Adding upstream version 13.4.upstream/13.4upstream

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

1 files changed, 872 insertions, 0 deletions

diff --git a/src/common/pg_lzcompress.c b/src/common/pg_lzcompress.c
new file mode 100644
index 0000000..f9c2982
--- /dev/null
+++ b/src/common/pg_lzcompress.c
@@ -0,0 +1,872 @@
+/* ----------
+ * pg_lzcompress.c -
+ *
+ *		This is an implementation of LZ compression for PostgreSQL.
+ *		It uses a simple history table and generates 2-3 byte tags
+ *		capable of backward copy information for 3-273 bytes with
+ *		a max offset of 4095.
+ *
+ *		Entry routines:
+ *
+ *			int32
+ *			pglz_compress(const char *source, int32 slen, char *dest,
+ *						  const PGLZ_Strategy *strategy);
+ *
+ *				source is the input data to be compressed.
+ *
+ *				slen is the length of the input data.
+ *
+ *				dest is the output area for the compressed result.
+ *					It must be at least as big as PGLZ_MAX_OUTPUT(slen).
+ *
+ *				strategy is a pointer to some information controlling
+ *					the compression algorithm. If NULL, the compiled
+ *					in default strategy is used.
+ *
+ *				The return value is the number of bytes written in the
+ *				buffer dest, or -1 if compression fails; in the latter
+ *				case the contents of dest are undefined.
+ *
+ *			int32
+ *			pglz_decompress(const char *source, int32 slen, char *dest,
+ *							int32 rawsize, bool check_complete)
+ *
+ *				source is the compressed input.
+ *
+ *				slen is the length of the compressed input.
+ *
+ *				dest is the area where the uncompressed data will be
+ *					written to. It is the callers responsibility to
+ *					provide enough space.
+ *
+ *					The data is written to buff exactly as it was handed
+ *					to pglz_compress(). No terminating zero byte is added.
+ *
+ *				rawsize is the length of the uncompressed data.
+ *
+ *				check_complete is a flag to let us know if -1 should be
+ *					returned in cases where we don't reach the end of the
+ *					source or dest buffers, or not.  This should be false
+ *					if the caller is asking for only a partial result and
+ *					true otherwise.
+ *
+ *				The return value is the number of bytes written in the
+ *				buffer dest, or -1 if decompression fails.
+ *
+ *		The decompression algorithm and internal data format:
+ *
+ *			It is made with the compressed data itself.
+ *
+ *			The data representation is easiest explained by describing
+ *			the process of decompression.
+ *
+ *			If compressed_size == rawsize, then the data
+ *			is stored uncompressed as plain bytes. Thus, the decompressor
+ *			simply copies rawsize bytes to the destination.
+ *
+ *			Otherwise the first byte tells what to do the next 8 times.
+ *			We call this the control byte.
+ *
+ *			An unset bit in the control byte means, that one uncompressed
+ *			byte follows, which is copied from input to output.
+ *
+ *			A set bit in the control byte means, that a tag of 2-3 bytes
+ *			follows. A tag contains information to copy some bytes, that
+ *			are already in the output buffer, to the current location in
+ *			the output. Let's call the three tag bytes T1, T2 and T3. The
+ *			position of the data to copy is coded as an offset from the
+ *			actual output position.
+ *
+ *			The offset is in the upper nibble of T1 and in T2.
+ *			The length is in the lower nibble of T1.
+ *
+ *			So the 16 bits of a 2 byte tag are coded as
+ *
+ *				7---T1--0  7---T2--0
+ *				OOOO LLLL  OOOO OOOO
+ *
+ *			This limits the offset to 1-4095 (12 bits) and the length
+ *			to 3-18 (4 bits) because 3 is always added to it. To emit
+ *			a tag of 2 bytes with a length of 2 only saves one control
+ *			bit. But we lose one byte in the possible length of a tag.
