Extract `buzhash` logic into a dedicated module under `chunkers`

Moved `buzhash` implementation from `chunker` to a new `buzhash` module for better separation of concerns. Updated imports, adjusted `setup.py` and build configuration accordingly. Removed deprecated `Chunker` definitions from `chunker.pyi`.

.gitignore

@@ -6,6 +6,7 @@ dist
 src/borg/compress.c
 src/borg/crypto/low_level.c
 src/borg/item.c
+src/borg/chunkers/buzhash.c
 src/borg/chunkers/chunker.c
 src/borg/chunkers/reader.c
 src/borg/checksums.c

scripts/make.py

@@ -542,6 +542,7 @@ class BuildMan:
 cython_sources = """
 src/borg/compress.pyx
 src/borg/crypto/low_level.pyx
+src/borg/chunkers/buzhash.pyx
 src/borg/chunkers/chunker.pyx
 src/borg/chunkers/reader.pyx
 src/borg/hashindex.pyx

setup.py

@@ -50,6 +50,7 @@ cflags = ["-Wall", "-Wextra", "-Wpointer-arith", "-Wno-unreachable-code-fallthro
 
 compress_source = "src/borg/compress.pyx"
 crypto_ll_source = "src/borg/crypto/low_level.pyx"
+buzhash_source = "src/borg/chunkers/buzhash.pyx"
 chunker_source = "src/borg/chunkers/chunker.pyx"
 reader_source = "src/borg/chunkers/reader.pyx"
 hashindex_source = "src/borg/hashindex.pyx"
@@ -65,6 +66,7 @@ platform_windows_source = "src/borg/platform/windows.pyx"
 cython_sources = [
     compress_source,
     crypto_ll_source,
+    buzhash_source,
     chunker_source,
     reader_source,
     hashindex_source,
@@ -184,6 +186,7 @@ if not on_rtd:
         Extension("borg.compress", **compress_ext_kwargs),
         Extension("borg.hashindex", [hashindex_source], extra_compile_args=cflags),
         Extension("borg.item", [item_source], extra_compile_args=cflags),
+        Extension("borg.chunkers.buzhash", [buzhash_source], extra_compile_args=cflags, undef_macros=["NDEBUG"]),
         Extension("borg.chunkers.chunker", [chunker_source], extra_compile_args=cflags, undef_macros=["NDEBUG"]),
         Extension("borg.chunkers.reader", [reader_source], extra_compile_args=cflags, undef_macros=["NDEBUG"]),
         Extension("borg.checksums", **checksums_ext_kwargs),

src/borg/chunkers/__init__.py

@@ -1,3 +1,4 @@
+from .buzhash import *  # noqa
 from .chunker import *  # noqa
 from .reader import *  # noqa
 

src/borg/chunkers/buzhash.pyi

@@ -0,0 +1,20 @@
+from typing import List, Iterator, BinaryIO
+
+from .reader import fmap_entry
+
+API_VERSION: str
+
+def buzhash(data: bytes, seed: int) -> int: ...
+def buzhash_update(sum: int, remove: int, add: int, len: int, seed: int) -> int: ...
+
+class Chunker:
+    def __init__(
+        self,
+        seed: int,
+        chunk_min_exp: int,
+        chunk_max_exp: int,
+        hash_mask_bits: int,
+        hash_window_size: int,
+        sparse: bool = False,
+    ) -> None: ...
+    def chunkify(self, fd: BinaryIO = None, fh: int = -1, fmap: List[fmap_entry] = None) -> Iterator: ...

