lmdb: MDB_LIFORECLAIM & MDB_COALESCE modes.

Reclaim FreeDB in LIFO order - this is a main feature.
Also aim to coalesce small FreeDFB records.

Change-Id: I76aa062ef59359616e5b697233ec47cbd14f43cd

lmdb.h

@@ -292,6 +292,10 @@ typedef void (MDB_rel_func)(MDB_val *item, void *oldptr, void *newptr, void *rel
 #define MDB_NORDAHEAD	0x800000
 	/** don't initialize malloc'd memory before writing to datafile */
 #define MDB_NOMEMINIT	0x1000000
+	/** aim to coalesce FreeDB records */
+#define MDB_COALESCE	0x2000000
+	/** LIFO policy for reclaiming FreeDB records */
+#define MDB_LIFORECLAIM	0x4000000
 /** @} */
 
 /**	@defgroup	mdb_dbi_open	Database Flags
@@ -593,6 +597,12 @@ int  mdb_env_create(MDB_env **env);
 	 *		caller is expected to overwrite all of the memory that was
 	 *		reserved in that case.
 	 *		This flag may be changed at any time using #mdb_env_set_flags().
+	 *  <li>#MDB_COALESCE
+	 *		Aim to coalesce records while reclaiming FreeDB.
+	 *		This flag may be changed at any time using #mdb_env_set_flags().
+	 *  <li>#MDB_LIFORECLAIM
+	 *		LIFO policy for reclaiming FreeDB records. This significantly reduce
+	 *		write IPOS in case MDB_NOSYNC with periodically checkpoints.
 	 * </ul>
 	 * @param[in] mode The UNIX permissions to set on created files. This parameter
 	 * is ignored on Windows.

mdb.c

@@ -992,6 +992,8 @@ struct MDB_txn {
 	 */
 	txnid_t		mt_txnid;
 	MDB_env		*mt_env;		/**< the DB environment */
+	/** The list of reclaimed txns from freeDB */
+	MDB_IDL		mt_lifo_reclaimed;
 	/** The list of pages that became unused during this transaction.
 	 */
 	MDB_IDL		mt_free_pgs;
@@ -1100,6 +1102,7 @@ struct MDB_cursor {
 #define C_DEL	0x08			/**< last op was a cursor_del */
 #define C_SPLITTING	0x20		/**< Cursor is in page_split */
 #define C_UNTRACK	0x40		/**< Un-track cursor when closing */
+#define C_RECLAIMING	0x80		/**< FreeDB lookup is prohibited */
 /** @} */
 	unsigned int	mc_flags;	/**< @ref mdb_cursor */
 	MDB_page	*mc_pg[CURSOR_STACK];	/**< stack of pushed pages */
@@ -2056,31 +2059,25 @@ mdb_page_dirty(MDB_txn *txn, MDB_page *mp)
  *  will always be satisfied by a single contiguous chunk of memory.
  * @return 0 on success, non-zero on failure.
  */
+
 static int
 mdb_page_alloc(MDB_cursor *mc, int num, MDB_page **mp)
 {
-#ifdef MDB_PARANOID	/* Seems like we can ignore this now */
-	/* Get at most <Max_retries> more freeDB records once me_pghead
-	 * has enough pages.  If not enough, use new pages from the map.
-	 * If <Paranoid> and mc is updating the freeDB, only get new
-	 * records if me_pghead is empty. Then the freelist cannot play
-	 * catch-up with itself by growing while trying to save it.
-	 */
-	enum { Paranoid = 1, Max_retries = 500 };
-#else
-	enum { Paranoid = 0, Max_retries = INT_MAX /*infinite*/ };
-#endif
 	int rc, retry = num * 60;
 	MDB_txn *txn = mc->mc_txn;
 	MDB_env *env = txn->mt_env;
 	pgno_t pgno, *mop = env->me_pghead;
-	unsigned i, j, mop_len = mop ? mop[0] : 0, n2 = num-1;
+	unsigned i = 0, j, mop_len = mop ? mop[0] : 0, n2 = num-1;
 	MDB_page *np;
-	txnid_t oldest = 0, last;
+	txnid_t oldest = 0, last = 0;
 	MDB_cursor_op op;
 	MDB_cursor m2;
 	int found_old = 0;
+	unsigned enought = env->me_maxfree_1pg / 2;
 