+ *
+ *			In the actual implementation, the 2 byte tag's length is
+ *			limited to 3-17, because the value 0xF in the length nibble
+ *			has special meaning. It means, that the next following
+ *			byte (T3) has to be added to the length value of 18. That
+ *			makes total limits of 1-4095 for offset and 3-273 for length.
+ *
+ *			Now that we have successfully decoded a tag. We simply copy
+ *			the output that occurred <offset> bytes back to the current
+ *			output location in the specified <length>. Thus, a
+ *			sequence of 200 spaces (think about bpchar fields) could be
+ *			coded in 4 bytes. One literal space and a three byte tag to
+ *			copy 199 bytes with a -1 offset. Whow - that's a compression
+ *			rate of 98%! Well, the implementation needs to save the
+ *			original data size too, so we need another 4 bytes for it
+ *			and end up with a total compression rate of 96%, what's still
+ *			worth a Whow.
+ *
+ *		The compression algorithm
+ *
+ *			The following uses numbers used in the default strategy.
+ *
+ *			The compressor works best for attributes of a size between
+ *			1K and 1M. For smaller items there's not that much chance of
+ *			redundancy in the character sequence (except for large areas
+ *			of identical bytes like trailing spaces) and for bigger ones
+ *			our 4K maximum look-back distance is too small.
+ *
+ *			The compressor creates a table for lists of positions.
+ *			For each input position (except the last 3), a hash key is
+ *			built from the 4 next input bytes and the position remembered
+ *			in the appropriate list. Thus, the table points to linked
+ *			lists of likely to be at least in the first 4 characters
+ *			matching strings. This is done on the fly while the input
+ *			is compressed into the output area.  Table entries are only
+ *			kept for the last 4096 input positions, since we cannot use
+ *			back-pointers larger than that anyway.  The size of the hash
+ *			table is chosen based on the size of the input - a larger table
+ *			has a larger startup cost, as it needs to be initialized to
+ *			zero, but reduces the number of hash collisions on long inputs.
+ *
+ *			For each byte in the input, its hash key (built from this
+ *			byte and the next 3) is used to find the appropriate list
+ *			in the table. The lists remember the positions of all bytes
+ *			that had the same hash key in the past in increasing backward
+ *			offset order. Now for all entries in the used lists, the
+ *			match length is computed by comparing the characters from the
+ *			entries position with the characters from the actual input
+ *			position.
+ *
+ *			The compressor starts with a so called "good_match" of 128.
+ *			It is a "prefer speed against compression ratio" optimizer.
+ *			So if the first entry looked at already has 128 or more
+ *			matching characters, the lookup stops and that position is
+ *			used for the next tag in the output.
+ *
+ *			For each subsequent entry in the history list, the "good_match"
+ *			is lowered by 10%. So the compressor will be more happy with
+ *			short matches the farer it has to go back in the history.
+ *			Another "speed against ratio" preference characteristic of
+ *			the algorithm.
+ *
+ *			Thus there are 3 stop conditions for the lookup of matches:
+ *
+ *				- a match >= good_match is found
+ *				- there are no more history entries to look at
+ *				- the next history entry is already too far back
+ *				  to be coded into a tag.
+ *
+ *			Finally the match algorithm checks that at least a match
+ *			of 3 or more bytes has been found, because that is the smallest
+ *			amount of copy information to code into a tag. If so, a tag
+ *			is omitted and all the input bytes covered by that are just
+ *			scanned for the history add's, otherwise a literal character
+ *			is omitted and only his history entry added.
+ *
+ *		Acknowledgments:
+ *
+ *			Many thanks to Adisak Pochanayon, who's article about SLZ
+ *			inspired me to write the PostgreSQL compression this way.
