetForkNum(ba) < BufTagGetForkNum(bb))
		return -1;
	if (BufTagGetForkNum(ba) > BufTagGetForkNum(bb))
		return 1;

	if (ba->blockNum < bb->blockNum)
		return -1;
	if (ba->blockNum > bb->blockNum)
		return 1;

	return 0;
}

/*
 * Comparator determining the writeout order in a checkpoint.
 *
 * It is important that tablespaces are compared first, the logic balancing
 * writes between tablespaces relies on it.
 */
static inline int
ckpt_buforder_comparator(const CkptSortItem *a, const CkptSortItem *b)
{
	/* compare tablespace */
	if (a->tsId < b->tsId)
		return -1;
	else if (a->tsId > b->tsId)
		return 1;
	/* compare relation */
	if (a->relNumber < b->relNumber)
		return -1;
	else if (a->relNumber > b->relNumber)
		return 1;
	/* compare fork */
	else if (a->forkNum < b->forkNum)
		return -1;
	else if (a->forkNum > b->forkNum)
		return 1;
	/* compare block number */
	else if (a->blockNum < b->blockNum)
		return -1;
	else if (a->blockNum > b->blockNum)
		return 1;
	/* equal page IDs are unlikely, but not impossible */
	return 0;
}

/*
 * Comparator for a Min-Heap over the per-tablespace checkpoint completion
 * progress.
 */
static int
ts_ckpt_progress_comparator(Datum a, Datum b, void *arg)
{
	CkptTsStatus *sa = (CkptTsStatus *) DatumGetPointer(a);
	CkptTsStatus *sb = (CkptTsStatus *) DatumGetPointer(b);

	/* we want a min-heap, so return 1 for the a < b */
	if (sa->progress < sb->progress)
		return 1;
	else if (sa->progress == sb->progress)
		return 0;
	else
		return -1;
}

/*
 * Initialize a writeback context, discarding potential previous state.
 *
 * *max_pending is a pointer instead of an immediate value, so the coalesce
 * limits can easily changed by the GUC mechanism, and so calling code does
 * not have to check the current configuration. A value of 0 means that no
 * writeback control will be performed.
 */
void
WritebackContextInit(WritebackContext *context, int *max_pending)
{
	Assert(*max_pending <= WRITEBACK_MAX_PENDING_FLUSHES);

	context->max_pending = max_pending;
	context->nr_pending = 0;
}

/*
 * Add buffer to list of pending writeback requests.
 */
void
ScheduleBufferTagForWriteback(WritebackContext *wb_context, IOContext io_context,
							  BufferTag *tag)
{
	PendingWriteback *pending;

	/*
	 * As pg_flush_data() doesn't do anything with fsync disabled, there's no
	 * point in tracking in that case.
	 */
	if (io_direct_flags & IO_DIRECT_DATA ||
		!enableFsync)
		return;

	/*
	 * Add buffer to the pending writeback array, unless writeback control is
	 * disabled.
	 */
	if (*wb_context->max_pending > 0)
	{
		Assert(*wb_context->max_pending <= WRITEBACK_MAX_PENDING_FLUSHES);

		pending = &wb_context->pending_writebacks[wb_context->nr_pending++];

		pending->tag = *tag;
	}

	/*
	 * Perform pending flushes if the writeback limit is exceeded. This
	 * includes the case where previously an item has been added, but control
	 * is now disabled.
	 */
	if (wb_context->nr_pending >= *wb_context->max_pending)
		IssuePendingWritebacks(wb_context, io_context);
}

#define ST_SORT sort_pending_writebacks
#define ST_ELEMENT_TYPE PendingWriteback
#define ST_COMPARE(a, b) buffertag_comparator(&a->tag, &b->tag)
#define ST_SCOPE static
#define ST_DEFINE
#include "lib/sort_template.h"

/*
 * Issue all pending writeback requests, previously scheduled with
 * ScheduleBufferTagForWriteback, to the OS.
 *
 * Because this is only used to improve the OSs IO scheduling we try to never
 * error out - it's just a hint.
 */
void
IssuePendingWritebacks(WritebackContext *wb_context, IOContext io_context)
{
	instr_time	io_start;
	int			i;

	if (wb_context->nr_pending == 0)
		return;

	/*
	 * Executing the writes in-order can make them a lot faster, and allows to
	 * merge writeback requests to consecutive blocks into larger writebacks.
	 */
	sort_pending_writebacks(wb_context->pending_writebacks,
							wb_context->nr_pending);

	io_start = pgstat_prepare_io_time(track_io_timing);

	/*
	 * Coalesce neighbouring writes, but nothing else. For that we iterate
	 * through the, now sorted, array of pending flushes, and look forward to
	 * find all neighbouring (or identical) writes.
	 */
	for (i = 0; i < wb_context->nr_pending; i++)
	{
		PendingWriteback *cur;
		PendingWriteback *next;
		SMgrRelation reln;
		int			ahead;
		BufferTag	tag;
		RelFileLocator currlocator;
		Size		nblocks = 1;

		cur = &wb_context->pending_writebacks[i];
		tag = cur->tag;
		currlocator = BufTagGetRelFileLocator(&tag);

		/*
		 * Peek ahead, into following writeback requests, to see if they can
		 * be combined with the current one.
		 */
		for (ahead = 0; i + ahead + 1 < wb_context->nr_pending; ahead++)
		{

			next = &wb_context->pending_writebacks[i + ahead + 1];

			/* different file, stop */
			if (!RelFileLocatorEquals(currlocator,
									  BufTagGetRelFileLocator(&next->tag)) ||
				BufTagGetForkNum(&cur->tag) != BufTagGetForkNum(&next->tag))
				break;

