Fixed lfs_dir_fetchmatch not understanding overwritten tags

Sometimes small, single line code change hides behind it a complicated
story. This is one of those times.

If you look at this diff, you may note that this is a case of
lfs_dir_fetchmatch not correctly handling a tag that invalidates a
callback used to search for some condition, in this case a search for a
parent, which is invalidated by a later dir tag overwritting the
previous dir pair.

But how can this happen? Dir-pair-tags are only overwritten during
relocations (when a block goes bad or exceeds the block_cycles config
option for dynamic wear-leveling). Other dir operations create new
directory entries. And the only lfs_dir_fetchmatch condition that relies
on overwrites (as opposed to proper deletes) is when we need to find a
directory's parent, an operation that only occurs during a _different_
relocation. And a false _positive_, can only happen if we don't have a
parent. Which is really unlikely when we search for directory parents!

This bug and minimal test case was found by Matthew Renzelmann. In a
unfortunate series of events, first a file creation causes a directory
split to occur. This creates a new, orphaned metadata-pair containing
our new file. However, the revision count on this metadata-pair
indicates the pair is due for relocation as a part of wear-leveling.
Normally, this is fine, even though this metadata-pair has no parent,
the lfs_dir_find should return ENOENT and continue without error.
However, here we get hit by our fetchmatch bug. A previous, unrelated
relocation overwrites a pair which just happens to contain the block
allocated for a new metadata-pair. When we search for a parent,
lfs_dir_fetchmatch incorrectly finds this old, outdated metadata pair
and incorrectly tells our orphan it's found its parent.

As you can imagine the orphan's dissapointment must be immense.

So an unfortunately timed dir split triggers a relocation which
incorrectly finds a previously written parent that has been outdated
by another relocation.

As a solution we can outdate our found tag if it is overwritten by
an exact match during lfs_dir_fetchmatch.

As a part of this I started adding a new set of tests: tests/test_relocations,
for aggressive relocations tests. This is already by appended to by
another PR. I suspect relocations is relatively under-tested and is
becoming more important due to recent improvements in wear-leveling.

.travis.yml

@@ -20,6 +20,7 @@ script:
   # run tests with a few different configurations
   - make test QUIET=1 CFLAGS+="-DLFS_READ_SIZE=1      -DLFS_CACHE_SIZE=4"
   - make test QUIET=1 CFLAGS+="-DLFS_READ_SIZE=512    -DLFS_CACHE_SIZE=512 -DLFS_BLOCK_CYCLES=16"
+  - make test QUIET=1 CFLAGS+="-DLFS_READ_SIZE=8      -DLFS_CACHE_SIZE=16  -DLFS_BLOCK_CYCLES=2"
   - make test QUIET=1 CFLAGS+="-DLFS_BLOCK_COUNT=1023 -DLFS_LOOKAHEAD_SIZE=256"
 
   - make clean test QUIET=1 CFLAGS+="-DLFS_INLINE_MAX=0"

DESIGN.md

@@ -254,7 +254,7 @@ have weaknesses that limit their usefulness. But if we merge the two they can
 mutually solve each other's limitations.
 
 This is the idea behind littlefs. At the sub-block level, littlefs is built
-out of small, two blocks logs that provide atomic updates to metadata anywhere
+out of small, two block logs that provide atomic updates to metadata anywhere
 on the filesystem. At the super-block level, littlefs is a CObW tree of blocks
 that can be evicted on demand.
 
@@ -676,7 +676,7 @@ block, this cost is fairly reasonable.
 ---
 
 This is a new data structure, so we still have several questions. What is the
-storage overage? Can the number of pointers exceed the size of a block? How do
+storage overhead? Can the number of pointers exceed the size of a block? How do
 we store a CTZ skip-list in our metadata pairs?
 
 To find the storage overhead, we can look at the data structure as multiple
@@ -742,8 +742,8 @@ where:
 2. popcount(![x]) = the number of bits that are 1 in ![x]
 
 Initial tests of this surprising property seem to hold. As ![n] approaches
-infinity, we end up with an average overhead of 2 pointers, which matches what
-our assumption from earlier. During iteration, the popcount function seems to
+infinity, we end up with an average overhead of 2 pointers, which matches our
+assumption from earlier. During iteration, the popcount function seems to
 handle deviations from this average. Of course, just to make sure I wrote a
 quick script that verified this property for all 32-bit integers.
 
@@ -767,7 +767,7 @@ overflow, but we can avoid this by rearranging the equation a bit:
 
 ![off = N - (B-2w/8)n - (w/8)popcount(n)][ctz-formula7]
 
-Our solution requires quite a bit of math, but computer are very good at math.
+Our solution requires quite a bit of math, but computers are very good at math.
 Now we can find both our block index and offset from a size in _O(1)_, letting
 us store CTZ skip-lists with only a pointer and size.
 
@@ -850,7 +850,7 @@ nearly every write to the filesystem.
 
 Normally, block allocation involves some sort of free list or bitmap stored on
 the filesystem that is updated with free blocks. However, with power
-resilience, keeping these structure consistent becomes difficult. It doesn't
+resilience, keeping these structures consistent becomes difficult. It doesn't
 help that any mistake in updating these structures can result in lost blocks
 that are impossible to recover.
 
@@ -894,9 +894,9 @@ high-risk error conditions.
 ---
 
 Our block allocator needs to find free blocks efficiently. You could traverse
-through every block on storage and check each one against our filesystem tree,
-however the runtime would be abhorrent. We need to somehow collect multiple
-blocks each traversal.
+through every block on storage and check each one against our filesystem tree;
+however, the runtime would be abhorrent. We need to somehow collect multiple
+blocks per traversal.
 
 Looking at existing designs, some larger filesystems that use a similar "drop
 it on the floor" strategy store a bitmap of the entire storage in [RAM]. This
@@ -920,8 +920,8 @@ a brute force traversal. Instead of a bitmap the size of storage, we keep track
 of a small, fixed-size bitmap called the lookahead buffer. During block
 allocation, we take blocks from the lookahead buffer. If the lookahead buffer
 is empty, we scan the filesystem for more free blocks, populating our lookahead
-buffer. Each scan we use an increasing offset, circling the storage as blocks
-are allocated.
+buffer. In each scan we use an increasing offset, circling the storage as
+blocks are allocated.
 
