lmdb: seek to steady meta-page on db-open.

This is 7/9 for https://github.com/ReOpen/ReOpenLDAP/issues/1 and https://github.com/ReOpen/ReOpenLDAP/issues/2 Change-Id: If59a6bfc7c6198e6b1e85f4bdc5b534ecff03123
2025-10-31 03:29:01 +08:00 · 2015-05-12 21:10:43 +03:00
parent 633f2a10ad
commit 21705fd9a2
1 changed files with 122 additions and 66 deletions
--- a/mdb.c
+++ b/mdb.c
@@ -1891,21 +1891,62 @@ static uint64_t mdb_meta_sign(MDB_meta *meta) {
 	return (sign > MDB_DATASIGN_WEAK) ? sign : ~sign;
 }

-/** Check both meta pages to see which one is newer.
- * @param[in] env the environment handle
- * @return pointer to last meta-page.
- */
-static MDB_meta*
-mdb_env_meta_head(const MDB_env *env) {
+static MDB_meta* mdb_meta_head_w(MDB_env *env) {
 	MDB_meta* a = METAPAGE_1(env);
 	MDB_meta* b = METAPAGE_2(env);
-	return (a->mm_txnid > b->mm_txnid) ? a : b;
+	txnid_t head_txnid = env->me_txns->mti_txnid;
+
+	mdb_assert(env, a->mm_txnid != b->mm_txnid || head_txnid == 0);
+	if (likely(a->mm_txnid == head_txnid))
+		return a;
+	if (likely(b->mm_txnid == head_txnid))
+		return b;
+
+	mdb_assert(env, head_txnid == a->mm_txnid || head_txnid == b->mm_txnid);
+	env->me_flags |= MDB_FATAL_ERROR;
+	return a;
+}
+
+static MDB_meta* mdb_meta_head_r(MDB_env *env) {
+	MDB_meta* a = METAPAGE_1(env);
+	MDB_meta* b = METAPAGE_2(env), *h;
+	txnid_t head_txnid;
+	int loop = 0, rc;
+
+	do {
+		head_txnid = env->me_txns->mti_txnid;
+
+		mdb_assert(env, a->mm_txnid != b->mm_txnid || head_txnid == 0);
+		if (likely(a->mm_txnid == head_txnid))
+			return a;
+		if (likely(b->mm_txnid == head_txnid))
+			return b;
+
+		/* LY: got a race on env->me_txns->mti_txnid with mdb_env_sync0() */
+#if defined(__i386__) || defined(__x86_64__)
+		__asm__ __volatile__("pause");
+#endif
+		mdb_coherent_barrier();
+		if (loop > 2)
+			pthread_yield();
+	} while (++loop < 5);
+
+	rc = mdb_mutex_lock(env, MDB_MUTEX(env, w));
+	h = mdb_meta_head_w(env);
+	if (rc == 0)
+		mdb_mutex_unlock(env, MDB_MUTEX(env, w));
+	return h;
 }

 static MDB_meta* mdb_env_meta_flipflop(const MDB_env *env, MDB_meta* meta) {
 	return (meta == METAPAGE_1(env)) ? METAPAGE_2(env) : METAPAGE_1(env);
 }

+static int mdb_meta_lt(MDB_meta* a, MDB_meta* b) {
+	return (META_IS_STEADY(a) == META_IS_STEADY(b))
+			? a->mm_txnid < b->mm_txnid : META_IS_STEADY(b);
+}
+
 /** Find oldest txnid still referenced. */
 static txnid_t
 mdb_find_oldest(MDB_env *env, int *laggard)
@@ -1941,7 +1982,7 @@ mdb_oomkick(MDB_env *env)
 {
 	int reader, retry;
 	txnid_t snap, oldest = mdb_find_oldest(env, &reader);
-	MDB_meta* head = mdb_env_meta_head(env);
+	MDB_meta* head = mdb_meta_head_w(env);
 	MDB_meta* tail = mdb_env_meta_flipflop(env, head);

 	if (META_IS_WEAK(head) && oldest == tail->mm_txnid) {
@@ -1980,12 +2021,12 @@ mdb_oomkick(MDB_env *env)
 			continue;