			/* ok, block queued twice, skip */
			if (cur->tag.blockNum == next->tag.blockNum)
				continue;

			/* only merge consecutive writes */
			if (cur->tag.blockNum + 1 != next->tag.blockNum)
				break;

			nblocks++;
			cur = next;
		}

		i += ahead;

		/* and finally tell the kernel to write the data to storage */
		reln = smgropen(currlocator, INVALID_PROC_NUMBER);
		smgrwriteback(reln, BufTagGetForkNum(&tag), tag.blockNum, nblocks);
	}

	/*
	 * Assume that writeback requests are only issued for buffers containing
	 * blocks of permanent relations.
	 */
	pgstat_count_io_op_time(IOOBJECT_RELATION, io_context,
							IOOP_WRITEBACK, io_start, wb_context->nr_pending, 0);

	wb_context->nr_pending = 0;
}

/* ResourceOwner callbacks */

static void
ResOwnerReleaseBufferIO(Datum res)
{
	Buffer		buffer = DatumGetInt32(res);

	AbortBufferIO(buffer);
}

static char *
ResOwnerPrintBufferIO(Datum res)
{
	Buffer		buffer = DatumGetInt32(res);

	return psprintf("lost track of buffer IO on buffer %d", buffer);
}

/*
 * Release buffer as part of resource owner cleanup. This will only be called
 * if the buffer is pinned. If this backend held the content lock at the time
 * of the error we also need to release that (note that it is not possible to
 * hold a content lock without a pin).
 */
static void
ResOwnerReleaseBuffer(Datum res)
{
	Buffer		buffer = DatumGetInt32(res);

	/* Like ReleaseBuffer, but don't call ResourceOwnerForgetBuffer */
	if (!BufferIsValid(buffer))
		elog(ERROR, "bad buffer ID: %d", buffer);

	if (BufferIsLocal(buffer))
		UnpinLocalBufferNoOwner(buffer);
	else
	{
		PrivateRefCountEntry *ref;

		ref = GetPrivateRefCountEntry(buffer, false);

		/* not having a private refcount would imply resowner corruption */
		Assert(ref != NULL);

		/*
		 * If the buffer was locked at the time of the resowner release,
		 * release the lock now. This should only happen after errors.
		 */
		if (ref->data.lockmode != BUFFER_LOCK_UNLOCK)
		{
			BufferDesc *buf = GetBufferDescriptor(buffer - 1);

			HOLD_INTERRUPTS();	/* match the upcoming RESUME_INTERRUPTS */
			BufferLockUnlock(buffer, buf);
		}

		UnpinBufferNoOwner(GetBufferDescriptor(buffer - 1));
	}
}

static char *
ResOwnerPrintBuffer(Datum res)
{
	return DebugPrintBufferRefcount(DatumGetInt32(res));
}

/*
 * Helper function to evict unpinned buffer whose buffer header lock is
 * already acquired.
 */
static bool
EvictUnpinnedBufferInternal(BufferDesc *desc, bool *buffer_flushed)
{
	uint64		buf_state;
	bool		result;

	*buffer_flushed = false;

	buf_state = pg_atomic_read_u64(&(desc->state));
	Assert(buf_state & BM_LOCKED);

	if ((buf_state & BM_VALID) == 0)
	{
		UnlockBufHdr(desc);
		return false;
	}

	/* Check that it's not pinned already. */
	if (BUF_STATE_GET_REFCOUNT(buf_state) > 0)
	{
		UnlockBufHdr(desc);
		return false;
	}

	PinBuffer_Locked(desc);		/* releases spinlock */

	/* If it was dirty, try to clean it once. */
	if (buf_state & BM_DIRTY)
	{
		FlushUnlockedBuffer(desc, NULL, IOOBJECT_RELATION, IOCONTEXT_NORMAL);
		*buffer_flushed = true;
	}

	/* This will return false if it becomes dirty or someone else pins it. */
	result = InvalidateVictimBuffer(desc);

	UnpinBuffer(desc);

	return result;
}

/*
 * Try to evict the current block in a shared buffer.
 *
 * This function is intended for testing/development use only!
 *
 * To succeed, the buffer must not be pinned on entry, so if the caller had a
 * particular block in mind, it might already have been replaced by some other
 * block by the time this function runs.  It's also unpinned on return, so the
 * buffer might be occupied again by the time control is returned, potentially
 * even by the same block.  This inherent raciness without other interlocking
 * makes the function unsuitable for non-testing usage.
 *
 * *buffer_flushed is set to true if the buffer was dirty and has been
 * flushed, false otherwise.  However, *buffer_flushed=true does not
 * necessarily mean that we flushed the buffer, it could have been flushed by
 * someone else.
 *
 * Returns true if the buffer was valid and it has now been made invalid.
 * Returns false if it wasn't valid, if it couldn't be evicted due to a pin,
 * or if the buffer becomes dirty again while we're trying to write it out.
 */
bool
EvictUnpinnedBuffer(Buffer buf, bool *buffer_flushed)
{
	BufferDesc *desc;

	Assert(BufferIsValid(buf) && !BufferIsLocal(buf));