 Here's what it might look like to allocate 4 blocks on a decently busy
 filesystem with a 32 bit lookahead and a total of 128 blocks (512 KiB
@@ -950,7 +950,7 @@ alloc = 112     lookahead:                         ffff8000
 ```
 
 This lookahead approach has a runtime complexity of _O(n²)_ to completely
-scan storage, however, bitmaps are surprisingly compact, and in practice only
+scan storage; however, bitmaps are surprisingly compact, and in practice only
 one or two passes are usually needed to find free blocks. Additionally, the
 performance of the allocator can be optimized by adjusting the block size or
 size of the lookahead buffer, trading either write granularity or RAM for
@@ -1173,9 +1173,9 @@ We may find that the new block is also bad, but hopefully after repeating this
 cycle we'll eventually find a new block where a write succeeds. If we don't,
 that means that all blocks in our storage are bad, and we've reached the end of
 our device's usable life. At this point, littlefs will return an "out of space"
-error, which is technically true, there are no more good blocks, but as an
-added benefit also matches the error condition expected by users of dynamically
-sized data.
+error. This is technically true, as there are no more good blocks, but as an
+added benefit it also matches the error condition expected by users of
+dynamically sized data.
 
 ---
 
@@ -1187,7 +1187,7 @@ original data even after it has been corrupted. One such mechanism for this is
 ECC is an extension to the idea of a checksum. Where a checksum such as CRC can
 detect that an error has occurred in the data, ECC can detect and actually
 correct some amount of errors. However, there is a limit to how many errors ECC
-can detect, call the [Hamming bound][wikipedia-hamming-bound]. As the number of
+can detect: the [Hamming bound][wikipedia-hamming-bound]. As the number of
 errors approaches the Hamming bound, we may still be able to detect errors, but
 can no longer fix the data. If we've reached this point the block is
 unrecoverable.
@@ -1202,7 +1202,7 @@ chip itself.
 In littlefs, ECC is entirely optional. Read errors can instead be prevented
 proactively by wear leveling. But it's important to note that ECC can be used
 at the block device level to modestly extend the life of a device. littlefs
-respects any errors reported by the block device, allow a block device to
+respects any errors reported by the block device, allowing a block device to
 provide additional aggressive error detection.
 
 ---
@@ -1231,7 +1231,7 @@ Generally, wear leveling algorithms fall into one of two categories:
    we need to consider all blocks, including blocks that already contain data.
 
 As a tradeoff for code size and complexity, littlefs (currently) only provides
-dynamic wear leveling. This is a best efforts solution. Wear is not distributed
+dynamic wear leveling. This is a best effort solution. Wear is not distributed
 perfectly, but it is distributed among the free blocks and greatly extends the
 life of a device.
 
@@ -1378,7 +1378,7 @@ We can make several improvements. First, instead of giving each file its own
 metadata pair, we can store multiple files in a single metadata pair. One way
 to do this is to directly associate a directory with a metadata pair (or a
 linked list of metadata pairs). This makes it easy for multiple files to share
-the directory's metadata pair for logging and reduce the collective storage
+the directory's metadata pair for logging and reduces the collective storage
 overhead.
 
 The strict binding of metadata pairs and directories also gives users
@@ -1816,12 +1816,12 @@ while manipulating the directory tree (foreshadowing!).
 
 ## The move problem
 
-We have one last challenge. The move problem. Phrasing the problem is simple:
+We have one last challenge: the move problem. Phrasing the problem is simple:
 
 How do you atomically move a file between two directories?
 
 In littlefs we can atomically commit to directories, but we can't create
-an atomic commit that span multiple directories. The filesystem must go
+an atomic commit that spans multiple directories. The filesystem must go
 through a minimum of two distinct states to complete a move.
 
 To make matters worse, file moves are a common form of synchronization for
@@ -1831,13 +1831,13 @@ atomic moves right.
 So what can we do?
 
 - We definitely can't just let power-loss result in duplicated or lost files.
-  This could easily break user's code and would only reveal itself in extreme
+  This could easily break users' code and would only reveal itself in extreme
   cases. We were only able to be lazy about the threaded linked-list because
   it isn't user facing and we can handle the corner cases internally.
 
-- Some filesystems propagate COW operations up the tree until finding a common
-  parent. Unfortunately this interacts poorly with our threaded tree and brings
-  back the issue of upward propagation of wear.
+- Some filesystems propagate COW operations up the tree until a common parent
+  is found. Unfortunately this interacts poorly with our threaded tree and
+  brings back the issue of upward propagation of wear.
 
 - In a previous version of littlefs we tried to solve this problem by going
   back and forth between the source and destination, marking and unmarking the
@@ -1852,7 +1852,7 @@ introduction of a mechanism called "global state".
 ---
 