+ *
+ *			Jan Wieck
+ *
+ * Copyright (c) 1999-2020, PostgreSQL Global Development Group
+ *
+ * src/common/pg_lzcompress.c
+ * ----------
+ */
+#ifndef FRONTEND
+#include "postgres.h"
+#else
+#include "postgres_fe.h"
+#endif
+
+#include <limits.h>
+
+#include "common/pg_lzcompress.h"
+
+
+/* ----------
+ * Local definitions
+ * ----------
+ */
+#define PGLZ_MAX_HISTORY_LISTS	8192	/* must be power of 2 */
+#define PGLZ_HISTORY_SIZE		4096
+#define PGLZ_MAX_MATCH			273
+
+
+/* ----------
+ * PGLZ_HistEntry -
+ *
+ *		Linked list for the backward history lookup
+ *
+ * All the entries sharing a hash key are linked in a doubly linked list.
+ * This makes it easy to remove an entry when it's time to recycle it
+ * (because it's more than 4K positions old).
+ * ----------
+ */
+typedef struct PGLZ_HistEntry
+{
+	struct PGLZ_HistEntry *next;	/* links for my hash key's list */
+	struct PGLZ_HistEntry *prev;
+	int			hindex;			/* my current hash key */
+	const char *pos;			/* my input position */
+} PGLZ_HistEntry;
+
+
+/* ----------
+ * The provided standard strategies
+ * ----------
+ */
+static const PGLZ_Strategy strategy_default_data = {
+	32,							/* Data chunks less than 32 bytes are not
+								 * compressed */
+	INT_MAX,					/* No upper limit on what we'll try to
+								 * compress */
+	25,							/* Require 25% compression rate, or not worth
+								 * it */
+	1024,						/* Give up if no compression in the first 1KB */
+	128,						/* Stop history lookup if a match of 128 bytes
+								 * is found */
+	10							/* Lower good match size by 10% at every loop
+								 * iteration */
+};
+const PGLZ_Strategy *const PGLZ_strategy_default = &strategy_default_data;
+
+
+static const PGLZ_Strategy strategy_always_data = {
+	0,							/* Chunks of any size are compressed */
+	INT_MAX,
+	0,							/* It's enough to save one single byte */
+	INT_MAX,					/* Never give up early */
+	128,						/* Stop history lookup if a match of 128 bytes
+								 * is found */
+	6							/* Look harder for a good match */
+};
+const PGLZ_Strategy *const PGLZ_strategy_always = &strategy_always_data;
+
+
+/* ----------
+ * Statically allocated work arrays for history
+ * ----------
+ */
+static int16 hist_start[PGLZ_MAX_HISTORY_LISTS];
+static PGLZ_HistEntry hist_entries[PGLZ_HISTORY_SIZE + 1];
+
+/*
+ * Element 0 in hist_entries is unused, and means 'invalid'. Likewise,
+ * INVALID_ENTRY_PTR in next/prev pointers mean 'invalid'.
+ */
+#define INVALID_ENTRY			0
+#define INVALID_ENTRY_PTR		(&hist_entries[INVALID_ENTRY])
+
+/* ----------
+ * pglz_hist_idx -
+ *
+ *		Computes the history table slot for the lookup by the next 4
+ *		characters in the input.
+ *
+ * NB: because we use the next 4 characters, we are not guaranteed to
+ * find 3-character matches; they very possibly will be in the wrong
+ * hash list.  This seems an acceptable tradeoff for spreading out the
+ * hash keys more.
+ * ----------
+ */
+#define pglz_hist_idx(_s,_e, _mask) (										\
+			((((_e) - (_s)) < 4) ? (int) (_s)[0] :							\
+			 (((_s)[0] << 6) ^ ((_s)[1] << 4) ^								\
+			  ((_s)[2] << 2) ^ (_s)[3])) & (_mask)				\
+		)
+
+
+/* ----------
+ * pglz_hist_add -
+ *
+ *		Adds a new entry to the history table.
+ *
+ * If _recycle is true, then we are recycling a previously used entry,
+ * and must first delink it from its old hashcode's linked list.