	/* Make sure we can pin the buffer. */
	ResourceOwnerEnlarge(CurrentResourceOwner);
	ReservePrivateRefCountEntry();

	desc = GetBufferDescriptor(buf - 1);
	LockBufHdr(desc);

	return EvictUnpinnedBufferInternal(desc, buffer_flushed);
}

/*
 * Try to evict all the shared buffers.
 *
 * This function is intended for testing/development use only! See
 * EvictUnpinnedBuffer().
 *
 * The buffers_* parameters are mandatory and indicate the total count of
 * buffers that:
 * - buffers_evicted - were evicted
 * - buffers_flushed - were flushed
 * - buffers_skipped - could not be evicted
 */
void
EvictAllUnpinnedBuffers(int32 *buffers_evicted, int32 *buffers_flushed,
						int32 *buffers_skipped)
{
	*buffers_evicted = 0;
	*buffers_skipped = 0;
	*buffers_flushed = 0;

	for (int buf = 1; buf <= NBuffers; buf++)
	{
		BufferDesc *desc = GetBufferDescriptor(buf - 1);
		uint64		buf_state;
		bool		buffer_flushed;

		CHECK_FOR_INTERRUPTS();

		buf_state = pg_atomic_read_u64(&desc->state);
		if (!(buf_state & BM_VALID))
			continue;

		ResourceOwnerEnlarge(CurrentResourceOwner);
		ReservePrivateRefCountEntry();

		LockBufHdr(desc);

		if (EvictUnpinnedBufferInternal(desc, &buffer_flushed))
			(*buffers_evicted)++;
		else
			(*buffers_skipped)++;

		if (buffer_flushed)
			(*buffers_flushed)++;
	}
}

/*
 * Try to evict all the shared buffers containing provided relation's pages.
 *
 * This function is intended for testing/development use only! See
 * EvictUnpinnedBuffer().
 *
 * The caller must hold at least AccessShareLock on the relation to prevent
 * the relation from being dropped.
 *
 * The buffers_* parameters are mandatory and indicate the total count of
 * buffers that:
 * - buffers_evicted - were evicted
 * - buffers_flushed - were flushed
 * - buffers_skipped - could not be evicted
 */
void
EvictRelUnpinnedBuffers(Relation rel, int32 *buffers_evicted,
						int32 *buffers_flushed, int32 *buffers_skipped)
{
	Assert(!RelationUsesLocalBuffers(rel));

	*buffers_skipped = 0;
	*buffers_evicted = 0;
	*buffers_flushed = 0;

	for (int buf = 1; buf <= NBuffers; buf++)
	{
		BufferDesc *desc = GetBufferDescriptor(buf - 1);
		uint64		buf_state = pg_atomic_read_u64(&(desc->state));
		bool		buffer_flushed;

		CHECK_FOR_INTERRUPTS();

		/* An unlocked precheck should be safe and saves some cycles. */
		if ((buf_state & BM_VALID) == 0 ||
			!BufTagMatchesRelFileLocator(&desc->tag, &rel->rd_locator))
			continue;

		/* Make sure we can pin the buffer. */
		ResourceOwnerEnlarge(CurrentResourceOwner);
		ReservePrivateRefCountEntry();

		buf_state = LockBufHdr(desc);

		/* recheck, could have changed without the lock */
		if ((buf_state & BM_VALID) == 0 ||
			!BufTagMatchesRelFileLocator(&desc->tag, &rel->rd_locator))
		{
			UnlockBufHdr(desc);
			continue;
		}

		if (EvictUnpinnedBufferInternal(desc, &buffer_flushed))
			(*buffers_evicted)++;
		else
			(*buffers_skipped)++;

		if (buffer_flushed)
			(*buffers_flushed)++;
	}
}

/*
 * Helper function to mark unpinned buffer dirty whose buffer header lock is
 * already acquired.
 */
static bool
MarkDirtyUnpinnedBufferInternal(Buffer buf, BufferDesc *desc,
								bool *buffer_already_dirty)
{
	uint64		buf_state;
	bool		result = false;

	*buffer_already_dirty = false;

	buf_state = pg_atomic_read_u64(&(desc->state));
	Assert(buf_state & BM_LOCKED);

	if ((buf_state & BM_VALID) == 0)
	{
		UnlockBufHdr(desc);
		return false;
	}

	/* Check that it's not pinned already. */
	if (BUF_STATE_GET_REFCOUNT(buf_state) > 0)
	{
		UnlockBufHdr(desc);
		return false;
	}

	/* Pin the buffer and then release the buffer spinlock */
	PinBuffer_Locked(desc);

	/* If it was not already dirty, mark it as dirty. */
	if (!(buf_state & BM_DIRTY))
	{
		BufferLockAcquire(buf, desc, BUFFER_LOCK_EXCLUSIVE);
		MarkBufferDirty(buf);
		result = true;
		BufferLockUnlock(buf, desc);
	}
	else
		*buffer_already_dirty = true;

	UnpinBuffer(desc);

	return result;
}

/*
 * Try to mark the provided shared buffer as dirty.
 *
 * This function is intended for testing/development use only!
 *
 * Same as EvictUnpinnedBuffer() but with MarkBufferDirty() call inside.
 *
 * The buffer_already_dirty parameter is mandatory and indicate if the buffer
 * could not be dirtied because it is already dirty.
 *
 * Returns true if the buffer has successfully been marked as dirty.
 */
bool
MarkDirtyUnpinnedBuffer(Buffer buf, bool *buffer_already_dirty)
{
	BufferDesc *desc;
	bool		buffer_dirtied = false;

	Assert(!BufferIsLocal(buf));