 Global state is a small set of state that can be updated from _any_ metadata
-pair. Combining global state with metadata pair's ability to update multiple
+pair. Combining global state with metadata pairs' ability to update multiple
 entries in one commit gives us a powerful tool for crafting complex atomic
 operations.
 
@@ -1910,7 +1910,7 @@ the filesystem is mounted.
 
 You may have noticed that global state is very expensive. We keep a copy in
 RAM and a delta in an unbounded number of metadata pairs. Even if we reset
-the global state to its initial value we can't easily clean up the deltas on
+the global state to its initial value, we can't easily clean up the deltas on
 disk. For this reason, it's very important that we keep the size of global
 state bounded and extremely small. But, even with a strict budget, global
 state is incredibly valuable.

Makefile

@@ -55,6 +55,7 @@ test: \
 	test_attrs \
 	test_move \
 	test_orphan \
+	test_relocations \
 	test_corrupt
 	@rm test.c
 test_%: tests/test_%.sh

README.md

@@ -53,6 +53,7 @@ const struct lfs_config cfg = {
     .block_count = 128,
     .cache_size = 16,
     .lookahead_size = 16,
+    .block_cycles = 500,
 };
 
 // entry point
@@ -109,7 +110,7 @@ directory functions, with the deviation that the allocation of filesystem
 structures must be provided by the user.
 
 All POSIX operations, such as remove and rename, are atomic, even in event
-of power-loss. Additionally, no file updates are not actually committed to
+of power-loss. Additionally, file updates are not actually committed to
 the filesystem until sync or close is called on the file.
 
 ## Other notes

emubd/lfs_emubd.c

@@ -102,6 +102,7 @@ int lfs_emubd_create(const struct lfs_config *cfg, const char *path) {
         if (res < 1) {
             err = -errno;
             LFS_TRACE("lfs_emubd_create -> %"PRId32, err);
+            fclose(f);
             return err;
         }
 
@@ -124,6 +125,7 @@ int lfs_emubd_create(const struct lfs_config *cfg, const char *path) {
         if (res < 1) {
             err = -errno;
             LFS_TRACE("lfs_emubd_create -> %"PRId32, err);
+            fclose(f);
             return err;
         }
 
@@ -178,6 +180,7 @@ int lfs_emubd_read(const struct lfs_config *cfg, lfs_block_t block,
         if (err) {
             err = -errno;
             LFS_TRACE("lfs_emubd_read -> %d", err);
+            fclose(f);
             return err;
         }
 
@@ -185,6 +188,7 @@ int lfs_emubd_read(const struct lfs_config *cfg, lfs_block_t block,
         if (res < size && !feof(f)) {
             err = -errno;
             LFS_TRACE("lfs_emubd_read -> %d", err);
+            fclose(f);
             return err;
         }
 
@@ -230,6 +234,7 @@ int lfs_emubd_prog(const struct lfs_config *cfg, lfs_block_t block,
     if (err) {
         err = -errno;
         LFS_TRACE("lfs_emubd_prog -> %d", err);
+        fclose(f);
         return err;
     }
 
@@ -237,6 +242,7 @@ int lfs_emubd_prog(const struct lfs_config *cfg, lfs_block_t block,
     if (res < size) {
         err = -errno;
         LFS_TRACE("lfs_emubd_prog -> %d", err);
+        fclose(f);
         return err;
     }
 
@@ -244,6 +250,7 @@ int lfs_emubd_prog(const struct lfs_config *cfg, lfs_block_t block,
     if (err) {
         err = -errno;
         LFS_TRACE("lfs_emubd_prog -> %d", err);
+        fclose(f);
         return err;
     }
 
@@ -252,6 +259,7 @@ int lfs_emubd_prog(const struct lfs_config *cfg, lfs_block_t block,
     if (res < 1) {
         err = -errno;
         LFS_TRACE("lfs_emubd_prog -> %d", err);
+        fclose(f);
         return err;
     }
 
@@ -340,6 +348,7 @@ int lfs_emubd_sync(const struct lfs_config *cfg) {
     if (res < 1) {
         int err = -errno;
         LFS_TRACE("lfs_emubd_sync -> %d", err);
+        fclose(f);
         return err;
     }
 
@@ -364,6 +373,7 @@ int lfs_emubd_sync(const struct lfs_config *cfg) {
     if (res < 1) {
         err = -errno;
         LFS_TRACE("lfs_emubd_sync -> %d", err);
+        fclose(f);
         return err;
     }
 
@@ -388,6 +398,7 @@ int lfs_emubd_sync(const struct lfs_config *cfg) {
     if (res < 1) {
         err = -errno;
         LFS_TRACE("lfs_emubd_sync -> %d", err);
+        fclose(f);
         return err;
     }
 

lfs.c

@@ -352,7 +352,7 @@ static inline bool lfs_gstate_hasmovehere(const struct lfs_gstate *a,
 
 static inline void lfs_gstate_xororphans(struct lfs_gstate *a,
         const struct lfs_gstate *b, bool orphans) {
-    a->tag ^= LFS_MKTAG(0x800, 0, 0) & (b->tag ^ (orphans << 31));
+    a->tag ^= LFS_MKTAG(0x800, 0, 0) & (b->tag ^ ((uint32_t)orphans << 31));
 }
 
 static inline void lfs_gstate_xormove(struct lfs_gstate *a,
@@ -846,7 +846,7 @@ static lfs_stag_t lfs_dir_fetchmatch(lfs_t *lfs,
                 }
 
                 // reset the next bit if we need to
-                ptag ^= (lfs_tag_chunk(tag) & 1U) << 31;
+                ptag ^= (lfs_tag_t)(lfs_tag_chunk(tag) & 1U) << 31;
 
                 // toss our crc into the filesystem seed for
                 // pseudorandom numbers
@@ -929,6 +929,11 @@ static lfs_stag_t lfs_dir_fetchmatch(lfs_t *lfs,
                 if (res == LFS_CMP_EQ) {
                     // found a match
                     tempbesttag = tag;
+                } else if ((LFS_MKTAG(0x7ff, 0x3ff, 0) & tag) ==
+                        (LFS_MKTAG(0x7ff, 0x3ff, 0) & tempbesttag)) {