+	/* mp == NULL when mdb_freelist_save() force reclaim to
+	 * get one more id for saving list of pages. */
+	if (mp) {
 		/* If there are any loose pages, just use them */
 		if (num == 1 && txn->mt_loose_pgs) {
 			np = txn->mt_loose_pgs;
@@ -2093,6 +2090,7 @@ mdb_page_alloc(MDB_cursor *mc, int num, MDB_page **mp)
 		}
 
 		*mp = NULL;
+	}
 
 	/* If our dirty list is already full, we can't do anything */
 	if (txn->mt_dirty_room == 0) {
@@ -2100,8 +2098,13 @@ mdb_page_alloc(MDB_cursor *mc, int num, MDB_page **mp)
 		goto fail;
 	}
 
+	int coalesce = (env->me_flags & MDB_COALESCE) ? 1 : 0;
+	if (coalesce && env->me_pgoldest == 0)
+		coalesce = 2;
+	const int lifo = (env->me_flags & MDB_LIFORECLAIM) != 0;
+
 oomkick_retry:;
-	for (op = MDB_FIRST;; op = MDB_NEXT) {
+	for (op = MDB_FIRST;; op = lifo ? MDB_PREV : MDB_NEXT) {
 		MDB_val key, data;
 		MDB_node *leaf;
 		pgno_t *idl;
@@ -2109,7 +2112,7 @@ oomkick_retry:;
 		/* Seek a big enough contiguous page range. Prefer
 		 * pages at the tail, just truncating the list.
 		 */
-		if (mop_len > n2) {
+		if (mp && mop_len > n2 && (! coalesce || op == MDB_FIRST)) {
 			i = mop_len;
 			do {
 				pgno = mop[i];
@@ -2122,23 +2125,42 @@ oomkick_retry:;
 
 		if (op == MDB_FIRST) {	/* 1st iteration */
 			/* Prepare to fetch more and coalesce */
-			last = env->me_pglast;
+			if (mc->mc_flags & C_RECLAIMING) {
+				/* If mc is updating the freeDB, then the freelist cannot play
+				 * catch-up with itself by growing while trying to save it.
+				 */
+				break;
+			}
 			oldest = env->me_pgoldest;
 			mdb_cursor_init(&m2, txn, FREE_DBI, NULL);
-			if (last) {
+			if (lifo) {
+				if (env->me_pglast > 1) {
+					/* Continue lookup from env->me_pglast */
+					last = env->me_pglast - 1;
 					op = MDB_SET_RANGE;
-				key.mv_data = &last; /* will look up last+1 */
+				} else {
+					oldest = mdb_find_oldest(txn);
+					env->me_pgoldest = oldest;
+					found_old = 1;
+					/* Begin from oldest reader if any */
+					if (oldest > 2) {
+						last = oldest - 1;
+						op = MDB_SET_RANGE;
+					}
+				}
+			} else if (env->me_pglast) {
+				/* Continue lookup from env->me_pglast */
+				last = env->me_pglast;
+				op = MDB_SET_RANGE;
+			}
+
+			key.mv_data = &last;
 			key.mv_size = sizeof(last);
 		}
-			if (Paranoid && mc->mc_dbi == FREE_DBI)
-				retry = -1;
-		}
-		if (Paranoid && retry < 0 && mop_len)
-			break;
 