	/* Make sure we can pin the buffer. */
	ResourceOwnerEnlarge(CurrentResourceOwner);
	ReservePrivateRefCountEntry();

	desc = GetBufferDescriptor(buf - 1);
	LockBufHdr(desc);

	buffer_dirtied = MarkDirtyUnpinnedBufferInternal(buf, desc, buffer_already_dirty);
	/* Both can not be true at the same time */
	Assert(!(buffer_dirtied && *buffer_already_dirty));

	return buffer_dirtied;
}

/*
 * Try to mark all the shared buffers containing provided relation's pages as
 * dirty.
 *
 * This function is intended for testing/development use only! See
 * MarkDirtyUnpinnedBuffer().
 *
 * The buffers_* parameters are mandatory and indicate the total count of
 * buffers that:
 * - buffers_dirtied - were dirtied
 * - buffers_already_dirty - were already dirty
 * - buffers_skipped - could not be dirtied because of a reason different
 * than a buffer being already dirty.
 */
void
MarkDirtyRelUnpinnedBuffers(Relation rel,
							int32 *buffers_dirtied,
							int32 *buffers_already_dirty,
							int32 *buffers_skipped)
{
	Assert(!RelationUsesLocalBuffers(rel));

	*buffers_dirtied = 0;
	*buffers_already_dirty = 0;
	*buffers_skipped = 0;

	for (int buf = 1; buf <= NBuffers; buf++)
	{
		BufferDesc *desc = GetBufferDescriptor(buf - 1);
		uint64		buf_state = pg_atomic_read_u64(&(desc->state));
		bool		buffer_already_dirty;

		CHECK_FOR_INTERRUPTS();

		/* An unlocked precheck should be safe and saves some cycles. */
		if ((buf_state & BM_VALID) == 0 ||
			!BufTagMatchesRelFileLocator(&desc->tag, &rel->rd_locator))
			continue;

		/* Make sure we can pin the buffer. */
		ResourceOwnerEnlarge(CurrentResourceOwner);
		ReservePrivateRefCountEntry();

		buf_state = LockBufHdr(desc);

		/* recheck, could have changed without the lock */
		if ((buf_state & BM_VALID) == 0 ||
			!BufTagMatchesRelFileLocator(&desc->tag, &rel->rd_locator))
		{
			UnlockBufHdr(desc);
			continue;
		}

		if (MarkDirtyUnpinnedBufferInternal(buf, desc, &buffer_already_dirty))
			(*buffers_dirtied)++;
		else if (buffer_already_dirty)
			(*buffers_already_dirty)++;
		else
			(*buffers_skipped)++;
	}
}

/*
 * Try to mark all the shared buffers as dirty.
 *
 * This function is intended for testing/development use only! See
 * MarkDirtyUnpinnedBuffer().
 *
 * See MarkDirtyRelUnpinnedBuffers() above for details about the buffers_*
 * parameters.
 */
void
MarkDirtyAllUnpinnedBuffers(int32 *buffers_dirtied,
							int32 *buffers_already_dirty,
							int32 *buffers_skipped)
{
	*buffers_dirtied = 0;
	*buffers_already_dirty = 0;
	*buffers_skipped = 0;

	for (int buf = 1; buf <= NBuffers; buf++)
	{
		BufferDesc *desc = GetBufferDescriptor(buf - 1);
		uint64		buf_state;
		bool		buffer_already_dirty;

		CHECK_FOR_INTERRUPTS();

		buf_state = pg_atomic_read_u64(&desc->state);
		if (!(buf_state & BM_VALID))
			continue;

		ResourceOwnerEnlarge(CurrentResourceOwner);
		ReservePrivateRefCountEntry();

		LockBufHdr(desc);

		if (MarkDirtyUnpinnedBufferInternal(buf, desc, &buffer_already_dirty))
			(*buffers_dirtied)++;
		else if (buffer_already_dirty)
			(*buffers_already_dirty)++;
		else
			(*buffers_skipped)++;
	}
}

/*
 * Generic implementation of the AIO handle staging callback for readv/writev
 * on local/shared buffers.
 *
 * Each readv/writev can target multiple buffers. The buffers have already
 * been registered with the IO handle.
 *
 * To make the IO ready for execution ("staging"), we need to ensure that the
 * targeted buffers are in an appropriate state while the IO is ongoing. For
 * that the AIO subsystem needs to have its own buffer pin, otherwise an error
 * in this backend could lead to this backend's buffer pin being released as
 * part of error handling, which in turn could lead to the buffer being
 * replaced while IO is ongoing.
 */
static pg_attribute_always_inline void
buffer_stage_common(PgAioHandle *ioh, bool is_write, bool is_temp)
{
	uint64	   *io_data;
	uint8		handle_data_len;
	PgAioWaitRef io_ref;
	BufferTag	first PG_USED_FOR_ASSERTS_ONLY = {0};

	io_data = pgaio_io_get_handle_data(ioh, &handle_data_len);

	pgaio_io_get_wref(ioh, &io_ref);

	/* iterate over all buffers affected by the vectored readv/writev */
	for (int i = 0; i < handle_data_len; i++)
	{
		Buffer		buffer = (Buffer) io_data[i];
		BufferDesc *buf_hdr = is_temp ?
			GetLocalBufferDescriptor(-buffer - 1)
			: GetBufferDescriptor(buffer - 1);
		uint64		buf_state;

		/*
		 * Check that all the buffers are actually ones that could conceivably
		 * be done in one IO, i.e. are sequential. This is the last
		 * buffer-aware code before IO is actually executed and confusion
		 * about which buffers are targeted by IO can be hard to debug, making
		 * it worth doing extra-paranoid checks.
		 */
		if (i == 0)
			first = buf_hdr->tag;
		else
		{
			Assert(buf_hdr->tag.relNumber == first.relNumber);
			Assert(buf_hdr->tag.blockNum == first.blockNum + i);
		}

		if (is_temp)
			buf_state = pg_atomic_read_u64(&buf_hdr->state);
		else
			buf_state = LockBufHdr(buf_hdr);