src/borg/chunkers/buzhash.pyx

@@ -0,0 +1,332 @@
+# cython: language_level=3
+
+API_VERSION = '1.2_01'
+
+import cython
+import time
+from cpython.bytes cimport PyBytes_AsString
+from libc.stdint cimport uint8_t, uint32_t
+from libc.stdlib cimport malloc, free
+from libc.string cimport memcpy, memmove
+
+from ..constants import CH_DATA, CH_ALLOC, CH_HOLE, zeros
+from .reader import FileReader, Chunk
+
+# Cyclic polynomial / buzhash
+#
+# https://en.wikipedia.org/wiki/Rolling_hash
+#
+# http://www.serve.net/buz/Notes.1st.year/HTML/C6/rand.012.html (by "BUZ", the inventor)
+#
+# http://www.dcs.gla.ac.uk/~hamer/cakes-talk.pdf (see buzhash slide)
+#
+# Some properties of buzhash / of this implementation:
+#
+# (1) the hash is designed for inputs <= 32 bytes, but the chunker uses it on a 4095 byte window;
+#     any repeating bytes at distance 32 within those 4095 bytes can cause cancellation within
+#     the hash function, e.g. in "X <any 31 bytes> X", the last X would cancel out the influence
+#     of the first X on the hash value.
+#
+# (2) the hash table is supposed to have (according to the BUZ) exactly a 50% distribution of
+#     0/1 bit values per position, but the hard coded table below doesn't fit that property.
+#
+# (3) if you would use a window size divisible by 64, the seed would cancel itself out completely.
+#     this is why we use a window size of 4095 bytes.
+#
+# Another quirk is that, even with the 4095 byte window, XORing the entire table by a constant
+# is equivalent to XORing the hash output with a different constant. but since the seed is stored
+# encrypted, i think it still serves its purpose.
+
+cdef uint32_t table_base[256]
+table_base = [
+    0xe7f831ec, 0xf4026465, 0xafb50cae, 0x6d553c7a, 0xd639efe3, 0x19a7b895, 0x9aba5b21, 0x5417d6d4,
+    0x35fd2b84, 0xd1f6a159, 0x3f8e323f, 0xb419551c, 0xf444cebf, 0x21dc3b80, 0xde8d1e36, 0x84a32436,
+    0xbeb35a9d, 0xa36f24aa, 0xa4e60186, 0x98d18ffe, 0x3f042f9e, 0xdb228bcd, 0x096474b7, 0x5c20c2f7,
+    0xf9eec872, 0xe8625275, 0xb9d38f80, 0xd48eb716, 0x22a950b4, 0x3cbaaeaa, 0xc37cddd3, 0x8fea6f6a,
+    0x1d55d526, 0x7fd6d3b3, 0xdaa072ee, 0x4345ac40, 0xa077c642, 0x8f2bd45b, 0x28509110, 0x55557613,
+    0xffc17311, 0xd961ffef, 0xe532c287, 0xaab95937, 0x46d38365, 0xb065c703, 0xf2d91d0f, 0x92cd4bb0,
+    0x4007c712, 0xf35509dd, 0x505b2f69, 0x557ead81, 0x310f4563, 0xbddc5be8, 0x9760f38c, 0x701e0205,
+    0x00157244, 0x14912826, 0xdc4ca32b, 0x67b196de, 0x5db292e8, 0x8c1b406b, 0x01f34075, 0xfa2520f7,
+    0x73bc37ab, 0x1e18bc30, 0xfe2c6cb3, 0x20c522d0, 0x5639e3db, 0x942bda35, 0x899af9d1, 0xced44035,
+    0x98cc025b, 0x255f5771, 0x70fefa24, 0xe928fa4d, 0x2c030405, 0xb9325590, 0x20cb63bd, 0xa166305d,
+    0x80e52c0a, 0xa8fafe2f, 0x1ad13f7d, 0xcfaf3685, 0x6c83a199, 0x7d26718a, 0xde5dfcd9, 0x79cf7355,
+    0x8979d7fb, 0xebf8c55e, 0xebe408e4, 0xcd2affba, 0xe483be6e, 0xe239d6de, 0x5dc1e9e0, 0x0473931f,
+    0x851b097c, 0xac5db249, 0x09c0f9f2, 0xd8d2f134, 0xe6f38e41, 0xb1c71bf1, 0x52b6e4db, 0x07224424,
+    0x6cf73e85, 0x4f25d89c, 0x782a7d74, 0x10a68dcd, 0x3a868189, 0xd570d2dc, 0x69630745, 0x9542ed86,
+    0x331cd6b2, 0xa84b5b28, 0x07879c9d, 0x38372f64, 0x7185db11, 0x25ba7c83, 0x01061523, 0xe6792f9f,
+    0xe5df07d1, 0x4321b47f, 0x7d2469d8, 0x1a3a4f90, 0x48be29a3, 0x669071af, 0x8ec8dd31, 0x0810bfbf,
+    0x813a06b4, 0x68538345, 0x65865ddc, 0x43a71b8e, 0x78619a56, 0x5a34451d, 0x5bdaa3ed, 0x71edc7e9,
+    0x17ac9a20, 0x78d10bfa, 0x6c1e7f35, 0xd51839d9, 0x240cbc51, 0x33513cc1, 0xd2b4f795, 0xccaa8186,
+    0x0babe682, 0xa33cf164, 0x18c643ea, 0xc1ca105f, 0x9959147a, 0x6d3d94de, 0x0b654fbe, 0xed902ca0,
+    0x7d835cb5, 0x99ba1509, 0x6445c922, 0x495e76c2, 0xf07194bc, 0xa1631d7e, 0x677076a5, 0x89fffe35,