+ *
+ * NOTE: beware of multiple evaluations of macro's arguments, and note that
+ * _hn and _recycle are modified in the macro.
+ * ----------
+ */
+#define pglz_hist_add(_hs,_he,_hn,_recycle,_s,_e, _mask)	\
+do {									\
+			int __hindex = pglz_hist_idx((_s),(_e), (_mask));				\
+			int16 *__myhsp = &(_hs)[__hindex];								\
+			PGLZ_HistEntry *__myhe = &(_he)[_hn];							\
+			if (_recycle) {													\
+				if (__myhe->prev == NULL)									\
+					(_hs)[__myhe->hindex] = __myhe->next - (_he);			\
+				else														\
+					__myhe->prev->next = __myhe->next;						\
+				if (__myhe->next != NULL)									\
+					__myhe->next->prev = __myhe->prev;						\
+			}																\
+			__myhe->next = &(_he)[*__myhsp];								\
+			__myhe->prev = NULL;											\
+			__myhe->hindex = __hindex;										\
+			__myhe->pos  = (_s);											\
+			/* If there was an existing entry in this hash slot, link */	\
+			/* this new entry to it. However, the 0th entry in the */		\
+			/* entries table is unused, so we can freely scribble on it. */ \
+			/* So don't bother checking if the slot was used - we'll */		\
+			/* scribble on the unused entry if it was not, but that's */	\
+			/* harmless. Avoiding the branch in this critical path */		\
+			/* speeds this up a little bit. */								\
+			/* if (*__myhsp != INVALID_ENTRY) */							\
+				(_he)[(*__myhsp)].prev = __myhe;							\
+			*__myhsp = _hn;													\
+			if (++(_hn) >= PGLZ_HISTORY_SIZE + 1) {							\
+				(_hn) = 1;													\
+				(_recycle) = true;											\
+			}																\
+} while (0)
+
+
+/* ----------
+ * pglz_out_ctrl -
+ *
+ *		Outputs the last and allocates a new control byte if needed.
+ * ----------
+ */
+#define pglz_out_ctrl(__ctrlp,__ctrlb,__ctrl,__buf) \
+do { \
+	if ((__ctrl & 0xff) == 0)												\
+	{																		\
+		*(__ctrlp) = __ctrlb;												\
+		__ctrlp = (__buf)++;												\
+		__ctrlb = 0;														\
+		__ctrl = 1;															\
+	}																		\
+} while (0)
+
+
+/* ----------
+ * pglz_out_literal -
+ *
+ *		Outputs a literal byte to the destination buffer including the
+ *		appropriate control bit.
+ * ----------
+ */
+#define pglz_out_literal(_ctrlp,_ctrlb,_ctrl,_buf,_byte) \
+do { \
+	pglz_out_ctrl(_ctrlp,_ctrlb,_ctrl,_buf);								\
+	*(_buf)++ = (unsigned char)(_byte);										\
+	_ctrl <<= 1;															\
+} while (0)
+
+
+/* ----------
+ * pglz_out_tag -
+ *
+ *		Outputs a backward reference tag of 2-4 bytes (depending on
+ *		offset and length) to the destination buffer including the
+ *		appropriate control bit.
+ * ----------
+ */
+#define pglz_out_tag(_ctrlp,_ctrlb,_ctrl,_buf,_len,_off) \
+do { \
+	pglz_out_ctrl(_ctrlp,_ctrlb,_ctrl,_buf);								\
+	_ctrlb |= _ctrl;														\
+	_ctrl <<= 1;															\
+	if (_len > 17)															\
+	{																		\
+		(_buf)[0] = (unsigned char)((((_off) & 0xf00) >> 4) | 0x0f);		\
+		(_buf)[1] = (unsigned char)(((_off) & 0xff));						\
+		(_buf)[2] = (unsigned char)((_len) - 18);							\
+		(_buf) += 3;														\
+	} else {																\
+		(_buf)[0] = (unsigned char)((((_off) & 0xf00) >> 4) | ((_len) - 3)); \
+		(_buf)[1] = (unsigned char)((_off) & 0xff);							\
+		(_buf) += 2;														\
+	}																		\
+} while (0)
+
+
+/* ----------
+ * pglz_find_match -
+ *
+ *		Lookup the history table if the actual input stream matches
+ *		another sequence of characters, starting somewhere earlier
+ *		in the input buffer.