		/* verify the buffer is in the expected state */
		Assert(buf_state & BM_TAG_VALID);
		if (is_write)
		{
			Assert(buf_state & BM_VALID);
			Assert(buf_state & BM_DIRTY);
		}
		else
		{
			Assert(!(buf_state & BM_VALID));
			Assert(!(buf_state & BM_DIRTY));
		}

		/* temp buffers don't use BM_IO_IN_PROGRESS */
		if (!is_temp)
			Assert(buf_state & BM_IO_IN_PROGRESS);

		Assert(BUF_STATE_GET_REFCOUNT(buf_state) >= 1);

		/*
		 * Reflect that the buffer is now owned by the AIO subsystem.
		 *
		 * For local buffers: This can't be done just via LocalRefCount, as
		 * one might initially think, as this backend could error out while
		 * AIO is still in progress, releasing all the pins by the backend
		 * itself.
		 *
		 * This pin is released again in TerminateBufferIO().
		 */
		buf_hdr->io_wref = io_ref;

		if (is_temp)
		{
			buf_state += BUF_REFCOUNT_ONE;
			pg_atomic_unlocked_write_u64(&buf_hdr->state, buf_state);
		}
		else
			UnlockBufHdrExt(buf_hdr, buf_state, 0, 0, 1);

		/*
		 * Ensure the content lock that prevents buffer modifications while
		 * the buffer is being written out is not released early due to an
		 * error.
		 */
		if (is_write && !is_temp)
		{
			Assert(BufferLockHeldByMe(buf_hdr));

			/*
			 * Lock is now owned by AIO subsystem.
			 */
			BufferLockDisown(buffer, buf_hdr);
		}

		/*
		 * Stop tracking this buffer via the resowner - the AIO system now
		 * keeps track.
		 */
		if (!is_temp)
			ResourceOwnerForgetBufferIO(CurrentResourceOwner, buffer);
	}
}

/*
 * Decode readv errors as encoded by buffer_readv_encode_error().
 */
static inline void
buffer_readv_decode_error(PgAioResult result,
						  bool *zeroed_any,
						  bool *ignored_any,
						  uint8 *zeroed_or_error_count,
						  uint8 *checkfail_count,
						  uint8 *first_off)
{
	uint32		rem_error = result.error_data;

	/* see static asserts in buffer_readv_encode_error */
#define READV_COUNT_BITS	7
#define READV_COUNT_MASK	((1 << READV_COUNT_BITS) - 1)

	*zeroed_any = rem_error & 1;
	rem_error >>= 1;

	*ignored_any = rem_error & 1;
	rem_error >>= 1;

	*zeroed_or_error_count = rem_error & READV_COUNT_MASK;
	rem_error >>= READV_COUNT_BITS;

	*checkfail_count = rem_error & READV_COUNT_MASK;
	rem_error >>= READV_COUNT_BITS;

	*first_off = rem_error & READV_COUNT_MASK;
	rem_error >>= READV_COUNT_BITS;
}

/*
 * Helper to encode errors for buffer_readv_complete()
 *
 * Errors are encoded as follows:
 * - bit 0 indicates whether any page was zeroed (1) or not (0)
 * - bit 1 indicates whether any checksum failure was ignored (1) or not (0)
 * - next READV_COUNT_BITS bits indicate the number of errored or zeroed pages
 * - next READV_COUNT_BITS bits indicate the number of checksum failures
 * - next READV_COUNT_BITS bits indicate the first offset of the first page
 *   that was errored or zeroed or, if no errors/zeroes, the first ignored
 *   checksum
 */
static inline void
buffer_readv_encode_error(PgAioResult *result,
						  bool is_temp,
						  bool zeroed_any,
						  bool ignored_any,
						  uint8 error_count,
						  uint8 zeroed_count,
						  uint8 checkfail_count,
						  uint8 first_error_off,
						  uint8 first_zeroed_off,
						  uint8 first_ignored_off)
{

	uint8		shift = 0;
	uint8		zeroed_or_error_count =
		error_count > 0 ? error_count : zeroed_count;
	uint8		first_off;

	StaticAssertDecl(PG_IOV_MAX <= 1 << READV_COUNT_BITS,
					 "PG_IOV_MAX is bigger than reserved space for error data");
	StaticAssertDecl((1 + 1 + 3 * READV_COUNT_BITS) <= PGAIO_RESULT_ERROR_BITS,
					 "PGAIO_RESULT_ERROR_BITS is insufficient for buffer_readv");

	/*
	 * We only have space to encode one offset - but luckily that's good
	 * enough. If there is an error, the error is the interesting offset, same
	 * with a zeroed buffer vs an ignored buffer.
	 */
	if (error_count > 0)
		first_off = first_error_off;
	else if (zeroed_count > 0)
		first_off = first_zeroed_off;
	else
		first_off = first_ignored_off;

	Assert(!zeroed_any || error_count == 0);

	result->error_data = 0;

	result->error_data |= zeroed_any << shift;
	shift += 1;

	result->error_data |= ignored_any << shift;
	shift += 1;

	result->error_data |= ((uint32) zeroed_or_error_count) << shift;
	shift += READV_COUNT_BITS;

	result->error_data |= ((uint32) checkfail_count) << shift;
	shift += READV_COUNT_BITS;

	result->error_data |= ((uint32) first_off) << shift;
	shift += READV_COUNT_BITS;

	result->id = is_temp ? PGAIO_HCB_LOCAL_BUFFER_READV :
		PGAIO_HCB_SHARED_BUFFER_READV;

	if (error_count > 0)
		result->status = PGAIO_RS_ERROR;
	else
		result->status = PGAIO_RS_WARNING;