+                    // found an identical tag, but contents didn't match
+                    // this must mean that our besttag has been overwritten
+                    tempbesttag = -1;
                 } else if (res == LFS_CMP_GT &&
                         lfs_tag_id(tag) <= lfs_tag_id(tempbesttag)) {
                     // found a greater match, keep track to keep things sorted
@@ -975,9 +980,10 @@ static lfs_stag_t lfs_dir_fetchmatch(lfs_t *lfs,
 
 static int lfs_dir_fetch(lfs_t *lfs,
         lfs_mdir_t *dir, const lfs_block_t pair[2]) {
-    // note, mask=-1, tag=0 can never match a tag since this
+    // note, mask=-1, tag=-1 can never match a tag since this
     // pattern has the invalid bit set
-    return lfs_dir_fetchmatch(lfs, dir, pair, -1, 0, NULL, NULL, NULL);
+    return (int)lfs_dir_fetchmatch(lfs, dir, pair,
+            (lfs_tag_t)-1, (lfs_tag_t)-1, NULL, NULL, NULL);
 }
 
 static int lfs_dir_getgstate(lfs_t *lfs, const lfs_mdir_t *dir,
@@ -1010,7 +1016,7 @@ static int lfs_dir_getinfo(lfs_t *lfs, lfs_mdir_t *dir,
     lfs_stag_t tag = lfs_dir_get(lfs, dir, LFS_MKTAG(0x780, 0x3ff, 0),
             LFS_MKTAG(LFS_TYPE_NAME, id, lfs->name_max+1), info->name);
     if (tag < 0) {
-        return tag;
+        return (int)tag;
     }
 
     info->type = lfs_tag_type3(tag);
@@ -1019,7 +1025,7 @@ static int lfs_dir_getinfo(lfs_t *lfs, lfs_mdir_t *dir,
     tag = lfs_dir_get(lfs, dir, LFS_MKTAG(0x700, 0x3ff, 0),
             LFS_MKTAG(LFS_TYPE_STRUCT, id, sizeof(ctz)), &ctz);
     if (tag < 0) {
-        return tag;
+        return (int)tag;
     }
     lfs_ctz_fromle32(&ctz);
 
@@ -1062,7 +1068,7 @@ static int lfs_dir_find_match(void *data,
     return LFS_CMP_EQ;
 }
 
-static int lfs_dir_find(lfs_t *lfs, lfs_mdir_t *dir,
+static lfs_stag_t lfs_dir_find(lfs_t *lfs, lfs_mdir_t *dir,
         const char **path, uint16_t *id) {
     // we reduce path to a single name if we can find it
     const char *name = *path;
@@ -1275,7 +1281,7 @@ static int lfs_dir_commitcrc(lfs_t *lfs, struct lfs_commit *commit) {
         }
 
         commit->off += sizeof(tag)+lfs_tag_size(tag);
-        commit->ptag = tag ^ (reset << 31);
+        commit->ptag = tag ^ ((lfs_tag_t)reset << 31);
         commit->crc = LFS_BLOCK_NULL; // reset crc for next "commit"
     }
 
@@ -1377,6 +1383,7 @@ static int lfs_dir_split(lfs_t *lfs,
         lfs_mdir_t *dir, const struct lfs_mattr *attrs, int attrcount,
         lfs_mdir_t *source, uint16_t split, uint16_t end) {
     // create tail directory
+    lfs_alloc_ack(lfs);
     lfs_mdir_t tail;
     int err = lfs_dir_alloc(lfs, &tail);
     if (err) {
@@ -1503,9 +1510,13 @@ static int lfs_dir_compact(lfs_t *lfs,
                 }
             }
 #ifdef LFS_MIGRATE
-        } else if (lfs_pair_cmp(dir->pair, lfs->root) == 0 && lfs->lfs1) {
-            // we can't relocate our root during migrations, as this would
-            // cause the superblock to get updated, which would clobber v1
+        } else if (lfs->lfs1) {
+            // do not proactively relocate blocks during migrations, this
+            // can cause a number of failure states such: clobbering the
+            // v1 superblock if we relocate root, and invalidating directory
+            // pointers if we relocate the head of a directory. On top of
+            // this, relocations increase the overall complexity of
+            // lfs_migration, which is already a delicate operation.
 #endif
         } else {
             // we're writing too much, time to relocate
@@ -1643,7 +1654,6 @@ relocate:
         if (err && (err != LFS_ERR_NOSPC && !exhausted)) {
             return err;
         }
-
         continue;
     }
 
@@ -1936,7 +1946,7 @@ int lfs_dir_open(lfs_t *lfs, lfs_dir_t *dir, const char *path) {
     LFS_TRACE("lfs_dir_open(%p, %p, \"%s\")", (void*)lfs, (void*)dir, path);
     lfs_stag_t tag = lfs_dir_find(lfs, &dir->m, &path, NULL);
     if (tag < 0) {
-        LFS_TRACE("lfs_dir_open -> %d", tag);
+        LFS_TRACE("lfs_dir_open -> %"PRId32, tag);
         return tag;
     }
 
@@ -1955,7 +1965,7 @@ int lfs_dir_open(lfs_t *lfs, lfs_dir_t *dir, const char *path) {
         lfs_stag_t res = lfs_dir_get(lfs, &dir->m, LFS_MKTAG(0x700, 0x3ff, 0),
                 LFS_MKTAG(LFS_TYPE_STRUCT, lfs_tag_id(tag), 8), pair);
         if (res < 0) {
-            LFS_TRACE("lfs_dir_open -> %d", res);
+            LFS_TRACE("lfs_dir_open -> %"PRId32, res);
             return res;
         }
         lfs_pair_fromle32(pair);
@@ -2064,10 +2074,14 @@ int lfs_dir_seek(lfs_t *lfs, lfs_dir_t *dir, lfs_off_t off) {
     dir->pos = lfs_min(2, off);
     off -= dir->pos;
 
-    while (off != 0) {
-        dir->id = lfs_min(dir->m.count, off);
-        dir->pos += dir->id;
-        off -= dir->id;
+    // skip superblock entry
+    dir->id = (off > 0 && lfs_pair_cmp(dir->head, lfs->root) == 0);
+
+    while (off > 0) {
+        int diff = lfs_min(dir->m.count - dir->id, off);
+        dir->id += diff;
+        dir->pos += diff;
+        off -= diff;
 
         if (dir->id == dir->m.count) {
             if (!dir->m.split) {
@@ -2080,6 +2094,8 @@ int lfs_dir_seek(lfs_t *lfs, lfs_dir_t *dir, lfs_off_t off) {
                 LFS_TRACE("lfs_dir_seek -> %d", err);
                 return err;
             }
+
+            dir->id = 0;
         }
     }
 
@@ -2103,8 +2119,6 @@ int lfs_dir_rewind(lfs_t *lfs, lfs_dir_t *dir) {
         return err;