+ * ----------
+ */
+static inline int
+pglz_find_match(int16 *hstart, const char *input, const char *end,
+				int *lenp, int *offp, int good_match, int good_drop, int mask)
+{
+	PGLZ_HistEntry *hent;
+	int16		hentno;
+	int32		len = 0;
+	int32		off = 0;
+
+	/*
+	 * Traverse the linked history list until a good enough match is found.
+	 */
+	hentno = hstart[pglz_hist_idx(input, end, mask)];
+	hent = &hist_entries[hentno];
+	while (hent != INVALID_ENTRY_PTR)
+	{
+		const char *ip = input;
+		const char *hp = hent->pos;
+		int32		thisoff;
+		int32		thislen;
+
+		/*
+		 * Stop if the offset does not fit into our tag anymore.
+		 */
+		thisoff = ip - hp;
+		if (thisoff >= 0x0fff)
+			break;
+
+		/*
+		 * Determine length of match. A better match must be larger than the
+		 * best so far. And if we already have a match of 16 or more bytes,
+		 * it's worth the call overhead to use memcmp() to check if this match
+		 * is equal for the same size. After that we must fallback to
+		 * character by character comparison to know the exact position where
+		 * the diff occurred.
+		 */
+		thislen = 0;
+		if (len >= 16)
+		{
+			if (memcmp(ip, hp, len) == 0)
+			{
+				thislen = len;
+				ip += len;
+				hp += len;
+				while (ip < end && *ip == *hp && thislen < PGLZ_MAX_MATCH)
+				{
+					thislen++;
+					ip++;
+					hp++;
+				}
+			}
+		}
+		else
+		{
+			while (ip < end && *ip == *hp && thislen < PGLZ_MAX_MATCH)
+			{
+				thislen++;
+				ip++;
+				hp++;
+			}
+		}
+
+		/*
+		 * Remember this match as the best (if it is)
+		 */
+		if (thislen > len)
+		{
+			len = thislen;
+			off = thisoff;
+		}
+
+		/*
+		 * Advance to the next history entry
+		 */
+		hent = hent->next;
+
+		/*
+		 * Be happy with lesser good matches the more entries we visited. But
+		 * no point in doing calculation if we're at end of list.
+		 */
+		if (hent != INVALID_ENTRY_PTR)
+		{
+			if (len >= good_match)
+				break;
+			good_match -= (good_match * good_drop) / 100;
+		}
+	}
+
+	/*
+	 * Return match information only if it results at least in one byte
+	 * reduction.
+	 */
+	if (len > 2)
+	{
+		*lenp = len;
+		*offp = off;
+		return 1;
+	}
+
+	return 0;
+}
+
+
+/* ----------
+ * pglz_compress -
+ *
+ *		Compresses source into dest using strategy. Returns the number of
+ *		bytes written in buffer dest, or -1 if compression fails.
+ * ----------
+ */
+int32
+pglz_compress(const char *source, int32 slen, char *dest,
+			  const PGLZ_Strategy *strategy)
+{
+	unsigned char *bp = (unsigned char *) dest;
+	unsigned char *bstart = bp;
+	int			hist_next = 1;
+	bool		hist_recycle = false;
+	const char *dp = source;
+	const char *dend = source + slen;
+	unsigned char ctrl_dummy = 0;
+	unsigned char *ctrlp = &ctrl_dummy;
+	unsigned char ctrlb = 0;
+	unsigned char ctrl = 0;
+	bool		found_match = false;
+	int32		match_len;
+	int32		match_off;
+	int32		good_match;
+	int32		good_drop;
+	int32		result_size;
+	int32		result_max;
+	int32		need_rate;
+	int			hashsz;
+	int			mask;
+
+	/*
+	 * Our fallback strategy is the default.