	/*
	 * The encoding is complicated enough to warrant cross-checking it against
	 * the decode function.
	 */
#ifdef USE_ASSERT_CHECKING
	{
		bool		zeroed_any_2,
					ignored_any_2;
		uint8		zeroed_or_error_count_2,
					checkfail_count_2,
					first_off_2;

		buffer_readv_decode_error(*result,
								  &zeroed_any_2, &ignored_any_2,
								  &zeroed_or_error_count_2,
								  &checkfail_count_2,
								  &first_off_2);
		Assert(zeroed_any == zeroed_any_2);
		Assert(ignored_any == ignored_any_2);
		Assert(zeroed_or_error_count == zeroed_or_error_count_2);
		Assert(checkfail_count == checkfail_count_2);
		Assert(first_off == first_off_2);
	}
#endif

#undef READV_COUNT_BITS
#undef READV_COUNT_MASK
}

/*
 * Helper for AIO readv completion callbacks, supporting both shared and temp
 * buffers. Gets called once for each buffer in a multi-page read.
 */
static pg_attribute_always_inline void
buffer_readv_complete_one(PgAioTargetData *td, uint8 buf_off, Buffer buffer,
						  uint8 flags, bool failed, bool is_temp,
						  bool *buffer_invalid,
						  bool *failed_checksum,
						  bool *ignored_checksum,
						  bool *zeroed_buffer)
{
	BufferDesc *buf_hdr = is_temp ?
		GetLocalBufferDescriptor(-buffer - 1)
		: GetBufferDescriptor(buffer - 1);
	BufferTag	tag = buf_hdr->tag;
	char	   *bufdata = BufferGetBlock(buffer);
	uint64		set_flag_bits;
	int			piv_flags;

	/* check that the buffer is in the expected state for a read */
#ifdef USE_ASSERT_CHECKING
	{
		uint64		buf_state = pg_atomic_read_u64(&buf_hdr->state);

		Assert(buf_state & BM_TAG_VALID);
		Assert(!(buf_state & BM_VALID));
		/* temp buffers don't use BM_IO_IN_PROGRESS */
		if (!is_temp)
			Assert(buf_state & BM_IO_IN_PROGRESS);
		Assert(!(buf_state & BM_DIRTY));
	}
#endif

	*buffer_invalid = false;
	*failed_checksum = false;
	*ignored_checksum = false;
	*zeroed_buffer = false;

	/*
	 * We ask PageIsVerified() to only log the message about checksum errors,
	 * as the completion might be run in any backend (or IO workers). We will
	 * report checksum errors in buffer_readv_report().
	 */
	piv_flags = PIV_LOG_LOG;

	/* the local zero_damaged_pages may differ from the definer's */
	if (flags & READ_BUFFERS_IGNORE_CHECKSUM_FAILURES)
		piv_flags |= PIV_IGNORE_CHECKSUM_FAILURE;

	/*
	 * If the buffers are marked for zero on error, we want to log that in
	 * case of a checksum failure.
	 */
	if (flags & READ_BUFFERS_ZERO_ON_ERROR)
		piv_flags |= PIV_ZERO_BUFFERS_ON_ERROR;

	/* Check for garbage data. */
	if (!failed)
	{
		/*
		 * If the buffer is not currently pinned by this backend, e.g. because
		 * we're completing this IO after an error, the buffer data will have
		 * been marked as inaccessible when the buffer was unpinned. The AIO
		 * subsystem holds a pin, but that doesn't prevent the buffer from
		 * having been marked as inaccessible. The completion might also be
		 * executed in a different process.
		 */
#ifdef USE_VALGRIND
		if (!BufferIsPinned(buffer))
			VALGRIND_MAKE_MEM_DEFINED(bufdata, BLCKSZ);
#endif

		if (!PageIsVerified((Page) bufdata, tag.blockNum, piv_flags,
							failed_checksum))
		{
			if (flags & READ_BUFFERS_ZERO_ON_ERROR)
			{
				memset(bufdata, 0, BLCKSZ);
				*zeroed_buffer = true;
			}
			else
			{
				*buffer_invalid = true;
				/* mark buffer as having failed */
				failed = true;
			}
		}
		else if (*failed_checksum)
			*ignored_checksum = true;

		/* undo what we did above */
#ifdef USE_VALGRIND
		if (!BufferIsPinned(buffer))
			VALGRIND_MAKE_MEM_NOACCESS(bufdata, BLCKSZ);
#endif

		/*
		 * Immediately log a message about the invalid page, but only to the
		 * server log. The reason to do so immediately is that this may be
		 * executed in a different backend than the one that originated the
		 * request. The reason to do so immediately is that the originator
		 * might not process the query result immediately (because it is busy
		 * doing another part of query processing) or at all (e.g. if it was
		 * cancelled or errored out due to another IO also failing). The
		 * definer of the IO will emit an ERROR or WARNING when processing the
		 * IO's results
		 *
		 * To avoid duplicating the code to emit these log messages, we reuse
		 * buffer_readv_report().
		 */
		if (*buffer_invalid || *failed_checksum || *zeroed_buffer)
		{
			PgAioResult result_one = {0};

			buffer_readv_encode_error(&result_one, is_temp,
									  *zeroed_buffer,
									  *ignored_checksum,
									  *buffer_invalid,
									  *zeroed_buffer ? 1 : 0,
									  *failed_checksum ? 1 : 0,
									  buf_off, buf_off, buf_off);
			pgaio_result_report(result_one, td, LOG_SERVER_ONLY);
		}
	}