 		rc = env->me_oom_func(env, pid, (void*) tid, oldest,
-			mdb_env_meta_head(env)->mm_txnid - oldest, retry);
+			mdb_meta_head_w(env)->mm_txnid - oldest, retry);
 		if (rc < 0)
 			break;

 		if (rc) {
-			r->mr_txnid = (txnid_t)-1;
+			r->mr_txnid = (txnid_t)-1L;
 			if (rc > 1) {
 				r->mr_tid = 0;
 				r->mr_pid = 0;
@@ -2505,7 +2546,7 @@ mdb_env_sync(MDB_env *env, int force)
 	if (unlikely(flags & (MDB_RDONLY | MDB_FATAL_ERROR)))
 		return EACCES;

-	head = mdb_env_meta_head(env);
+	head = mdb_meta_head_r(env);
 	if (force || head->mm_mapsize != env->me_mapsize)
 		flags &= MDB_WRITEMAP;

@@ -2523,7 +2564,7 @@ mdb_env_sync(MDB_env *env, int force)
 		} else if (fdatasync(env->me_fd))
 			return errno;
 		/* LY: head may be changed during the sync. */
-		head = mdb_env_meta_head(env);
+		head = mdb_meta_head_r(env);
 	}

 	if (! META_IS_WEAK(head) && env->me_sync_pending == 0
@@ -2537,7 +2578,7 @@ mdb_env_sync(MDB_env *env, int force)
 		return rc;

 	/* LY: head may be changed while the mutex has been acquired. */
-	head = mdb_env_meta_head(env);
+	head = mdb_meta_head_w(env);
 	rc = MDB_SUCCESS;
 	if (META_IS_WEAK(head) || env->me_sync_pending != 0
 			|| env->me_mapsize != head->mm_mapsize) {
@@ -2731,7 +2772,7 @@ mdb_txn_renew0(MDB_txn *txn)
 		}

 		do { /* LY: Retry on a race, ITS#7970. */
-			meta = mdb_env_meta_head(env);
+			meta = mdb_meta_head_r(env);
 			r->mr_txnid = meta->mm_txnid;
 			mdb_coherent_barrier();
 			memcpy(txn->mt_dbs, meta->mm_dbs, 2 * sizeof(MDB_db));
@@ -2746,7 +2787,7 @@ mdb_txn_renew0(MDB_txn *txn)
 		if (unlikely(rc))
 			return rc;

-		meta = mdb_env_meta_head(env);
+		meta = mdb_meta_head_w(env);
 		txn->mt_txnid = meta->mm_txnid;

 		/* Setup db info */
@@ -2991,7 +3032,7 @@ mdb_txn_straggler(MDB_txn *txn, int *percent)
 		return -1;

 	env = txn->mt_env;
-	meta = mdb_env_meta_head(env);
+	meta = mdb_meta_head_r(env);
 	if (percent) {
 		long cent = env->me_maxpg / 100;
 		long last = env->me_txn ? env->me_txn0->mt_next_pgno : meta->mm_last_pg;
@@ -3794,6 +3835,8 @@ mdb_env_read_header(MDB_env *env, MDB_meta *meta)
 	 * Read both meta pages so we can use the latest one.
 	 */

+	meta->mm_datasync_sign = MDB_DATASIGN_WEAK;
+	meta->mm_txnid = 0;
 	for (i=off=0; i<2; i++, off = meta->mm_psize) {
 		rc = pread(env->me_fd, &pbuf, Size, off);
 		if (rc != Size) {
@@ -3823,10 +3866,18 @@ mdb_env_read_header(MDB_env *env, MDB_meta *meta)
 			return MDB_VERSION_MISMATCH;
 		}

-		if (off == 0 || m->mm_txnid > meta->mm_txnid)
+		if (m->mm_datasync_sign > MDB_DATASIGN_WEAK && m->mm_datasync_sign != mdb_meta_sign(m))
+			continue;
+
+		if (mdb_meta_lt(meta, m))
 			*meta = *m;
 	}
-	return 0;
+
+	if (meta->mm_datasync_sign == MDB_DATASIGN_WEAK)
+		/* LY: Both meta-pages are weak. */
+		return MDB_CORRUPTED;
+
+	return MDB_SUCCESS;
 }