-		last++;
+		if (! lifo) {
 			/* Do not fetch more if the record will be too recent */
-		if (oldest <= last) {
+			if (op != MDB_FIRST && ++last >= oldest) {
 				if (!found_old) {
 					oldest = mdb_find_oldest(txn);
 					env->me_pgoldest = oldest;
@@ -2147,12 +2169,29 @@ oomkick_retry:;
 				if (oldest <= last)
 					break;
 			}
+		}
+
 		rc = mdb_cursor_get(&m2, &key, NULL, op);
+		if (rc == MDB_NOTFOUND && lifo) {
+			if (op == MDB_SET_RANGE)
+				continue;
+			env->me_pgoldest = mdb_find_oldest(txn);
+			found_old = 1;
+			if (oldest < env->me_pgoldest) {
+				oldest = env->me_pgoldest;
+				last = oldest - 1;
+				key.mv_data = &last;
+				key.mv_size = sizeof(last);
+				op = MDB_SET_RANGE;
+				rc = mdb_cursor_get(&m2, &key, NULL, op);
+			}
+		}
 		if (rc) {
 			if (rc == MDB_NOTFOUND)
 				break;
 			goto fail;
 		}
+
 		last = *(txnid_t*)key.mv_data;
 		if (oldest <= last) {
 			if (!found_old) {
@@ -2160,15 +2199,38 @@ oomkick_retry:;
 				env->me_pgoldest = oldest;
 				found_old = 1;
 			}
-			if (oldest <= last)
+			if (oldest <= last) {
+				if (lifo)
+					continue;
 				break;
 			}
+		}
+
+		if (lifo) {
+			if (txn->mt_lifo_reclaimed) {
+				for(i = txn->mt_lifo_reclaimed[0]; i > 0; --i)
+					if (txn->mt_lifo_reclaimed[i] == last)
+						break;
+				if (i)
+					continue;
+			}
+		}
+
 		np = m2.mc_pg[m2.mc_top];
 		leaf = NODEPTR(np, m2.mc_ki[m2.mc_top]);
 		if ((rc = mdb_node_read(txn, leaf, &data)) != MDB_SUCCESS)
 			return rc;
 
+		if (lifo && !txn->mt_lifo_reclaimed) {
+			txn->mt_lifo_reclaimed = mdb_midl_alloc(env->me_maxfree_1pg);
+			if (!txn->mt_lifo_reclaimed) {
+				rc = ENOMEM;
+				goto fail;
+			}
+		}
+
 		idl = (MDB_ID *) data.mv_data;
+		mdb_tassert(txn, idl[0] == 0 || data.mv_size == (idl[0] + 1) * sizeof(MDB_ID));
 		i = idl[0];
 		if (!mop) {
 			if (!(env->me_pghead = mop = mdb_midl_alloc(i))) {
@@ -2180,6 +2242,10 @@ oomkick_retry:;
 				goto fail;
 			mop = env->me_pghead;
 		}
+		if (lifo) {
+			if ((rc = mdb_midl_append(&txn->mt_lifo_reclaimed, last)) != 0)
+				goto fail;
+		}
 		env->me_pglast = last;
 #if (MDB_DEBUG) > 1
 		DPRINTF(("IDL read txn %"Z"u root %"Z"u num %u",
@@ -2190,6 +2256,31 @@ oomkick_retry:;
 		/* Merge in descending sorted order */
 		mdb_midl_xmerge(mop, idl);
 		mop_len = mop[0];
+
+		if (! mp) {
+			/* force reclaim mode */
+			return 0;
+		}
+
+		/* Don't try to coalesce too much. */
+		if (mop_len > MDB_IDL_UM_SIZE / 2)
+			break;
+		if (coalesce == 1 && (mop_len >= enought || i >= enought / 2))
+			coalesce = 0;
+	}
+
+	if (! mp) {
+		/* force reclaim mode */
+		return MDB_NOTFOUND;
+	}
+
+	if (mop_len > n2 && coalesce) {
+		i = mop_len;
+		do {
+			pgno = mop[i];
+			if (mop[i-n2] == pgno+n2)
+				goto search_done;
+		} while (--i > n2);
 	}
 