	/* Terminate I/O and set BM_VALID. */
	set_flag_bits = failed ? BM_IO_ERROR : BM_VALID;
	if (is_temp)
		TerminateLocalBufferIO(buf_hdr, false, set_flag_bits, true);
	else
		TerminateBufferIO(buf_hdr, false, set_flag_bits, false, true);

	/*
	 * Call the BUFFER_READ_DONE tracepoint in the callback, even though the
	 * callback may not be executed in the same backend that called
	 * BUFFER_READ_START. The alternative would be to defer calling the
	 * tracepoint to a later point (e.g. the local completion callback for
	 * shared buffer reads), which seems even less helpful.
	 */
	TRACE_POSTGRESQL_BUFFER_READ_DONE(tag.forkNum,
									  tag.blockNum,
									  tag.spcOid,
									  tag.dbOid,
									  tag.relNumber,
									  is_temp ? MyProcNumber : INVALID_PROC_NUMBER,
									  false);
}

/*
 * Perform completion handling of a single AIO read. This read may cover
 * multiple blocks / buffers.
 *
 * Shared between shared and local buffers, to reduce code duplication.
 */
static pg_attribute_always_inline PgAioResult
buffer_readv_complete(PgAioHandle *ioh, PgAioResult prior_result,
					  uint8 cb_data, bool is_temp)
{
	PgAioResult result = prior_result;
	PgAioTargetData *td = pgaio_io_get_target_data(ioh);
	uint8		first_error_off = 0;
	uint8		first_zeroed_off = 0;
	uint8		first_ignored_off = 0;
	uint8		error_count = 0;
	uint8		zeroed_count = 0;
	uint8		ignored_count = 0;
	uint8		checkfail_count = 0;
	uint64	   *io_data;
	uint8		handle_data_len;

	if (is_temp)
	{
		Assert(td->smgr.is_temp);
		Assert(pgaio_io_get_owner(ioh) == MyProcNumber);
	}
	else
		Assert(!td->smgr.is_temp);

	/*
	 * Iterate over all the buffers affected by this IO and call the
	 * per-buffer completion function for each buffer.
	 */
	io_data = pgaio_io_get_handle_data(ioh, &handle_data_len);
	for (uint8 buf_off = 0; buf_off < handle_data_len; buf_off++)
	{
		Buffer		buf = io_data[buf_off];
		bool		failed;
		bool		failed_verification = false;
		bool		failed_checksum = false;
		bool		zeroed_buffer = false;
		bool		ignored_checksum = false;

		Assert(BufferIsValid(buf));

		/*
		 * If the entire I/O failed on a lower-level, each buffer needs to be
		 * marked as failed. In case of a partial read, the first few buffers
		 * may be ok.
		 */
		failed =
			prior_result.status == PGAIO_RS_ERROR
			|| prior_result.result <= buf_off;

		buffer_readv_complete_one(td, buf_off, buf, cb_data, failed, is_temp,
								  &failed_verification,
								  &failed_checksum,
								  &ignored_checksum,
								  &zeroed_buffer);

		/*
		 * Track information about the number of different kinds of error
		 * conditions across all pages, as there can be multiple pages failing
		 * verification as part of one IO.
		 */
		if (failed_verification && !zeroed_buffer && error_count++ == 0)
			first_error_off = buf_off;
		if (zeroed_buffer && zeroed_count++ == 0)
			first_zeroed_off = buf_off;
		if (ignored_checksum && ignored_count++ == 0)
			first_ignored_off = buf_off;
		if (failed_checksum)
			checkfail_count++;
	}

	/*
	 * If the smgr read succeeded [partially] and page verification failed for
	 * some of the pages, adjust the IO's result state appropriately.
	 */
	if (prior_result.status != PGAIO_RS_ERROR &&
		(error_count > 0 || ignored_count > 0 || zeroed_count > 0))
	{
		buffer_readv_encode_error(&result, is_temp,
								  zeroed_count > 0, ignored_count > 0,
								  error_count, zeroed_count, checkfail_count,
								  first_error_off, first_zeroed_off,
								  first_ignored_off);
		pgaio_result_report(result, td, DEBUG1);
	}

	/*
	 * For shared relations this reporting is done in
	 * shared_buffer_readv_complete_local().
	 */
	if (is_temp && checkfail_count > 0)
		pgstat_report_checksum_failures_in_db(td->smgr.rlocator.dbOid,
											  checkfail_count);

	return result;
}

/*
 * AIO error reporting callback for aio_shared_buffer_readv_cb and
 * aio_local_buffer_readv_cb.
 *
 * The error is encoded / decoded in buffer_readv_encode_error() /
 * buffer_readv_decode_error().
 */
static void
buffer_readv_report(PgAioResult result, const PgAioTargetData *td,
					int elevel)
{
	int			nblocks = td->smgr.nblocks;
	BlockNumber first = td->smgr.blockNum;
	BlockNumber last = first + nblocks - 1;
	ProcNumber	errProc =
		td->smgr.is_temp ? MyProcNumber : INVALID_PROC_NUMBER;
	RelPathStr	rpath =
		relpathbackend(td->smgr.rlocator, errProc, td->smgr.forkNum);
	bool		zeroed_any,
				ignored_any;
	uint8		zeroed_or_error_count,
				checkfail_count,
				first_off;
	uint8		affected_count;
	const char *msg_one,
			   *msg_mult,
			   *det_mult,
			   *hint_mult;

	buffer_readv_decode_error(result, &zeroed_any, &ignored_any,
							  &zeroed_or_error_count,
							  &checkfail_count,
							  &first_off);