+    0x1a49bcf3, 0x8e6c948a, 0x0144c917, 0x8d93aea1, 0x16f87ddf, 0xc8f25d49, 0x1fb11297, 0x27e750cd,
+    0x2f422da1, 0xdee89a77, 0x1534c643, 0x457b7b8b, 0xaf172f7a, 0x6b9b09d6, 0x33573f7f, 0xf14e15c4,
+    0x526467d5, 0xaf488241, 0x87c3ee0d, 0x33be490c, 0x95aa6e52, 0x43ec242e, 0xd77de99b, 0xd018334f,
+    0x5b78d407, 0x498eb66b, 0xb1279fa8, 0xb38b0ea6, 0x90718376, 0xe325dee2, 0x8e2f2cba, 0xcaa5bdec,
+    0x9d652c56, 0xad68f5cb, 0xa77591af, 0x88e37ee8, 0xf8faa221, 0xfcbbbe47, 0x4f407786, 0xaf393889,
+    0xf444a1d9, 0x15ae1a2f, 0x40aa7097, 0x6f9486ac, 0x29d232a3, 0xe47609e9, 0xe8b631ff, 0xba8565f4,
+    0x11288749, 0x46c9a838, 0xeb1b7cd8, 0xf516bbb1, 0xfb74fda0, 0x010996e6, 0x4c994653, 0x1d889512,
+    0x53dcd9a3, 0xdd074697, 0x1e78e17c, 0x637c98bf, 0x930bb219, 0xcf7f75b0, 0xcb9355fb, 0x9e623009,
+    0xe466d82c, 0x28f968d3, 0xfeb385d9, 0x238e026c, 0xb8ed0560, 0x0c6a027a, 0x3d6fec4b, 0xbb4b2ec2,
+    0xe715031c, 0xeded011d, 0xcdc4d3b9, 0xc456fc96, 0xdd0eea20, 0xb3df8ec9, 0x12351993, 0xd9cbb01c,
+    0x603147a2, 0xcf37d17d, 0xf7fcd9dc, 0xd8556fa3, 0x104c8131, 0x13152774, 0xb4715811, 0x6a72c2c9,
+    0xc5ae37bb, 0xa76ce12a, 0x8150d8f3, 0x2ec29218, 0xa35f0984, 0x48c0647e, 0x0b5ff98c, 0x71893f7b
+]
+
+# This seems to be the most reliable way to inline this code, using a C preprocessor macro:
+cdef extern from *:
+   """
+   #define BARREL_SHIFT(v, shift) (((v) << (shift)) | ((v) >> (((32 - (shift)) & 0x1f))))
+   """
+   uint32_t BARREL_SHIFT(uint32_t v, uint32_t shift)
+
+
+@cython.boundscheck(False)  # Deactivate bounds checking
+@cython.wraparound(False)  # Deactivate negative indexing.
+cdef uint32_t* buzhash_init_table(uint32_t seed):
+    """Initialize the buzhash table with the given seed."""
+    cdef int i
+    cdef uint32_t* table = <uint32_t*>malloc(1024)  # 256 * sizeof(uint32_t)
+    for i in range(256):
+        table[i] = table_base[i] ^ seed
+    return table
+
+
+@cython.boundscheck(False)  # Deactivate bounds checking
+@cython.wraparound(False)  # Deactivate negative indexing.
+@cython.cdivision(True)  # Use C division/modulo semantics for integer division.
+cdef uint32_t _buzhash(const unsigned char* data, size_t len, const uint32_t* h):
+    """Calculate the buzhash of the given data."""
+    cdef uint32_t i
+    cdef uint32_t sum = 0, imod
+    for i in range(len - 1, 0, -1):
+        imod = i & 0x1f
+        sum ^= BARREL_SHIFT(h[data[0]], imod)
+        data += 1
+    return sum ^ h[data[0]]
+
+
+@cython.boundscheck(False)  # Deactivate bounds checking
+@cython.wraparound(False)  # Deactivate negative indexing.
+@cython.cdivision(True)  # Use C division/modulo semantics for integer division.
+cdef uint32_t _buzhash_update(uint32_t sum, unsigned char remove, unsigned char add, size_t len, const uint32_t* h):
+    """Update the buzhash with a new byte."""
+    cdef uint32_t lenmod = len & 0x1f
+    return BARREL_SHIFT(sum, 1) ^ BARREL_SHIFT(h[remove], lenmod) ^ h[add]
+
+
+cdef class Chunker:
+    """
+    Content-Defined Chunker, variable chunk sizes.
+
+    This chunker makes quite some effort to cut mostly chunks of the same-content, even if
+    the content moves to a different offset inside the file. It uses the buzhash
+    rolling-hash algorithm to identify the chunk cutting places by looking at the
+    content inside the moving window and computing the rolling hash value over the
+    window contents. If the last n bits of the rolling hash are 0, a chunk is cut.
+    Additionally it obeys some more criteria, like a minimum and maximum chunk size.
+    It also uses a per-repo random seed to avoid some chunk length fingerprinting attacks.
+    """
+    cdef uint32_t chunk_mask
+    cdef uint32_t* table
+    cdef uint8_t* data
+    cdef object _fd  # Python object for file descriptor