 	/* Use new pages from the map when nothing suitable in the freeDB */
@@ -2197,7 +2288,7 @@ oomkick_retry:;
 	pgno = txn->mt_next_pgno;
 	if (pgno + num > env->me_maxpg) {
 		DPUTS("DB size maxed out");
-		if (mdb_oomkick_laggard(env))
+		if ((mc->mc_flags & C_RECLAIMING) == 0 && mdb_oomkick_laggard(env))
 			goto oomkick_retry;
 		rc = MDB_MAP_FULL;
 		goto fail;
@@ -2766,6 +2857,8 @@ mdb_txn_renew0(MDB_txn *txn)
 		txn->mt_free_pgs = env->me_free_pgs;
 		txn->mt_free_pgs[0] = 0;
 		txn->mt_spill_pgs = NULL;
+		if (txn->mt_lifo_reclaimed)
+			txn->mt_lifo_reclaimed[0] = 0;
 		env->me_txn = txn;
 		memcpy(txn->mt_dbiseqs, env->me_dbiseqs, env->me_maxdbs * sizeof(unsigned int));
 	}
@@ -2870,6 +2963,7 @@ mdb_txn_begin(MDB_env *env, MDB_txn *parent, unsigned int flags, MDB_txn **ret)
 			txn->mt_dbiseqs = (unsigned int *)(txn->mt_cursors + env->me_maxdbs);
 			txn->mt_dbflags = (unsigned char *)(txn->mt_dbiseqs + env->me_maxdbs);
 		}
+		txn->mt_dbxs = env->me_dbxs;
 	}
 	txn->mt_env = env;
 
@@ -3025,6 +3119,14 @@ mdb_txn_reset0(MDB_txn *txn, const char *act)
 			mdb_dlist_free(txn);
 		}
 
+		if (txn->mt_lifo_reclaimed) {
+			txn->mt_lifo_reclaimed[0] = 0;
+			if (txn != env->me_txn0) {
+				mdb_midl_free(txn->mt_lifo_reclaimed);
+				txn->mt_lifo_reclaimed = NULL;
+			}
+		}
+
 		if (!txn->mt_parent) {
 			if (mdb_midl_shrink(&txn->mt_free_pgs))
 				env->me_free_pgs = txn->mt_free_pgs;
@@ -3095,10 +3197,12 @@ mdb_freelist_save(MDB_txn *txn)
 	txnid_t	pglast = 0, head_id = 0;
 	pgno_t	freecnt = 0, *free_pgs, *mop;
 	ssize_t	head_room = 0, total_room = 0, mop_len, clean_limit;
+	unsigned cleanup_idx = 0, refill_idx = 0;
+	const int lifo = (env->me_flags & MDB_LIFORECLAIM) != 0;
 
 	mdb_cursor_init(&mc, txn, FREE_DBI, NULL);
 
-	if (env->me_pghead) {
+	if (! lifo && env->me_pghead) {
 		/* Make sure first page of freeDB is touched and on freelist */
 		rc = mdb_page_search(&mc, NULL, MDB_PS_FIRST|MDB_PS_MODIFY);
 		if (rc && rc != MDB_NOTFOUND)
@@ -3128,20 +3232,59 @@ mdb_freelist_save(MDB_txn *txn)
 		pgno_t *pgs;
 		ssize_t j;
 
+		if (! lifo) {
 			/* If using records from freeDB which we have not yet
 			 * deleted, delete them and any we reserved for me_pghead.
 			 */
 			while (pglast < env->me_pglast) {
+				/* The great answer is 42, and seems to be enough to prevent search in
+				 * mdb_page_alloc() during a deleting, when freeDB tree is unbalanced. */
+				while (!env->me_pghead || env->me_pghead[0] < 42) {
+					if (mdb_page_alloc(&mc, 0, NULL))
+						break;
+				}
 				rc = mdb_cursor_first(&mc, &key, NULL);
 				if (rc)
 					return rc;
 				pglast = head_id = *(txnid_t *)key.mv_data;
 				total_room = head_room = 0;
+				more = 1;
 				mdb_tassert(txn, pglast <= env->me_pglast);
+				mc.mc_flags |= C_RECLAIMING;
 				rc = mdb_cursor_del(&mc, 0);
+				mc.mc_flags &= ~C_RECLAIMING;
 				if (rc)
 					return rc;
 			}
+		} else if (txn->mt_lifo_reclaimed) {
+again:
+			/* LY: cleanup reclaimed records. */
+			while(cleanup_idx < txn->mt_lifo_reclaimed[0]) {
+				pglast = txn->mt_lifo_reclaimed[++cleanup_idx];
+				key.mv_data = &pglast;
+				key.mv_size = sizeof(pglast);
+				/* The great answer is 42, and seems to be enough to prevent search in
+				 * mdb_page_alloc() during a deleting, when freeDB tree is unbalanced. */
+				while (!env->me_pghead || env->me_pghead[0] < 42) {
+					if (mdb_page_alloc(&mc, 0, NULL)) {
+						rc = mdb_page_search(&mc, &key, MDB_PS_MODIFY);
+						if (rc && rc != MDB_NOTFOUND)
+							goto ballout;
+						break;
+					}
+				}
+				rc = mdb_cursor_get(&mc, &key, NULL, MDB_SET);
+				if (rc != MDB_NOTFOUND) {
+					if (rc)
+						goto ballout;
+					mc.mc_flags |= C_RECLAIMING;
+					rc = mdb_cursor_del(&mc, 0);
+					mc.mc_flags &= ~C_RECLAIMING;
+					if (rc)
+						goto ballout;
+				}
+			}
+		}
 