	/*
	 * Treat a read that had both zeroed buffers *and* ignored checksums as a
	 * special case, it's too irregular to be emitted the same way as the
	 * other cases.
	 */
	if (zeroed_any && ignored_any)
	{
		Assert(zeroed_any && ignored_any);
		Assert(nblocks > 1);	/* same block can't be both zeroed and ignored */
		Assert(result.status != PGAIO_RS_ERROR);
		affected_count = zeroed_or_error_count;

		ereport(elevel,
				errcode(ERRCODE_DATA_CORRUPTED),
				errmsg("zeroing %u page(s) and ignoring %u checksum failure(s) among blocks %u..%u of relation \"%s\"",
					   affected_count, checkfail_count, first, last, rpath.str),
				affected_count > 1 ?
				errdetail("Block %u held the first zeroed page.",
						  first + first_off) : 0,
				errhint_plural("See server log for details about the other %d invalid block.",
							   "See server log for details about the other %d invalid blocks.",
							   affected_count + checkfail_count - 1,
							   affected_count + checkfail_count - 1));
		return;
	}

	/*
	 * The other messages are highly repetitive. To avoid duplicating a long
	 * and complicated ereport(), gather the translated format strings
	 * separately and then do one common ereport.
	 */
	if (result.status == PGAIO_RS_ERROR)
	{
		Assert(!zeroed_any);	/* can't have invalid pages when zeroing them */
		affected_count = zeroed_or_error_count;
		msg_one = _("invalid page in block %u of relation \"%s\"");
		msg_mult = _("%u invalid pages among blocks %u..%u of relation \"%s\"");
		det_mult = _("Block %u held the first invalid page.");
		hint_mult = _("See server log for the other %u invalid block(s).");
	}
	else if (zeroed_any && !ignored_any)
	{
		affected_count = zeroed_or_error_count;
		msg_one = _("invalid page in block %u of relation \"%s\"; zeroing out page");
		msg_mult = _("zeroing out %u invalid pages among blocks %u..%u of relation \"%s\"");
		det_mult = _("Block %u held the first zeroed page.");
		hint_mult = _("See server log for the other %u zeroed block(s).");
	}
	else if (!zeroed_any && ignored_any)
	{
		affected_count = checkfail_count;
		msg_one = _("ignoring checksum failure in block %u of relation \"%s\"");
		msg_mult = _("ignoring %u checksum failures among blocks %u..%u of relation \"%s\"");
		det_mult = _("Block %u held the first ignored page.");
		hint_mult = _("See server log for the other %u ignored block(s).");
	}
	else
		pg_unreachable();

	ereport(elevel,
			errcode(ERRCODE_DATA_CORRUPTED),
			affected_count == 1 ?
			errmsg_internal(msg_one, first + first_off, rpath.str) :
			errmsg_internal(msg_mult, affected_count, first, last, rpath.str),
			affected_count > 1 ? errdetail_internal(det_mult, first + first_off) : 0,
			affected_count > 1 ? errhint_internal(hint_mult, affected_count - 1) : 0);
}

static void
shared_buffer_readv_stage(PgAioHandle *ioh, uint8 cb_data)
{
	buffer_stage_common(ioh, false, false);
}

static PgAioResult
shared_buffer_readv_complete(PgAioHandle *ioh, PgAioResult prior_result,
							 uint8 cb_data)
{
	return buffer_readv_complete(ioh, prior_result, cb_data, false);
}

/*
 * We need a backend-local completion callback for shared buffers, to be able
 * to report checksum errors correctly. Unfortunately that can only safely
 * happen if the reporting backend has previously called
 * pgstat_prepare_report_checksum_failure(), which we can only guarantee in
 * the backend that started the IO. Hence this callback.
 */
static PgAioResult
shared_buffer_readv_complete_local(PgAioHandle *ioh, PgAioResult prior_result,
								   uint8 cb_data)
{
	bool		zeroed_any,
				ignored_any;
	uint8		zeroed_or_error_count,
				checkfail_count,
				first_off;

	if (prior_result.status == PGAIO_RS_OK)
		return prior_result;

	buffer_readv_decode_error(prior_result,
							  &zeroed_any,
							  &ignored_any,
							  &zeroed_or_error_count,
							  &checkfail_count,
							  &first_off);

	if (checkfail_count)
	{
		PgAioTargetData *td = pgaio_io_get_target_data(ioh);

		pgstat_report_checksum_failures_in_db(td->smgr.rlocator.dbOid,
											  checkfail_count);
	}

	return prior_result;
}

static void
local_buffer_readv_stage(PgAioHandle *ioh, uint8 cb_data)
{
	buffer_stage_common(ioh, false, true);
}

static PgAioResult
local_buffer_readv_complete(PgAioHandle *ioh, PgAioResult prior_result,
							uint8 cb_data)
{
	return buffer_readv_complete(ioh, prior_result, cb_data, true);
}

/* readv callback is passed READ_BUFFERS_* flags as callback data */
const PgAioHandleCallbacks aio_shared_buffer_readv_cb = {
	.stage = shared_buffer_readv_stage,
	.complete_shared = shared_buffer_readv_complete,
	/* need a local callback to report checksum failures */
	.complete_local = shared_buffer_readv_complete_local,
	.report = buffer_readv_report,
};

/* readv callback is passed READ_BUFFERS_* flags as callback data */
const PgAioHandleCallbacks aio_local_buffer_readv_cb = {
	.stage = local_buffer_readv_stage,

	/*
	 * Note that this, in contrast to the shared_buffers case, uses
	 * complete_local, as only the issuing backend has access to the required
	 * datastructures. This is important in case the IO completion may be
	 * consumed incidentally by another backend.
	 */
	.complete_local = local_buffer_readv_complete,
	.report = buffer_readv_report,
};