+    cdef int fh
+    cdef int done, eof
+    cdef size_t min_size, buf_size, window_size, remaining, position, last
+    cdef long long bytes_read, bytes_yielded  # off_t in C, using long long for compatibility
+    cdef readonly float chunking_time
+    cdef object file_reader  # FileReader instance
+    cdef size_t reader_block_size
+    cdef bint sparse
+
+    def __cinit__(self, int seed, int chunk_min_exp, int chunk_max_exp, int hash_mask_bits, int hash_window_size, bint sparse=False):
+        min_size = 1 << chunk_min_exp
+        max_size = 1 << chunk_max_exp
+        assert max_size <= len(zeros)
+        # see chunker_process, first while loop condition, first term must be able to get True:
+        assert hash_window_size + min_size + 1 <= max_size, "too small max_size"
+
+        self.window_size = hash_window_size
+        self.chunk_mask = (1 << hash_mask_bits) - 1
+        self.min_size = min_size
+        self.table = buzhash_init_table(seed & 0xffffffff)
+        self.buf_size = max_size
+        self.data = <uint8_t*>malloc(self.buf_size)
+        self.fh = -1
+        self.done = 0
+        self.eof = 0
+        self.remaining = 0
+        self.position = 0
+        self.last = 0
+        self.bytes_read = 0
+        self.bytes_yielded = 0
+        self._fd = None
+        self.chunking_time = 0.0
+        self.reader_block_size = 1024 * 1024
+        self.sparse = sparse
+
+    def __dealloc__(self):
+        """Free the chunker's resources."""
+        if self.table != NULL:
+            free(self.table)
+            self.table = NULL
+        if self.data != NULL:
+            free(self.data)
+            self.data = NULL
+
+    cdef int fill(self) except 0:
+        """Fill the chunker's buffer with more data."""
+        cdef ssize_t n
+        cdef object chunk
+
+        # Move remaining data to the beginning of the buffer
+        memmove(self.data, self.data + self.last, self.position + self.remaining - self.last)
+        self.position -= self.last
+        self.last = 0
+        n = self.buf_size - self.position - self.remaining
+
+        if self.eof or n == 0:
+            return 1
+
+        # Use FileReader to read data
+        chunk = self.file_reader.read(n)
+        n = chunk.meta["size"]
+
+        if n > 0:
+            # Only copy data if it's not a hole
+            if chunk.meta["allocation"] == CH_DATA:
+                # Copy data from chunk to our buffer
+                memcpy(self.data + self.position + self.remaining, <const unsigned char*>PyBytes_AsString(chunk.data), n)
+            else:
+                # For holes, fill with zeros
+                memcpy(self.data + self.position + self.remaining, <const unsigned char*>PyBytes_AsString(zeros[:n]), n)
+
+            self.remaining += n
+            self.bytes_read += n
+        else:
+            self.eof = 1
+
+        return 1
+
+    cdef object process(self) except *:
+        """Process the chunker's buffer and return the next chunk."""
+        cdef uint32_t sum, chunk_mask = self.chunk_mask
+        cdef size_t n, old_last, min_size = self.min_size, window_size = self.window_size
+        cdef uint8_t* p
+        cdef uint8_t* stop_at
+        cdef size_t did_bytes
+
+        if self.done:
+            if self.bytes_read == self.bytes_yielded:
+                raise StopIteration
+            else:
+                raise Exception("chunkifier byte count mismatch")
+
+        while self.remaining < min_size + window_size + 1 and not self.eof:  # see assert in Chunker init
+            if not self.fill():
+                return None
+
+        # Here we either are at eof...
+        if self.eof:
+            self.done = 1