+	 */
+	if (strategy == NULL)
+		strategy = PGLZ_strategy_default;
+
+	/*
+	 * If the strategy forbids compression (at all or if source chunk size out
+	 * of range), fail.
+	 */
+	if (strategy->match_size_good <= 0 ||
+		slen < strategy->min_input_size ||
+		slen > strategy->max_input_size)
+		return -1;
+
+	/*
+	 * Limit the match parameters to the supported range.
+	 */
+	good_match = strategy->match_size_good;
+	if (good_match > PGLZ_MAX_MATCH)
+		good_match = PGLZ_MAX_MATCH;
+	else if (good_match < 17)
+		good_match = 17;
+
+	good_drop = strategy->match_size_drop;
+	if (good_drop < 0)
+		good_drop = 0;
+	else if (good_drop > 100)
+		good_drop = 100;
+
+	need_rate = strategy->min_comp_rate;
+	if (need_rate < 0)
+		need_rate = 0;
+	else if (need_rate > 99)
+		need_rate = 99;
+
+	/*
+	 * Compute the maximum result size allowed by the strategy, namely the
+	 * input size minus the minimum wanted compression rate.  This had better
+	 * be <= slen, else we might overrun the provided output buffer.
+	 */
+	if (slen > (INT_MAX / 100))
+	{
+		/* Approximate to avoid overflow */
+		result_max = (slen / 100) * (100 - need_rate);
+	}
+	else
+		result_max = (slen * (100 - need_rate)) / 100;
+
+	/*
+	 * Experiments suggest that these hash sizes work pretty well. A large
+	 * hash table minimizes collision, but has a higher startup cost. For a
+	 * small input, the startup cost dominates. The table size must be a power
+	 * of two.
+	 */
+	if (slen < 128)
+		hashsz = 512;
+	else if (slen < 256)
+		hashsz = 1024;
+	else if (slen < 512)
+		hashsz = 2048;
+	else if (slen < 1024)
+		hashsz = 4096;
+	else
+		hashsz = 8192;
+	mask = hashsz - 1;
+
+	/*
+	 * Initialize the history lists to empty.  We do not need to zero the
+	 * hist_entries[] array; its entries are initialized as they are used.
+	 */
+	memset(hist_start, 0, hashsz * sizeof(int16));
+
+	/*
+	 * Compress the source directly into the output buffer.
+	 */
+	while (dp < dend)
+	{
+		/*
+		 * If we already exceeded the maximum result size, fail.
+		 *
+		 * We check once per loop; since the loop body could emit as many as 4
+		 * bytes (a control byte and 3-byte tag), PGLZ_MAX_OUTPUT() had better
+		 * allow 4 slop bytes.
+		 */
+		if (bp - bstart >= result_max)
+			return -1;
+
+		/*
+		 * If we've emitted more than first_success_by bytes without finding
+		 * anything compressible at all, fail.  This lets us fall out
+		 * reasonably quickly when looking at incompressible input (such as
+		 * pre-compressed data).
+		 */
+		if (!found_match && bp - bstart >= strategy->first_success_by)
+			return -1;
+
+		/*
+		 * Try to find a match in the history
+		 */
+		if (pglz_find_match(hist_start, dp, dend, &match_len,
+							&match_off, good_match, good_drop, mask))
+		{
+			/*
+			 * Create the tag and add history entries for all matched
+			 * characters.