 		/* Save the IDL of pages freed by this txn, to a single record */
 		if (freecnt < txn->mt_free_pgs[0]) {
@@ -3149,7 +3292,7 @@ mdb_freelist_save(MDB_txn *txn)
 				/* Make sure last page of freeDB is touched and on freelist */
 				rc = mdb_page_search(&mc, NULL, MDB_PS_LAST|MDB_PS_MODIFY);
 				if (rc && rc != MDB_NOTFOUND)
-					return rc;
+					goto ballout;
 			}
 			free_pgs = txn->mt_free_pgs;
 			/* Write to last page of freeDB */
@@ -3160,7 +3303,7 @@ mdb_freelist_save(MDB_txn *txn)
 				data.mv_size = MDB_IDL_SIZEOF(free_pgs);
 				rc = mdb_cursor_put(&mc, &key, &data, MDB_RESERVE);
 				if (rc)
-					return rc;
+					goto ballout;
 				/* Retry if mt_free_pgs[] grew during the Put() */
 				free_pgs = txn->mt_free_pgs;
 			} while (freecnt < free_pgs[0]);
@@ -3181,6 +3324,9 @@ mdb_freelist_save(MDB_txn *txn)
 		mop = env->me_pghead;
 		mop_len = (mop ? mop[0] : 0) + txn->mt_loose_count;
 
+		if (mop_len && refill_idx == 0)
+			refill_idx = 1;
+
 		/* Reserve records for me_pghead[]. Split it if multi-page,
 		 * to avoid searching freeDB for a page range. Use keys in
 		 * range [1,me_pglast]: Smaller than txnid of oldest reader.
@@ -3191,8 +3337,46 @@ mdb_freelist_save(MDB_txn *txn)
 		} else if (head_room >= maxfree_1pg && head_id > 1) {
 			/* Keep current record (overflow page), add a new one */
 			head_id--;
+			refill_idx++;
 			head_room = 0;
 		}
+
+		if (lifo) {
+			if (refill_idx > (txn->mt_lifo_reclaimed ? txn->mt_lifo_reclaimed[0] : 0)) {
+				/* LY: need a more txn-id for save page list. */
+				rc = mdb_page_alloc(&mc, 0, NULL);
+				if (rc == 0)
+					/* LY: ок, reclaimed from freedb. */
+					continue;
+				if (rc != MDB_NOTFOUND)
+					/* LY: other troubles... */
+					goto ballout;
+
+				/* LY: freedb is empty, will look any free txn-id in high2low order. */
+				if (env->me_pglast < 1) {
+					/* LY: not any txn in the past of freedb. */
+					rc = MDB_MAP_FULL;
+					goto ballout;
+				}
+
+				if (! txn->mt_lifo_reclaimed) {
+					txn->mt_lifo_reclaimed = mdb_midl_alloc(env->me_maxfree_1pg);
+					if (! txn->mt_lifo_reclaimed) {
+						rc = ENOMEM;
+						goto ballout;
+					}
+				}
+				/* LY: append the list. */
+				rc = mdb_midl_append(&txn->mt_lifo_reclaimed, env->me_pglast - 1);
+				if (rc)
+					goto ballout;
+				--env->me_pglast;
+				/* LY: note that freeDB cleanup is not needed. */
+				++cleanup_idx;
+			}
+			head_id = txn->mt_lifo_reclaimed[refill_idx];
+		}
+
 		/* (Re)write {key = head_id, IDL length = head_room} */
 		total_room -= head_room;
 		head_room = mop_len - total_room;
@@ -3203,13 +3387,14 @@ mdb_freelist_save(MDB_txn *txn)
 		} else if (head_room < 0) {
 			/* Rare case, not bothering to delete this record */
 			head_room = 0;
+			continue;
 		}
 		key.mv_size = sizeof(head_id);
 		key.mv_data = &head_id;
 		data.mv_size = (head_room + 1) * sizeof(pgno_t);
 		rc = mdb_cursor_put(&mc, &key, &data, MDB_RESERVE);
 		if (rc)
-			return rc;
+			goto ballout;
 		/* IDL is initially empty, zero out at least the length */
 		pgs = (pgno_t *)data.mv_data;
 		j = head_room > clean_limit ? head_room : 0;
@@ -3219,6 +3404,8 @@ mdb_freelist_save(MDB_txn *txn)
 		total_room += head_room;
 	}
 