+            if self.remaining:
+                self.bytes_yielded += self.remaining
+                # Return a memory view of the remaining data
+                return memoryview((self.data + self.position)[:self.remaining])
+            else:
+                if self.bytes_read == self.bytes_yielded:
+                    raise StopIteration
+                else:
+                    raise Exception("chunkifier byte count mismatch")
+
+        # ... or we have at least min_size + window_size + 1 bytes remaining.
+        # We do not want to "cut" a chunk smaller than min_size and the hash
+        # window starts at the potential cutting place.
+        self.position += min_size
+        self.remaining -= min_size
+        sum = _buzhash(self.data + self.position, window_size, self.table)
+
+        while self.remaining > self.window_size and (sum & chunk_mask) and not (self.eof and self.remaining <= window_size):
+            p = self.data + self.position
+            stop_at = p + self.remaining - window_size
+
+            while p < stop_at and (sum & chunk_mask):
+                sum = _buzhash_update(sum, p[0], p[window_size], window_size, self.table)
+                p += 1
+
+            did_bytes = p - (self.data + self.position)
+            self.position += did_bytes
+            self.remaining -= did_bytes
+
+            if self.remaining <= window_size:
+                if not self.fill():
+                    return None
+
+        if self.remaining <= window_size:
+            self.position += self.remaining
+            self.remaining = 0
+
+        old_last = self.last
+        self.last = self.position
+        n = self.last - old_last
+        self.bytes_yielded += n
+
+        # Return a memory view of the chunk
+        return memoryview((self.data + old_last)[:n])
+
+    def chunkify(self, fd, fh=-1, fmap=None):
+        """
+        Cut a file into chunks.
+
+        :param fd: Python file object
+        :param fh: OS-level file handle (if available),
+                   defaults to -1 which means not to use OS-level fd.
+        :param fmap: a file map, same format as generated by sparsemap
+        """
+        self._fd = fd
+        self.fh = fh
+        self.file_reader = FileReader(fd=fd, fh=fh, read_size=self.reader_block_size, sparse=self.sparse, fmap=fmap)
+        self.done = 0
+        self.remaining = 0
+        self.bytes_read = 0
+        self.bytes_yielded = 0
+        self.position = 0
+        self.last = 0
+        self.eof = 0
+        return self
+
+    def __iter__(self):
+        return self
+
+    def __next__(self):
+        started_chunking = time.monotonic()
+        data = self.process()
+        got = len(data)
+        # we do not have SEEK_DATA/SEEK_HOLE support in chunker_process C code,
+        # but we can just check if data was all-zero (and either came from a hole
+        # or from stored zeros - we can not detect that here).
+        if zeros.startswith(data):
+            data = None
+            allocation = CH_ALLOC
+        else:
+            allocation = CH_DATA
+        self.chunking_time += time.monotonic() - started_chunking
+        return Chunk(data, size=got, allocation=allocation)
+
+
+def buzhash(data, unsigned long seed):
+    cdef uint32_t *table
+    cdef uint32_t sum
+    table = buzhash_init_table(seed & 0xffffffff)
+    sum = _buzhash(<const unsigned char *> data, len(data), table)
+    free(table)
+    return sum
+
+
+def buzhash_update(uint32_t sum, unsigned char remove, unsigned char add, size_t len, unsigned long seed):
+    cdef uint32_t *table
+    table = buzhash_init_table(seed & 0xffffffff)
+    sum = _buzhash_update(sum, remove, add, len, table)
+    free(table)
+    return sum