     }
 
-    dir->m.pair[0] = dir->head[0];
-    dir->m.pair[1] = dir->head[1];
     dir->id = 0;
     dir->pos = 0;
     LFS_TRACE("lfs_dir_rewind -> %d", 0);
@@ -2738,14 +2752,14 @@ lfs_ssize_t lfs_file_read(lfs_t *lfs, lfs_file_t *file,
         // flush out any writes
         int err = lfs_file_flush(lfs, file);
         if (err) {
-            LFS_TRACE("lfs_file_read -> %"PRId32, err);
+            LFS_TRACE("lfs_file_read -> %d", err);
             return err;
         }
     }
 
     if (file->pos >= file->ctz.size) {
         // eof if past end
-        LFS_TRACE("lfs_file_read -> %"PRId32, 0);
+        LFS_TRACE("lfs_file_read -> %d", 0);
         return 0;
     }
 
@@ -2761,7 +2775,7 @@ lfs_ssize_t lfs_file_read(lfs_t *lfs, lfs_file_t *file,
                         file->ctz.head, file->ctz.size,
                         file->pos, &file->block, &file->off);
                 if (err) {
-                    LFS_TRACE("lfs_file_read -> %"PRId32, err);
+                    LFS_TRACE("lfs_file_read -> %d", err);
                     return err;
                 }
             } else {
@@ -2781,7 +2795,7 @@ lfs_ssize_t lfs_file_read(lfs_t *lfs, lfs_file_t *file,
                     LFS_MKTAG(LFS_TYPE_INLINESTRUCT, file->id, 0),
                     file->off, data, diff);
             if (err) {
-                LFS_TRACE("lfs_file_read -> %"PRId32, err);
+                LFS_TRACE("lfs_file_read -> %d", err);
                 return err;
             }
         } else {
@@ -2789,7 +2803,7 @@ lfs_ssize_t lfs_file_read(lfs_t *lfs, lfs_file_t *file,
                     NULL, &file->cache, lfs->cfg->block_size,
                     file->block, file->off, data, diff);
             if (err) {
-                LFS_TRACE("lfs_file_read -> %"PRId32, err);
+                LFS_TRACE("lfs_file_read -> %d", err);
                 return err;
             }
         }
@@ -2818,7 +2832,7 @@ lfs_ssize_t lfs_file_write(lfs_t *lfs, lfs_file_t *file,
         // drop any reads
         int err = lfs_file_flush(lfs, file);
         if (err) {
-            LFS_TRACE("lfs_file_write -> %"PRId32, err);
+            LFS_TRACE("lfs_file_write -> %d", err);
             return err;
         }
     }
@@ -2829,7 +2843,7 @@ lfs_ssize_t lfs_file_write(lfs_t *lfs, lfs_file_t *file,
 
     if (file->pos + size > lfs->file_max) {
         // Larger than file limit?
-        LFS_TRACE("lfs_file_write -> %"PRId32, LFS_ERR_FBIG);
+        LFS_TRACE("lfs_file_write -> %d", LFS_ERR_FBIG);
         return LFS_ERR_FBIG;
     }
 
@@ -2855,7 +2869,7 @@ lfs_ssize_t lfs_file_write(lfs_t *lfs, lfs_file_t *file,
         int err = lfs_file_outline(lfs, file);
         if (err) {
             file->flags |= LFS_F_ERRED;
-            LFS_TRACE("lfs_file_write -> %"PRId32, err);
+            LFS_TRACE("lfs_file_write -> %d", err);
             return err;
         }
     }
@@ -2872,7 +2886,7 @@ lfs_ssize_t lfs_file_write(lfs_t *lfs, lfs_file_t *file,
                             file->pos-1, &file->block, &file->off);
                     if (err) {
                         file->flags |= LFS_F_ERRED;
-                        LFS_TRACE("lfs_file_write -> %"PRId32, err);
+                        LFS_TRACE("lfs_file_write -> %d", err);
                         return err;
                     }
 
@@ -2887,7 +2901,7 @@ lfs_ssize_t lfs_file_write(lfs_t *lfs, lfs_file_t *file,
                         &file->block, &file->off);
                 if (err) {
                     file->flags |= LFS_F_ERRED;
-                    LFS_TRACE("lfs_file_write -> %"PRId32, err);
+                    LFS_TRACE("lfs_file_write -> %d", err);
                     return err;
                 }
             } else {
@@ -2908,7 +2922,7 @@ lfs_ssize_t lfs_file_write(lfs_t *lfs, lfs_file_t *file,
                     goto relocate;
                 }
                 file->flags |= LFS_F_ERRED;
-                LFS_TRACE("lfs_file_write -> %"PRId32, err);
+                LFS_TRACE("lfs_file_write -> %d", err);
                 return err;
             }
 
@@ -2917,7 +2931,7 @@ relocate:
             err = lfs_file_relocate(lfs, file);
             if (err) {
                 file->flags |= LFS_F_ERRED;
-                LFS_TRACE("lfs_file_write -> %"PRId32, err);
+                LFS_TRACE("lfs_file_write -> %d", err);
                 return err;
             }
         }
@@ -2944,7 +2958,7 @@ lfs_soff_t lfs_file_seek(lfs_t *lfs, lfs_file_t *file,
     // write out everything beforehand, may be noop if rdonly
     int err = lfs_file_flush(lfs, file);
     if (err) {
-        LFS_TRACE("lfs_file_seek -> %"PRId32, err);
+        LFS_TRACE("lfs_file_seek -> %d", err);
         return err;
     }
 
@@ -2960,7 +2974,7 @@ lfs_soff_t lfs_file_seek(lfs_t *lfs, lfs_file_t *file,
 
     if (npos > lfs->file_max) {
         // file position out of range
-        LFS_TRACE("lfs_file_seek -> %"PRId32, LFS_ERR_INVAL);
+        LFS_TRACE("lfs_file_seek -> %d", LFS_ERR_INVAL);
         return LFS_ERR_INVAL;
     }
 
@@ -3006,10 +3020,10 @@ int lfs_file_truncate(lfs_t *lfs, lfs_file_t *file, lfs_off_t size) {
     } else if (size > oldsize) {
         // flush+seek if not already at end
         if (file->pos != oldsize) {
-            int err = lfs_file_seek(lfs, file, 0, LFS_SEEK_END);
-            if (err < 0) {
-                LFS_TRACE("lfs_file_truncate -> %d", err);