+			 */
+			pglz_out_tag(ctrlp, ctrlb, ctrl, bp, match_len, match_off);
+			while (match_len--)
+			{
+				pglz_hist_add(hist_start, hist_entries,
+							  hist_next, hist_recycle,
+							  dp, dend, mask);
+				dp++;			/* Do not do this ++ in the line above! */
+				/* The macro would do it four times - Jan.  */
+			}
+			found_match = true;
+		}
+		else
+		{
+			/*
+			 * No match found. Copy one literal byte.
+			 */
+			pglz_out_literal(ctrlp, ctrlb, ctrl, bp, *dp);
+			pglz_hist_add(hist_start, hist_entries,
+						  hist_next, hist_recycle,
+						  dp, dend, mask);
+			dp++;				/* Do not do this ++ in the line above! */
+			/* The macro would do it four times - Jan.  */
+		}
+	}
+
+	/*
+	 * Write out the last control byte and check that we haven't overrun the
+	 * output size allowed by the strategy.
+	 */
+	*ctrlp = ctrlb;
+	result_size = bp - bstart;
+	if (result_size >= result_max)
+		return -1;
+
+	/* success */
+	return result_size;
+}
+
+
+/* ----------
+ * pglz_decompress -
+ *
+ *		Decompresses source into dest. Returns the number of bytes
+ *		decompressed into the destination buffer, or -1 if the
+ *		compressed data is corrupted.
+ *
+ *		If check_complete is true, the data is considered corrupted
+ *		if we don't exactly fill the destination buffer.  Callers that
+ *		are extracting a slice typically can't apply this check.
+ * ----------
+ */
+int32
+pglz_decompress(const char *source, int32 slen, char *dest,
+				int32 rawsize, bool check_complete)
+{
+	const unsigned char *sp;
+	const unsigned char *srcend;
+	unsigned char *dp;
+	unsigned char *destend;
+
+	sp = (const unsigned char *) source;
+	srcend = ((const unsigned char *) source) + slen;
+	dp = (unsigned char *) dest;
+	destend = dp + rawsize;
+
+	while (sp < srcend && dp < destend)
+	{
+		/*
+		 * Read one control byte and process the next 8 items (or as many as
+		 * remain in the compressed input).
+		 */
+		unsigned char ctrl = *sp++;
+		int			ctrlc;
+
+		for (ctrlc = 0; ctrlc < 8 && sp < srcend && dp < destend; ctrlc++)
+		{
+			if (ctrl & 1)
+			{
+				/*
+				 * Set control bit means we must read a match tag. The match
+				 * is coded with two bytes. First byte uses lower nibble to
+				 * code length - 3. Higher nibble contains upper 4 bits of the
+				 * offset. The next following byte contains the lower 8 bits
+				 * of the offset. If the length is coded as 18, another
+				 * extension tag byte tells how much longer the match really
+				 * was (0-255).
+				 */
+				int32		len;
+				int32		off;
+
+				len = (sp[0] & 0x0f) + 3;
+				off = ((sp[0] & 0xf0) << 4) | sp[1];
+				sp += 2;
+				if (len == 18)
+					len += *sp++;
+
+				/*
+				 * Check for corrupt data: if we fell off the end of the
+				 * source, or if we obtained off = 0, we have problems.  (We
+				 * must check this, else we risk an infinite loop below in the
+				 * face of corrupt data.)
+				 */
+				if (unlikely(sp > srcend || off == 0))
+					return -1;
+
+				/*
+				 * Don't emit more data than requested.
+				 */
+				len = Min(len, destend - dp);
+
+				/*
+				 * Now we copy the bytes specified by the tag from OUTPUT to
+				 * OUTPUT (copy len bytes from dp - off to dp).  The copied
+				 * areas could overlap, so to avoid undefined behavior in
+				 * memcpy(), be careful to copy only non-overlapping regions.
+				 *
+				 * Note that we cannot use memmove() instead, since while its
+				 * behavior is well-defined, it's also not what we want.