src/borg/chunkers/chunker.pyi

@@ -4,9 +4,6 @@ from .reader import fmap_entry
 
 API_VERSION: str
 
-def buzhash(data: bytes, seed: int) -> int: ...
-def buzhash_update(sum: int, remove: int, add: int, len: int, seed: int) -> int: ...
-
 class ChunkerFailing:
     def __init__(self, block_size: int, map: str) -> None: ...
     def chunkify(self, fd: BinaryIO = None, fh: int = -1) -> Iterator: ...
@@ -14,15 +11,3 @@ class ChunkerFailing:
 class ChunkerFixed:
     def __init__(self, block_size: int, header_size: int = 0, sparse: bool = False) -> None: ...
     def chunkify(self, fd: BinaryIO = None, fh: int = -1, fmap: List[fmap_entry] = None) -> Iterator: ...
-
-class Chunker:
-    def __init__(
-        self,
-        seed: int,
-        chunk_min_exp: int,
-        chunk_max_exp: int,
-        hash_mask_bits: int,
-        hash_window_size: int,
-        sparse: bool = False,
-    ) -> None: ...
-    def chunkify(self, fd: BinaryIO = None, fh: int = -1, fmap: List[fmap_entry] = None) -> Iterator: ...

src/borg/chunkers/chunker.pyx

@@ -2,14 +2,9 @@
 
 API_VERSION = '1.2_01'
 
-import cython
 import os
 import errno
 import time
-from cpython.bytes cimport PyBytes_AsString
-from libc.stdint cimport uint8_t, uint32_t
-from libc.stdlib cimport malloc, free
-from libc.string cimport memcpy, memmove
 