-                return err;
+            lfs_soff_t res = lfs_file_seek(lfs, file, 0, LFS_SEEK_END);
+            if (res < 0) {
+                LFS_TRACE("lfs_file_truncate -> %"PRId32, res);
+                return (int)res;
             }
         }
 
@@ -3017,17 +3031,17 @@ int lfs_file_truncate(lfs_t *lfs, lfs_file_t *file, lfs_off_t size) {
         while (file->pos < size) {
             lfs_ssize_t res = lfs_file_write(lfs, file, &(uint8_t){0}, 1);
             if (res < 0) {
-                LFS_TRACE("lfs_file_truncate -> %d", res);
-                return res;
+                LFS_TRACE("lfs_file_truncate -> %"PRId32, res);
+                return (int)res;
             }
         }
     }
 
     // restore pos
-    int err = lfs_file_seek(lfs, file, pos, LFS_SEEK_SET);
-    if (err < 0) {
-      LFS_TRACE("lfs_file_truncate -> %d", err);
-      return err;
+    lfs_soff_t res = lfs_file_seek(lfs, file, pos, LFS_SEEK_SET);
+    if (res < 0) {
+      LFS_TRACE("lfs_file_truncate -> %"PRId32, res);
+      return (int)res;
     }
 
     LFS_TRACE("lfs_file_truncate -> %d", 0);
@@ -3046,8 +3060,8 @@ int lfs_file_rewind(lfs_t *lfs, lfs_file_t *file) {
     LFS_TRACE("lfs_file_rewind(%p, %p)", (void*)lfs, (void*)file);
     lfs_soff_t res = lfs_file_seek(lfs, file, 0, LFS_SEEK_SET);
     if (res < 0) {
-        LFS_TRACE("lfs_file_rewind -> %d", res);
-        return res;
+        LFS_TRACE("lfs_file_rewind -> %"PRId32, res);
+        return (int)res;
     }
 
     LFS_TRACE("lfs_file_rewind -> %d", 0);
@@ -3075,8 +3089,8 @@ int lfs_stat(lfs_t *lfs, const char *path, struct lfs_info *info) {
     lfs_mdir_t cwd;
     lfs_stag_t tag = lfs_dir_find(lfs, &cwd, &path, NULL);
     if (tag < 0) {
-        LFS_TRACE("lfs_stat -> %d", tag);
-        return tag;
+        LFS_TRACE("lfs_stat -> %"PRId32, tag);
+        return (int)tag;
     }
 
     int err = lfs_dir_getinfo(lfs, &cwd, lfs_tag_id(tag), info);
@@ -3096,8 +3110,8 @@ int lfs_remove(lfs_t *lfs, const char *path) {
     lfs_mdir_t cwd;
     lfs_stag_t tag = lfs_dir_find(lfs, &cwd, &path, NULL);
     if (tag < 0 || lfs_tag_id(tag) == 0x3ff) {
-        LFS_TRACE("lfs_remove -> %d", (tag < 0) ? tag : LFS_ERR_INVAL);
-        return (tag < 0) ? tag : LFS_ERR_INVAL;
+        LFS_TRACE("lfs_remove -> %"PRId32, (tag < 0) ? tag : LFS_ERR_INVAL);
+        return (tag < 0) ? (int)tag : LFS_ERR_INVAL;
     }
 
     lfs_mdir_t dir;
@@ -3107,8 +3121,8 @@ int lfs_remove(lfs_t *lfs, const char *path) {
         lfs_stag_t res = lfs_dir_get(lfs, &cwd, LFS_MKTAG(0x700, 0x3ff, 0),
                 LFS_MKTAG(LFS_TYPE_STRUCT, lfs_tag_id(tag), 8), pair);
         if (res < 0) {
-            LFS_TRACE("lfs_remove -> %d", res);
-            return res;
+            LFS_TRACE("lfs_remove -> %"PRId32, res);
+            return (int)res;
         }
         lfs_pair_fromle32(pair);
 
@@ -3170,8 +3184,8 @@ int lfs_rename(lfs_t *lfs, const char *oldpath, const char *newpath) {
     lfs_mdir_t oldcwd;
     lfs_stag_t oldtag = lfs_dir_find(lfs, &oldcwd, &oldpath, NULL);
     if (oldtag < 0 || lfs_tag_id(oldtag) == 0x3ff) {
-        LFS_TRACE("lfs_rename -> %d", (oldtag < 0) ? oldtag : LFS_ERR_INVAL);
-        return (oldtag < 0) ? oldtag : LFS_ERR_INVAL;
+        LFS_TRACE("lfs_rename -> %"PRId32, (oldtag < 0) ? oldtag : LFS_ERR_INVAL);
+        return (oldtag < 0) ? (int)oldtag : LFS_ERR_INVAL;
     }
 
     // find new entry
@@ -3180,8 +3194,8 @@ int lfs_rename(lfs_t *lfs, const char *oldpath, const char *newpath) {
     lfs_stag_t prevtag = lfs_dir_find(lfs, &newcwd, &newpath, &newid);
     if ((prevtag < 0 || lfs_tag_id(prevtag) == 0x3ff) &&
             !(prevtag == LFS_ERR_NOENT && newid != 0x3ff)) {
-        LFS_TRACE("lfs_rename -> %d", (prevtag < 0) ? prevtag : LFS_ERR_INVAL);
-        return (prevtag < 0) ? prevtag : LFS_ERR_INVAL;
+        LFS_TRACE("lfs_rename -> %"PRId32, (prevtag < 0) ? prevtag : LFS_ERR_INVAL);
+        return (prevtag < 0) ? (int)prevtag : LFS_ERR_INVAL;
     }
 
     lfs_mdir_t prevdir;
@@ -3201,8 +3215,8 @@ int lfs_rename(lfs_t *lfs, const char *oldpath, const char *newpath) {
         lfs_stag_t res = lfs_dir_get(lfs, &newcwd, LFS_MKTAG(0x700, 0x3ff, 0),
                 LFS_MKTAG(LFS_TYPE_STRUCT, newid, 8), prevpair);
         if (res < 0) {
-            LFS_TRACE("lfs_rename -> %d", res);
-            return res;
+            LFS_TRACE("lfs_rename -> %"PRId32, res);
+            return (int)res;
         }
         lfs_pair_fromle32(prevpair);
 
@@ -3296,7 +3310,7 @@ lfs_ssize_t lfs_getattr(lfs_t *lfs, const char *path,
         id = 0;
         int err = lfs_dir_fetch(lfs, &cwd, lfs->root);
         if (err) {
-            LFS_TRACE("lfs_getattr -> %"PRId32, err);
+            LFS_TRACE("lfs_getattr -> %d", err);
             return err;
         }
     }
@@ -3307,7 +3321,7 @@ lfs_ssize_t lfs_getattr(lfs_t *lfs, const char *path,
             buffer);
     if (tag < 0) {
         if (tag == LFS_ERR_NOENT) {
-            LFS_TRACE("lfs_getattr -> %"PRId32, LFS_ERR_NOATTR);
+            LFS_TRACE("lfs_getattr -> %d", LFS_ERR_NOATTR);
             return LFS_ERR_NOATTR;
         }
 
@@ -3369,6 +3383,12 @@ static int lfs_init(lfs_t *lfs, const struct lfs_config *cfg) {
     lfs->cfg = cfg;
     int err = 0;
 
+    // validate that the lfs-cfg sizes were initiated properly before
+    // performing any arithmetic logics with them
+    LFS_ASSERT(lfs->cfg->read_size != 0);