+	mdb_tassert(txn, cleanup_idx == (txn->mt_lifo_reclaimed ? txn->mt_lifo_reclaimed[0] : 0));
+
 	/* Return loose page numbers to me_pghead, though usually none are
 	 * left at this point.  The pages themselves remain in dirty_list.
 	 */
@@ -3228,7 +3415,7 @@ mdb_freelist_save(MDB_txn *txn)
 		MDB_IDL loose;
 		/* Room for loose pages + temp IDL with same */
 		if ((rc = mdb_midl_need(&env->me_pghead, 2*count+1)) != 0)
-			return rc;
+			goto ballout;
 		mop = env->me_pghead;
 		loose = mop + MDB_IDL_ALLOCLEN(mop) - count;
 		for (count = 0; mp; mp = NEXT_LOOSE_PAGE(mp))
@@ -3247,27 +3434,75 @@ mdb_freelist_save(MDB_txn *txn)
 		MDB_val key, data;
 
 		mop += mop_len;
+		if (! lifo) {
 			rc = mdb_cursor_first(&mc, &key, &data);
-		for (; !rc; rc = mdb_cursor_next(&mc, &key, &data, MDB_NEXT)) {
+			if (rc)
-			txnid_t id = *(txnid_t *)key.mv_data;
+				goto ballout;
-			ssize_t	len = (ssize_t)(data.mv_size / sizeof(MDB_ID)) - 1;
+		}
+
+		for(;;) {
+			txnid_t id;
+			ssize_t	len;
 			MDB_ID save;
 
-			mdb_tassert(txn, len >= 0 && id <= env->me_pglast);
+			if (! lifo) {
+				id = *(txnid_t *)key.mv_data;
+				mdb_tassert(txn, id <= env->me_pglast);
+			} else {
+				mdb_tassert(txn, refill_idx > 0 && refill_idx <= txn->mt_lifo_reclaimed[0]);
+				id = txn->mt_lifo_reclaimed[refill_idx--];
 				key.mv_data = &id;
-			if (len > mop_len) {
+				key.mv_size = sizeof(id);
+				rc = mdb_cursor_get(&mc, &key, &data, MDB_SET);
+				if (rc)
+					goto ballout;
+			}
+			mdb_tassert(txn, cleanup_idx == (txn->mt_lifo_reclaimed ? txn->mt_lifo_reclaimed[0] : 0));
+
+			len = (ssize_t)(data.mv_size / sizeof(MDB_ID)) - 1;
+			mdb_tassert(txn, len >= 0);
+			if (len > mop_len)
 				len = mop_len;
 			data.mv_size = (len + 1) * sizeof(MDB_ID);
-			}
+			key.mv_data = &id;
+			key.mv_size = sizeof(id);
 			data.mv_data = mop -= len;
+
 			save = mop[0];
 			mop[0] = len;
 			rc = mdb_cursor_put(&mc, &key, &data, MDB_CURRENT);
+			mdb_tassert(txn, cleanup_idx == (txn->mt_lifo_reclaimed ? txn->mt_lifo_reclaimed[0] : 0));
 			mop[0] = save;
-			if (rc || !(mop_len -= len))
+			if (rc || (mop_len -= len) == 0)
-				break;
+				goto ballout;
+
+			if (! lifo) {
+				rc = mdb_cursor_next(&mc, &key, &data, MDB_NEXT);
+				if (rc)
+					goto ballout;
 			}
 		}
+	}
+
+ballout:
+	if (txn->mt_lifo_reclaimed) {
+		mdb_tassert(txn, rc || cleanup_idx == txn->mt_lifo_reclaimed[0]);
+		if (rc == 0 && cleanup_idx != txn->mt_lifo_reclaimed[0]) {
+			mdb_tassert(txn, cleanup_idx < txn->mt_lifo_reclaimed[0]);
+			/* LY: zeroed cleanup_idx to force cleanup & refill created freeDB records. */
+			cleanup_idx = 0;
+			/* LY: restart filling */
+			refill_idx = total_room = head_room = 0;
+			more = 1;
+			goto again;
+		}
+		txn->mt_lifo_reclaimed[0] = 0;
+		if (txn != env->me_txn0) {
+			mdb_midl_free(txn->mt_lifo_reclaimed);
+			txn->mt_lifo_reclaimed = NULL;
+		}
+	}
+
 	return rc;
 }
 