 from ..constants import CH_DATA, CH_ALLOC, CH_HOLE, zeros
 from .reader import FileReader, Chunk
@@ -124,323 +119,3 @@ class ChunkerFixed:
                 break  # EOF
             assert size <= self.block_size
             yield chunk
-
-
-# Cyclic polynomial / buzhash
-#
-# https://en.wikipedia.org/wiki/Rolling_hash
-#
-# http://www.serve.net/buz/Notes.1st.year/HTML/C6/rand.012.html (by "BUZ", the inventor)
-#
-# http://www.dcs.gla.ac.uk/~hamer/cakes-talk.pdf (see buzhash slide)
-#
-# Some properties of buzhash / of this implementation:
-#
-# (1) the hash is designed for inputs <= 32 bytes, but the chunker uses it on a 4095 byte window;
-#     any repeating bytes at distance 32 within those 4095 bytes can cause cancellation within
-#     the hash function, e.g. in "X <any 31 bytes> X", the last X would cancel out the influence
-#     of the first X on the hash value.
-#
-# (2) the hash table is supposed to have (according to the BUZ) exactly a 50% distribution of
-#     0/1 bit values per position, but the hard coded table below doesn't fit that property.
-#
-# (3) if you would use a window size divisible by 64, the seed would cancel itself out completely.
-#     this is why we use a window size of 4095 bytes.
-#
-# Another quirk is that, even with the 4095 byte window, XORing the entire table by a constant
-# is equivalent to XORing the hash output with a different constant. but since the seed is stored
-# encrypted, i think it still serves its purpose.
-
-cdef uint32_t table_base[256]
-table_base = [
-    0xe7f831ec, 0xf4026465, 0xafb50cae, 0x6d553c7a, 0xd639efe3, 0x19a7b895, 0x9aba5b21, 0x5417d6d4,
-    0x35fd2b84, 0xd1f6a159, 0x3f8e323f, 0xb419551c, 0xf444cebf, 0x21dc3b80, 0xde8d1e36, 0x84a32436,
-    0xbeb35a9d, 0xa36f24aa, 0xa4e60186, 0x98d18ffe, 0x3f042f9e, 0xdb228bcd, 0x096474b7, 0x5c20c2f7,
-    0xf9eec872, 0xe8625275, 0xb9d38f80, 0xd48eb716, 0x22a950b4, 0x3cbaaeaa, 0xc37cddd3, 0x8fea6f6a,
-    0x1d55d526, 0x7fd6d3b3, 0xdaa072ee, 0x4345ac40, 0xa077c642, 0x8f2bd45b, 0x28509110, 0x55557613,
-    0xffc17311, 0xd961ffef, 0xe532c287, 0xaab95937, 0x46d38365, 0xb065c703, 0xf2d91d0f, 0x92cd4bb0,
-    0x4007c712, 0xf35509dd, 0x505b2f69, 0x557ead81, 0x310f4563, 0xbddc5be8, 0x9760f38c, 0x701e0205,
-    0x00157244, 0x14912826, 0xdc4ca32b, 0x67b196de, 0x5db292e8, 0x8c1b406b, 0x01f34075, 0xfa2520f7,
-    0x73bc37ab, 0x1e18bc30, 0xfe2c6cb3, 0x20c522d0, 0x5639e3db, 0x942bda35, 0x899af9d1, 0xced44035,
-    0x98cc025b, 0x255f5771, 0x70fefa24, 0xe928fa4d, 0x2c030405, 0xb9325590, 0x20cb63bd, 0xa166305d,
-    0x80e52c0a, 0xa8fafe2f, 0x1ad13f7d, 0xcfaf3685, 0x6c83a199, 0x7d26718a, 0xde5dfcd9, 0x79cf7355,
-    0x8979d7fb, 0xebf8c55e, 0xebe408e4, 0xcd2affba, 0xe483be6e, 0xe239d6de, 0x5dc1e9e0, 0x0473931f,
-    0x851b097c, 0xac5db249, 0x09c0f9f2, 0xd8d2f134, 0xe6f38e41, 0xb1c71bf1, 0x52b6e4db, 0x07224424,
-    0x6cf73e85, 0x4f25d89c, 0x782a7d74, 0x10a68dcd, 0x3a868189, 0xd570d2dc, 0x69630745, 0x9542ed86,
-    0x331cd6b2, 0xa84b5b28, 0x07879c9d, 0x38372f64, 0x7185db11, 0x25ba7c83, 0x01061523, 0xe6792f9f,
-    0xe5df07d1, 0x4321b47f, 0x7d2469d8, 0x1a3a4f90, 0x48be29a3, 0x669071af, 0x8ec8dd31, 0x0810bfbf,
-    0x813a06b4, 0x68538345, 0x65865ddc, 0x43a71b8e, 0x78619a56, 0x5a34451d, 0x5bdaa3ed, 0x71edc7e9,
-    0x17ac9a20, 0x78d10bfa, 0x6c1e7f35, 0xd51839d9, 0x240cbc51, 0x33513cc1, 0xd2b4f795, 0xccaa8186,
-    0x0babe682, 0xa33cf164, 0x18c643ea, 0xc1ca105f, 0x9959147a, 0x6d3d94de, 0x0b654fbe, 0xed902ca0,
-    0x7d835cb5, 0x99ba1509, 0x6445c922, 0x495e76c2, 0xf07194bc, 0xa1631d7e, 0x677076a5, 0x89fffe35,
-    0x1a49bcf3, 0x8e6c948a, 0x0144c917, 0x8d93aea1, 0x16f87ddf, 0xc8f25d49, 0x1fb11297, 0x27e750cd,
-    0x2f422da1, 0xdee89a77, 0x1534c643, 0x457b7b8b, 0xaf172f7a, 0x6b9b09d6, 0x33573f7f, 0xf14e15c4,
-    0x526467d5, 0xaf488241, 0x87c3ee0d, 0x33be490c, 0x95aa6e52, 0x43ec242e, 0xd77de99b, 0xd018334f,
-    0x5b78d407, 0x498eb66b, 0xb1279fa8, 0xb38b0ea6, 0x90718376, 0xe325dee2, 0x8e2f2cba, 0xcaa5bdec,
-    0x9d652c56, 0xad68f5cb, 0xa77591af, 0x88e37ee8, 0xf8faa221, 0xfcbbbe47, 0x4f407786, 0xaf393889,
-    0xf444a1d9, 0x15ae1a2f, 0x40aa7097, 0x6f9486ac, 0x29d232a3, 0xe47609e9, 0xe8b631ff, 0xba8565f4,
-    0x11288749, 0x46c9a838, 0xeb1b7cd8, 0xf516bbb1, 0xfb74fda0, 0x010996e6, 0x4c994653, 0x1d889512,
-    0x53dcd9a3, 0xdd074697, 0x1e78e17c, 0x637c98bf, 0x930bb219, 0xcf7f75b0, 0xcb9355fb, 0x9e623009,
-    0xe466d82c, 0x28f968d3, 0xfeb385d9, 0x238e026c, 0xb8ed0560, 0x0c6a027a, 0x3d6fec4b, 0xbb4b2ec2,
-    0xe715031c, 0xeded011d, 0xcdc4d3b9, 0xc456fc96, 0xdd0eea20, 0xb3df8ec9, 0x12351993, 0xd9cbb01c,
-    0x603147a2, 0xcf37d17d, 0xf7fcd9dc, 0xd8556fa3, 0x104c8131, 0x13152774, 0xb4715811, 0x6a72c2c9,
-    0xc5ae37bb, 0xa76ce12a, 0x8150d8f3, 0x2ec29218, 0xa35f0984, 0x48c0647e, 0x0b5ff98c, 0x71893f7b
-]
-