+				 */
+				while (off < len)
+				{
+					/*
+					 * We can safely copy "off" bytes since that clearly
+					 * results in non-overlapping source and destination.
+					 */
+					memcpy(dp, dp - off, off);
+					len -= off;
+					dp += off;
+
+					/*----------
+					 * This bit is less obvious: we can double "off" after
+					 * each such step.  Consider this raw input:
+					 *		112341234123412341234
+					 * This will be encoded as 5 literal bytes "11234" and
+					 * then a match tag with length 16 and offset 4.  After
+					 * memcpy'ing the first 4 bytes, we will have emitted
+					 *		112341234
+					 * so we can double "off" to 8, then after the next step
+					 * we have emitted
+					 *		11234123412341234
+					 * Then we can double "off" again, after which it is more
+					 * than the remaining "len" so we fall out of this loop
+					 * and finish with a non-overlapping copy of the
+					 * remainder.  In general, a match tag with off < len
+					 * implies that the decoded data has a repeat length of
+					 * "off".  We can handle 1, 2, 4, etc repetitions of the
+					 * repeated string per memcpy until we get to a situation
+					 * where the final copy step is non-overlapping.
+					 *
+					 * (Another way to understand this is that we are keeping
+					 * the copy source point dp - off the same throughout.)
+					 *----------
+					 */
+					off += off;
+				}
+				memcpy(dp, dp - off, len);
+				dp += len;
+			}
+			else
+			{
+				/*
+				 * An unset control bit means LITERAL BYTE. So we just copy
+				 * one from INPUT to OUTPUT.
+				 */
+				*dp++ = *sp++;
+			}
+
+			/*
+			 * Advance the control bit
+			 */
+			ctrl >>= 1;
+		}
+	}
+
+	/*
+	 * If requested, check we decompressed the right amount.
+	 */
+	if (check_complete && (dp != destend || sp != srcend))
+		return -1;
+
+	/*
+	 * That's it.
+	 */
+	return (char *) dp - dest;
+}
+
+
+/* ----------
+ * pglz_max_compressed_size -
+ *
+ *		Calculate the maximum compressed size for a given amount of raw data.
+ *		Return the maximum size, or total compressed size if maximum size is
+ *		larger than total compressed size.
+ *
+ * We can't use PGLZ_MAX_OUTPUT for this purpose, because that's used to size
+ * the compression buffer (and abort the compression). It does not really say
+ * what's the maximum compressed size for an input of a given length, and it
+ * may happen that while the whole value is compressible (and thus fits into
+ * PGLZ_MAX_OUTPUT nicely), the prefix is not compressible at all.
+ * ----------
+ */
+int32
+pglz_maximum_compressed_size(int32 rawsize, int32 total_compressed_size)
+{
+	int64		compressed_size;
+
+	/*
+	 * pglz uses one control bit per byte, so if the entire desired prefix is
+	 * represented as literal bytes, we'll need (rawsize * 9) bits.  We care
+	 * about bytes though, so be sure to round up not down.
+	 *
+	 * Use int64 here to prevent overflow during calculation.
+	 */
+	compressed_size = ((int64) rawsize * 9 + 7) / 8;
+
+	/*
+	 * The above fails to account for a corner case: we could have compressed
+	 * data that starts with N-1 or N-2 literal bytes and then has a match tag
+	 * of 2 or 3 bytes.  It's therefore possible that we need to fetch 1 or 2
+	 * more bytes in order to have the whole match tag.  (Match tags earlier
+	 * in the compressed data don't cause a problem, since they should
+	 * represent more decompressed bytes than they occupy themselves.)
+	 */
+	compressed_size += 2;
+
+	/*
+	 * Maximum compressed size can't be larger than total compressed size.
+	 * (This also ensures that our result fits in int32.)
+	 */
+	compressed_size = Min(compressed_size, total_compressed_size);
+
+	return (int32) compressed_size;
+}