@@ -3460,6 +3695,18 @@ mdb_txn_commit(MDB_txn *txn)
 		MDB_IDL pspill;
 		unsigned x, y, len, ps_len;
 
+		/* Append our reclaim list to parent's */
+		if (txn->mt_lifo_reclaimed) {
+			if (parent->mt_lifo_reclaimed) {
+				rc = mdb_midl_append_list(&parent->mt_lifo_reclaimed, txn->mt_lifo_reclaimed);
+				if (rc)
+					goto fail;
+				mdb_midl_free(txn->mt_lifo_reclaimed);
+			} else
+				parent->mt_lifo_reclaimed = txn->mt_lifo_reclaimed;
+			txn->mt_lifo_reclaimed = NULL;
+		}
+
 		/* Append our free list to parent's */
 		rc = mdb_midl_append_list(&parent->mt_free_pgs, txn->mt_free_pgs);
 		if (rc)
@@ -4707,9 +4954,10 @@ fail:
 	 *	at runtime. Changing other flags requires closing the
 	 *	environment and re-opening it with the new flags.
 	 */
-#define	CHANGEABLE	(MDB_NOSYNC|MDB_NOMETASYNC|MDB_MAPASYNC|MDB_NOMEMINIT)
+#define	CHANGEABLE	(MDB_NOSYNC|MDB_NOMETASYNC|MDB_MAPASYNC| \
+    MDB_NOMEMINIT|MDB_COALESCE)
 #define	CHANGELESS	(MDB_FIXEDMAP|MDB_NOSUBDIR|MDB_RDONLY| \
-	MDB_WRITEMAP|MDB_NOTLS|MDB_NOLOCK|MDB_NORDAHEAD)
+    MDB_WRITEMAP|MDB_NOTLS|MDB_NOLOCK|MDB_NORDAHEAD|MDB_LIFORECLAIM)
 
 #if VALID_FLAGS & PERSISTENT_FLAGS & (CHANGEABLE|CHANGELESS)
 # error "Persistent DB flags & env flags overlap, but both go in mm_flags"
@@ -4879,6 +5127,8 @@ mdb_env_close0(MDB_env *env, int excl)
 	free(env->me_dbxs);
 	free(env->me_path);
 	free(env->me_dirty_list);
+	if (env->me_txn0)
+		mdb_midl_free(env->me_txn0->mt_lifo_reclaimed);
 	free(env->me_txn0);
 	mdb_midl_free(env->me_free_pgs);
 
@@ -7301,6 +7551,7 @@ mdb_cursor_init(MDB_cursor *mc, MDB_txn *txn, MDB_dbi dbi, MDB_xcursor *mx)
 	mc->mc_top = 0;
 	mc->mc_pg[0] = 0;
 	mc->mc_flags = 0;
+	mc->mc_ki[0] = 0;
 	if (txn->mt_dbs[dbi].md_flags & MDB_DUPSORT) {
 		mdb_tassert(txn, mx != NULL);
 		mc->mc_xcursor = mx;