+    LFS_ASSERT(lfs->cfg->prog_size != 0);
+    LFS_ASSERT(lfs->cfg->cache_size != 0);
+
     // check that block size is a multiple of cache size is a multiple
     // of prog and read sizes
     LFS_ASSERT(lfs->cfg->cache_size % lfs->cfg->read_size == 0);
@@ -3757,7 +3777,7 @@ int lfs_fs_traverse(lfs_t *lfs,
                 if (tag == LFS_ERR_NOENT) {
                     continue;
                 }
-                LFS_TRACE("lfs_fs_traverse -> %d", tag);
+                LFS_TRACE("lfs_fs_traverse -> %"PRId32, tag);
                 return tag;
             }
             lfs_ctz_fromle32(&ctz);
@@ -3881,6 +3901,12 @@ static int lfs_fs_relocate(lfs_t *lfs,
             d->m.pair[0] = newpair[0];
             d->m.pair[1] = newpair[1];
         }
+
+        if (d->type == LFS_TYPE_DIR &&
+                lfs_pair_cmp(oldpair, ((lfs_dir_t*)d)->head) == 0) {
+            ((lfs_dir_t*)d)->head[0] = newpair[0];
+            ((lfs_dir_t*)d)->head[1] = newpair[1];
+        }
     }
 
     // find parent
@@ -4066,11 +4092,11 @@ lfs_ssize_t lfs_fs_size(lfs_t *lfs) {
     lfs_size_t size = 0;
     int err = lfs_fs_traverse(lfs, lfs_fs_size_count, &size);
     if (err) {
-        LFS_TRACE("lfs_fs_size -> %"PRId32, err);
+        LFS_TRACE("lfs_fs_size -> %d", err);
         return err;
     }
 
-    LFS_TRACE("lfs_fs_size -> %"PRId32, err);
+    LFS_TRACE("lfs_fs_size -> %d", err);
     return size;
 }
 
@@ -4608,7 +4634,7 @@ int lfs_migrate(lfs_t *lfs, const struct lfs_config *cfg) {
                             id, entry1.d.nlen), name},
                         {LFS_MKTAG(
                             isdir ? LFS_TYPE_DIRSTRUCT : LFS_TYPE_CTZSTRUCT,
-                            id, sizeof(&entry1.d.u)), &entry1.d.u}));
+                            id, sizeof(entry1.d.u)), &entry1.d.u}));
                 lfs1_entry_fromle32(&entry1.d);
                 if (err) {
                     goto cleanup;
@@ -4631,7 +4657,7 @@ int lfs_migrate(lfs_t *lfs, const struct lfs_config *cfg) {
 
                 lfs_pair_tole32(dir2.pair);
                 err = lfs_dir_commit(lfs, &dir2, LFS_MKATTRS(
-                        {LFS_MKTAG(LFS_TYPE_SOFTTAIL, 0x3ff, 0),
+                        {LFS_MKTAG(LFS_TYPE_SOFTTAIL, 0x3ff, 8),
                             dir1.d.tail}));
                 lfs_pair_fromle32(dir2.pair);
                 if (err) {

tests/test_alloc.sh

@@ -250,6 +250,14 @@ scripts/test.py << TEST
     lfs_unmount(&lfs) => 0;
 TEST
 
+## Below, these tests depend _very_ heavily on the geometry of the
+## block device being tested, they should be removed and replaced
+## by generalized tests. For now we'll just skip if the geometry
+## is customized.
+
+if [[ ! $MAKEFLAGS =~ "LFS_BLOCK_CYCLES" ]]
+then
+
 echo "--- Chained dir exhaustion test ---"
 scripts/test.py << TEST
     lfs_mount(&lfs, &cfg) => 0;
@@ -481,4 +489,6 @@ scripts/test.py << TEST
     lfs_unmount(&lfs) => 0;
 TEST
 
+fi
+
 scripts/results.py

tests/test_paths.sh

@@ -179,7 +179,7 @@ echo "--- Really big path test ---"
 scripts/test.py << TEST
     lfs_mount(&lfs, &cfg) => 0;
     memset(path, 'w', LFS_NAME_MAX);
-    path[LFS_NAME_MAX+1] = '\0';
+    path[LFS_NAME_MAX] = '\0';
     lfs_mkdir(&lfs, path) => 0;
     lfs_remove(&lfs, path) => 0;
     lfs_file_open(&lfs, &file, path,
@@ -189,7 +189,7 @@ scripts/test.py << TEST
 
     memcpy(path, "coffee/", strlen("coffee/"));
     memset(path+strlen("coffee/"), 'w', LFS_NAME_MAX);
-    path[strlen("coffee/")+LFS_NAME_MAX+1] = '\0';
+    path[strlen("coffee/")+LFS_NAME_MAX] = '\0';
     lfs_mkdir(&lfs, path) => 0;
     lfs_remove(&lfs, path) => 0;
     lfs_file_open(&lfs, &file, path,

tests/test_relocations.sh

@@ -0,0 +1,139 @@
+#!/bin/bash
+set -eu
+export TEST_FILE=$0
+trap 'export TEST_LINE=$LINENO' DEBUG
+
+ITERATIONS=20
+COUNT=10
+
+echo "=== Relocation tests ==="
+rm -rf blocks
+scripts/test.py << TEST
+    lfs_format(&lfs, &cfg) => 0;
+    // fill up filesystem so only ~16 blocks are left
+    lfs_mount(&lfs, &cfg) => 0;
+    lfs_file_open(&lfs, &file, "padding", LFS_O_CREAT | LFS_O_WRONLY) => 0;
+    memset(buffer, 0, 512);
+    while (LFS_BLOCK_COUNT - lfs_fs_size(&lfs) > 16) {
+        lfs_file_write(&lfs, &file, buffer, 512) => 512;
+    }
+    lfs_file_close(&lfs, &file) => 0;
+    // make a child dir to use in bounded space
+    lfs_mkdir(&lfs, "child") => 0;
+    lfs_unmount(&lfs) => 0;
+TEST
+
+echo "--- Dangling split dir test ---"
+scripts/test.py << TEST
+    lfs_mount(&lfs, &cfg) => 0;
+    for (int j = 0; j < $ITERATIONS; j++) {
+        for (int i = 0; i < $COUNT; i++) {
+            sprintf(path, "child/test%03d_loooooooooooooooooong_name", i);
+            lfs_file_open(&lfs, &file, path, LFS_O_CREAT | LFS_O_WRONLY) => 0;
+            lfs_file_close(&lfs, &file) => 0;
+        }
+
+        lfs_dir_open(&lfs, &dir, "child") => 0;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        for (int i = 0; i < $COUNT; i++) {
+            sprintf(path, "test%03d_loooooooooooooooooong_name", i);
+            lfs_dir_read(&lfs, &dir, &info) => 1;
+            strcmp(info.name, path) => 0;
+        }
+        lfs_dir_read(&lfs, &dir, &info) => 0;
+        lfs_dir_close(&lfs, &dir) => 0;
+
+        if (j == $ITERATIONS-1) {
+            break;
+        }
+
+        for (int i = 0; i < $COUNT; i++) {
+            sprintf(path, "child/test%03d_loooooooooooooooooong_name", i);