 /** Fill in most of the zeroed #MDB_meta for an empty database environment */
@@ -3890,7 +3941,7 @@ static int
 mdb_env_sync0(MDB_env *env, unsigned flags, MDB_meta *pending)
 {
 	int rc;
-	MDB_meta* head = mdb_env_meta_head(env);
+	MDB_meta* head = mdb_meta_head_w(env);
 	size_t prev_mapsize = head->mm_mapsize;
 	MDB_meta* tail = META_IS_WEAK(head) ? head : mdb_env_meta_flipflop(env, head);
 	off_t offset = (char*) tail - env->me_map;
@@ -3898,8 +3949,7 @@ mdb_env_sync0(MDB_env *env, unsigned flags, MDB_meta *pending)
 	mdb_assert(env, (env->me_flags & (MDB_RDONLY | MDB_FATAL_ERROR)) == 0);
 	mdb_assert(env, META_IS_WEAK(head) || env->me_sync_pending != 0
 			   || env->me_mapsize != prev_mapsize);
-	mdb_assert(env, pending->mm_txnid > head->mm_txnid
-			   || (pending->mm_txnid == head->mm_txnid && META_IS_WEAK(head)));
+	mdb_assert(env, pending->mm_txnid > head->mm_txnid || META_IS_WEAK(head));
 	mdb_assert(env, pending->mm_txnid > tail->mm_txnid || META_IS_WEAK(tail));

 	MDB_meta* stay = mdb_env_meta_flipflop(env, tail);
@@ -3991,8 +4041,9 @@ mdb_env_sync0(MDB_env *env, unsigned flags, MDB_meta *pending)
 		}
 		mdb_invalidate_cache(env->me_map + offset, sizeof(MDB_meta));
 	}
+	env->me_txns->mti_txnid = pending->mm_txnid;

-	/* LY: step#3 - sync updated meta-pages. */
+	/* LY: step#3 - sync meta-pages. */
 	if ((flags & (MDB_NOSYNC | MDB_NOMETASYNC)) == 0) {
 		if (env->me_flags & MDB_WRITEMAP) {
 			char* ptr = env->me_map + (offset & ~(env->me_os_psize - 1));
@@ -4120,7 +4171,7 @@ mdb_env_set_mapsize(MDB_env *env, size_t size)
 		void *old;
 		if (env->me_txn)
 			return EINVAL;
-		meta = mdb_env_meta_head(env);
+		meta = mdb_meta_head_w(env);
 		if (!size)
 			size = meta->mm_mapsize;
 		{
@@ -4188,13 +4239,12 @@ mdb_fsize(HANDLE fd, size_t *size)
 /** Further setup required for opening an LMDB environment
 */
 static int ESECT
-mdb_env_open2(MDB_env *env)
+mdb_env_open2(MDB_env *env, MDB_meta *meta)
 {
 	unsigned flags = env->me_flags;
 	int i, newenv = 0, rc;
-	MDB_meta meta;

-	if ((i = mdb_env_read_header(env, &meta)) != 0) {
+	if ((i = mdb_env_read_header(env, meta)) != 0) {
 		if (i != ENOENT)
 			return i;
 		mdb_debug("new mdbenv");
@@ -4202,26 +4252,26 @@ mdb_env_open2(MDB_env *env)
 		env->me_psize = env->me_os_psize;
 		if (env->me_psize > MAX_PAGESIZE)
 			env->me_psize = MAX_PAGESIZE;
-		memset(&meta, 0, sizeof(meta));
-		mdb_env_init_meta0(env, &meta);
-		meta.mm_mapsize = DEFAULT_MAPSIZE;
+		memset(meta, 0, sizeof(*meta));
+		mdb_env_init_meta0(env, meta);
+		meta->mm_mapsize = DEFAULT_MAPSIZE;
 	} else {
-		env->me_psize = meta.mm_psize;
+		env->me_psize = meta->mm_psize;
 	}