+            lfs_remove(&lfs, path) => 0;
+        }
+    }
+    lfs_unmount(&lfs) => 0;
+TEST
+scripts/test.py << TEST
+    lfs_mount(&lfs, &cfg) => 0;
+    lfs_dir_open(&lfs, &dir, "child") => 0;
+    lfs_dir_read(&lfs, &dir, &info) => 1;
+    lfs_dir_read(&lfs, &dir, &info) => 1;
+    for (int i = 0; i < $COUNT; i++) {
+        sprintf(path, "test%03d_loooooooooooooooooong_name", i);
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        strcmp(info.name, path) => 0;
+    }
+    lfs_dir_read(&lfs, &dir, &info) => 0;
+    lfs_dir_close(&lfs, &dir) => 0;
+    for (int i = 0; i < $COUNT; i++) {
+        sprintf(path, "child/test%03d_loooooooooooooooooong_name", i);
+        lfs_remove(&lfs, path) => 0;
+    }
+    lfs_unmount(&lfs) => 0;
+TEST
+
+echo "--- Outdated head test ---"
+scripts/test.py << TEST
+    lfs_mount(&lfs, &cfg) => 0;
+    for (int j = 0; j < $ITERATIONS; j++) {
+        for (int i = 0; i < $COUNT; i++) {
+            sprintf(path, "child/test%03d_loooooooooooooooooong_name", i);
+            lfs_file_open(&lfs, &file, path, LFS_O_CREAT | LFS_O_WRONLY) => 0;
+            lfs_file_close(&lfs, &file) => 0;
+        }
+
+        lfs_dir_open(&lfs, &dir, "child") => 0;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        for (int i = 0; i < $COUNT; i++) {
+            sprintf(path, "test%03d_loooooooooooooooooong_name", i);
+            lfs_dir_read(&lfs, &dir, &info) => 1;
+            strcmp(info.name, path) => 0;
+            info.size => 0;
+
+            sprintf(path, "child/test%03d_loooooooooooooooooong_name", i);
+            lfs_file_open(&lfs, &file, path, LFS_O_WRONLY) => 0;
+            lfs_file_write(&lfs, &file, "hi", 2) => 2;
+            lfs_file_close(&lfs, &file) => 0;
+        }
+        lfs_dir_read(&lfs, &dir, &info) => 0;
+
+        lfs_dir_rewind(&lfs, &dir) => 0;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        for (int i = 0; i < $COUNT; i++) {
+            sprintf(path, "test%03d_loooooooooooooooooong_name", i);
+            lfs_dir_read(&lfs, &dir, &info) => 1;
+            strcmp(info.name, path) => 0;
+            info.size => 2;
+
+            sprintf(path, "child/test%03d_loooooooooooooooooong_name", i);
+            lfs_file_open(&lfs, &file, path, LFS_O_WRONLY) => 0;
+            lfs_file_write(&lfs, &file, "hi", 2) => 2;
+            lfs_file_close(&lfs, &file) => 0;
+        }
+        lfs_dir_read(&lfs, &dir, &info) => 0;
+
+        lfs_dir_rewind(&lfs, &dir) => 0;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        for (int i = 0; i < $COUNT; i++) {
+            sprintf(path, "test%03d_loooooooooooooooooong_name", i);
+            lfs_dir_read(&lfs, &dir, &info) => 1;
+            strcmp(info.name, path) => 0;
+            info.size => 2;
+        }
+        lfs_dir_read(&lfs, &dir, &info) => 0;
+        lfs_dir_close(&lfs, &dir) => 0;
+
+        for (int i = 0; i < $COUNT; i++) {
+            sprintf(path, "child/test%03d_loooooooooooooooooong_name", i);
+            lfs_remove(&lfs, path) => 0;
+        }
+    }
+    lfs_unmount(&lfs) => 0;
+TEST
+
+scripts/results.py

tests/test_seek.sh

@@ -428,4 +428,78 @@ scripts/test.py << TEST
 TEST
 done
 
+echo "--- Root seek test ---"
+./scripts/test.py << TEST
+    lfs_mount(&lfs, &cfg) => 0;
+    for (int i = 3; i < $MEDIUMSIZE; i++) {
+        sprintf(path, "hi%03d", i);
+        lfs_mkdir(&lfs, path) => 0;
+    }
+
+    lfs_dir_open(&lfs, &dir, "/") => 0;
+    for (int i = 0; i < $MEDIUMSIZE; i++) {
+        if (i == 0) {
+            sprintf(path, ".");
+        } else if (i == 1) {
+            sprintf(path, "..");
+        } else if (i == 2) {
+            sprintf(path, "hello");
+        } else {
+            sprintf(path, "hi%03d", i);
+        }
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        strcmp(path, info.name) => 0;
+    }
+    lfs_dir_read(&lfs, &dir, &info) => 0;
+    lfs_dir_close(&lfs, &dir) => 0;
+
+    for (int j = 0; j < $MEDIUMSIZE; j++) {
+        lfs_soff_t off = -1;
+
+        lfs_dir_open(&lfs, &dir, "/") => 0;
+        for (int i = 0; i < $MEDIUMSIZE; i++) {
+            if (i == 0) {
+                sprintf(path, ".");
+            } else if (i == 1) {
+                sprintf(path, "..");
+            } else if (i == 2) {
+                sprintf(path, "hello");
+            } else {
+                sprintf(path, "hi%03d", i);
+            }
+
+            if (i == j) {
+                off = lfs_dir_tell(&lfs, &dir);
+                off >= 0 => true;
+            }
+
+            lfs_dir_read(&lfs, &dir, &info) => 1;
+            strcmp(path, info.name) => 0;
+        }
+        lfs_dir_read(&lfs, &dir, &info) => 0;
+        lfs_dir_close(&lfs, &dir) => 0;
+
+        lfs_dir_open(&lfs, &dir, "/") => 0;
+        lfs_dir_seek(&lfs, &dir, off) => 0;
+        for (int i = j; i < $MEDIUMSIZE; i++) {
+            if (i == 0) {
+                sprintf(path, ".");
+            } else if (i == 1) {
+                sprintf(path, "..");
+            } else if (i == 2) {
+                sprintf(path, "hello");
+            } else {
+                sprintf(path, "hi%03d", i);
+            }
+
+            lfs_dir_read(&lfs, &dir, &info) => 1;
+            strcmp(path, info.name) => 0;
+        }
+        lfs_dir_read(&lfs, &dir, &info) => 0;
+        lfs_dir_close(&lfs, &dir) => 0;
+    }
+
+    lfs_unmount(&lfs) => 0;
+TEST
+
 scripts/results.py