 	/* Was a mapsize configured? */
 	if (!env->me_mapsize) {
-		env->me_mapsize = meta.mm_mapsize;
+		env->me_mapsize = meta->mm_mapsize;
 	}
 	{
 		/* Make sure mapsize >= committed data size.  Even when using
 		 * mm_mapsize, which could be broken in old files (ITS#7789).
 		 */
-		size_t minsize = (meta.mm_last_pg + 1) * meta.mm_psize;
+		size_t minsize = (meta->mm_last_pg + 1) * meta->mm_psize;
 		if (env->me_mapsize < minsize)
 			env->me_mapsize = minsize;
 	}
-	meta.mm_mapsize = env->me_mapsize;
+	meta->mm_mapsize = env->me_mapsize;

 	if (newenv && !(flags & MDB_FIXEDMAP)) {
 		/* mdb_env_map() may grow the datafile.  Write the metapages
@@ -4231,20 +4281,20 @@ mdb_env_open2(MDB_env *env)
 		 * program might end up doing that - one with a memory layout
 		 * and map address which does not suit the main program.
 		 */
-		rc = mdb_env_init_meta(env, &meta);
+		rc = mdb_env_init_meta(env, meta);
 		if (rc)
 			return rc;
 		newenv = 0;
 	}

-	rc = mdb_env_map(env, (flags & MDB_FIXEDMAP) ? meta.mm_address : NULL);
+	rc = mdb_env_map(env, (flags & MDB_FIXEDMAP) ? meta->mm_address : NULL);
 	if (rc)
 		return rc;

 	if (newenv) {
 		if (flags & MDB_FIXEDMAP)
-			meta.mm_address = env->me_map;
-		i = mdb_env_init_meta(env, &meta);
+			meta->mm_address = env->me_map;
+		i = mdb_env_init_meta(env, meta);
 		if (i != MDB_SUCCESS) {
 			return i;
 		}
@@ -4258,24 +4308,6 @@ mdb_env_open2(MDB_env *env)
 #endif
 	env->me_maxpg = env->me_mapsize / env->me_psize;

-#if MDB_DEBUG
-	{
-		MDB_meta *meta = mdb_env_meta_head(env);
-		MDB_db *db = &meta->mm_dbs[MAIN_DBI];
-		int toggle = ((char*) meta == PAGEDATA(env->me_map)) ? 0 : 1;
-
-		mdb_debug("opened database version %u, pagesize %u",
-			meta->mm_version, env->me_psize);
-		mdb_debug("using meta page %d, txn %zu", toggle, meta->mm_txnid);
-		mdb_debug("depth: %u",             db->md_depth);
-		mdb_debug("entries: %zu",        db->md_entries);
-		mdb_debug("branch pages: %zu",   db->md_branch_pages);
-		mdb_debug("leaf pages: %zu",     db->md_leaf_pages);
-		mdb_debug("overflow pages: %zu", db->md_overflow_pages);
-		mdb_debug("root: %zu",           db->md_root);
-	}
-#endif
-
 	return MDB_SUCCESS;
 }

@@ -4295,10 +4327,9 @@ mdb_env_reader_dest(void *ptr)

 /** Downgrade the exclusive lock on the region back to shared */
 static int ESECT
-mdb_env_share_locks(MDB_env *env, int *excl)
+mdb_env_share_locks(MDB_env *env, int *excl, MDB_meta *meta)
 {
 	struct flock lock_info;
-	MDB_meta *meta = mdb_env_meta_head(env);
 	int rc = 0;

 	env->me_txns->mti_txnid = meta->mm_txnid;
@@ -4507,7 +4538,7 @@ mdb_env_setup_locks(MDB_env *env, char *lpath, int mode, int *excl)

 		env->me_txns->mti_magic = MDB_MAGIC;
 		env->me_txns->mti_format = MDB_LOCK_FORMAT;
-		env->me_txns->mti_txnid = 0;
+		env->me_txns->mti_txnid = ~0L;
 		env->me_txns->mti_numreaders = 0;
 	} else {
 		if (env->me_txns->mti_magic != MDB_MAGIC) {
@@ -4629,10 +4660,11 @@ mdb_env_open(MDB_env *env, const char *path, unsigned flags, mode_t mode)
 			goto leave;
 	}

-	if ((rc = mdb_env_open2(env)) == MDB_SUCCESS) {
+	MDB_meta meta;
+	if ((rc = mdb_env_open2(env, &meta)) == MDB_SUCCESS) {
 		mdb_debug("opened dbenv %p", (void *) env);
 		if (excl > 0) {
-			rc = mdb_env_share_locks(env, &excl);
+			rc = mdb_env_share_locks(env, &excl, &meta);
 			if (rc)
 				goto leave;
 		}
@@ -4656,6 +4688,24 @@ mdb_env_open(MDB_env *env, const char *path, unsigned flags, mode_t mode)
 		}
 	}

+#if MDB_DEBUG
+	{
+		MDB_meta *meta = mdb_meta_head_r(env);
+		MDB_db *db = &meta->mm_dbs[MAIN_DBI];
+		int toggle = ((char*) meta == PAGEDATA(env->me_map)) ? 0 : 1;
+
+		mdb_debug("opened database version %u, pagesize %u",
+			meta->mm_version, env->me_psize);
+		mdb_debug("using meta page %d, txn %zu", toggle, meta->mm_txnid);
+		mdb_debug("depth: %u",             db->md_depth);
+		mdb_debug("entries: %zu",        db->md_entries);
+		mdb_debug("branch pages: %zu",   db->md_branch_pages);
+		mdb_debug("leaf pages: %zu",     db->md_leaf_pages);
+		mdb_debug("overflow pages: %zu", db->md_overflow_pages);
+		mdb_debug("root: %zu",           db->md_root);
+	}
+#endif
+
 leave:
 	if (rc)
 		mdb_env_close0(env);
@@ -8982,7 +9032,7 @@ mdb_env_stat(MDB_env *env, MDB_stat *arg)
 	if (env == NULL || arg == NULL)
 		return EINVAL;

-	meta = mdb_env_meta_head(env);
+	meta = mdb_meta_head_r(env);
 	return mdb_stat0(env, &meta->mm_dbs[MAIN_DBI], arg);
 }

@@ -8994,7 +9044,7 @@ mdb_env_info(MDB_env *env, MDB_envinfo *arg)
 	if (env == NULL || arg == NULL)
 		return EINVAL;

-	meta = mdb_env_meta_head(env);
+	meta = mdb_meta_head_r(env);
 	arg->me_mapaddr = meta->mm_address;
 	arg->me_mapsize = env->me_mapsize;
 	arg->me_maxreaders = env->me_maxreaders;
@@ -9513,7 +9563,6 @@ static int mdb_mutex_failed(MDB_env *env, pthread_mutex_t *mutex, int rc)
 #ifdef EOWNERDEAD
 	if (unlikely(rc == EOWNERDEAD)) {
 		int rlocked, rc2;
-		MDB_meta *meta;

 		/* We own the mutex. Clean up after dead previous owner. */
 		rc = MDB_SUCCESS;
@@ -9522,8 +9571,13 @@ static int mdb_mutex_failed(MDB_env *env, pthread_mutex_t *mutex, int rc)
 			/* Keep mti_txnid updated, otherwise next writer can
 			 * overwrite data which latest meta page refers to.
 			 */
-			meta = mdb_env_meta_head(env);
+	#if 0
+			/* LY: Hm, how this can happen, if the mti_txnid
+			 * is updating only at the finish of a successful commit ? */
+
+			MDB_meta *meta = mdb_env_meta_head(env);
 			env->me_txns->mti_txnid = meta->mm_txnid;
+	#endif
 			/* env is hosed if the dead thread was ours */
 			if (env->me_txn) {
 				env->me_flags |= MDB_FATAL_ERROR;
@@ -9544,8 +9598,10 @@ static int mdb_mutex_failed(MDB_env *env, pthread_mutex_t *mutex, int rc)
 #endif /* EOWNERDEAD */
 	if (unlikely(rc)) {
 		mdb_debug("lock mutex failed, %s", mdb_strerror(rc));
-		env->me_flags |= MDB_FATAL_ERROR;
-		rc = MDB_PANIC;
+		if (rc != EDEADLK) {
+			env->me_flags |= MDB_FATAL_ERROR;
+			rc = MDB_PANIC;
+		}
 	}

 	return rc;