-# This seems to be the most reliable way to inline this code, using a C preprocessor macro:
-cdef extern from *:
-   """
-   #define BARREL_SHIFT(v, shift) (((v) << (shift)) | ((v) >> (((32 - (shift)) & 0x1f))))
-   """
-   uint32_t BARREL_SHIFT(uint32_t v, uint32_t shift)
-
-
-@cython.boundscheck(False)  # Deactivate bounds checking
-@cython.wraparound(False)  # Deactivate negative indexing.
-cdef uint32_t* buzhash_init_table(uint32_t seed):
-    """Initialize the buzhash table with the given seed."""
-    cdef int i
-    cdef uint32_t* table = <uint32_t*>malloc(1024)  # 256 * sizeof(uint32_t)
-    for i in range(256):
-        table[i] = table_base[i] ^ seed
-    return table
-
-
-@cython.boundscheck(False)  # Deactivate bounds checking
-@cython.wraparound(False)  # Deactivate negative indexing.
-@cython.cdivision(True)  # Use C division/modulo semantics for integer division.
-cdef uint32_t _buzhash(const unsigned char* data, size_t len, const uint32_t* h):
-    """Calculate the buzhash of the given data."""
-    cdef uint32_t i
-    cdef uint32_t sum = 0, imod
-    for i in range(len - 1, 0, -1):
-        imod = i & 0x1f
-        sum ^= BARREL_SHIFT(h[data[0]], imod)
-        data += 1
-    return sum ^ h[data[0]]
-
-
-@cython.boundscheck(False)  # Deactivate bounds checking
-@cython.wraparound(False)  # Deactivate negative indexing.
-@cython.cdivision(True)  # Use C division/modulo semantics for integer division.
-cdef uint32_t _buzhash_update(uint32_t sum, unsigned char remove, unsigned char add, size_t len, const uint32_t* h):
-    """Update the buzhash with a new byte."""
-    cdef uint32_t lenmod = len & 0x1f
-    return BARREL_SHIFT(sum, 1) ^ BARREL_SHIFT(h[remove], lenmod) ^ h[add]
-
-
-cdef class Chunker:
-    """
-    Content-Defined Chunker, variable chunk sizes.
-
-    This chunker makes quite some effort to cut mostly chunks of the same-content, even if
-    the content moves to a different offset inside the file. It uses the buzhash
-    rolling-hash algorithm to identify the chunk cutting places by looking at the
-    content inside the moving window and computing the rolling hash value over the
-    window contents. If the last n bits of the rolling hash are 0, a chunk is cut.
-    Additionally it obeys some more criteria, like a minimum and maximum chunk size.
-    It also uses a per-repo random seed to avoid some chunk length fingerprinting attacks.
-    """
-    cdef uint32_t chunk_mask
-    cdef uint32_t* table
-    cdef uint8_t* data
-    cdef object _fd  # Python object for file descriptor
-    cdef int fh
-    cdef int done, eof
-    cdef size_t min_size, buf_size, window_size, remaining, position, last
-    cdef long long bytes_read, bytes_yielded  # off_t in C, using long long for compatibility
-    cdef readonly float chunking_time
-    cdef object file_reader  # FileReader instance
-    cdef size_t reader_block_size
-    cdef bint sparse
-
-    def __cinit__(self, int seed, int chunk_min_exp, int chunk_max_exp, int hash_mask_bits, int hash_window_size, bint sparse=False):
-        min_size = 1 << chunk_min_exp
-        max_size = 1 << chunk_max_exp
-        assert max_size <= len(zeros)
-        # see chunker_process, first while loop condition, first term must be able to get True:
-        assert hash_window_size + min_size + 1 <= max_size, "too small max_size"
-
-        self.window_size = hash_window_size
-        self.chunk_mask = (1 << hash_mask_bits) - 1
-        self.min_size = min_size
-        self.table = buzhash_init_table(seed & 0xffffffff)
-        self.buf_size = max_size
-        self.data = <uint8_t*>malloc(self.buf_size)
-        self.fh = -1
-        self.done = 0
-        self.eof = 0
-        self.remaining = 0
-        self.position = 0
-        self.last = 0
-        self.bytes_read = 0
-        self.bytes_yielded = 0
-        self._fd = None
-        self.chunking_time = 0.0
-        self.reader_block_size = 1024 * 1024
-        self.sparse = sparse
-
-    def __dealloc__(self):
-        """Free the chunker's resources."""
-        if self.table != NULL:
-            free(self.table)
-            self.table = NULL
-        if self.data != NULL:
-            free(self.data)
-            self.data = NULL
-
-    cdef int fill(self) except 0:
-        """Fill the chunker's buffer with more data."""
-        cdef ssize_t n
-        cdef object chunk
-
-        # Move remaining data to the beginning of the buffer
-        memmove(self.data, self.data + self.last, self.position + self.remaining - self.last)
-        self.position -= self.last
-        self.last = 0
-        n = self.buf_size - self.position - self.remaining
-
-        if self.eof or n == 0:
-            return 1
-
-        # Use FileReader to read data
-        chunk = self.file_reader.read(n)
-        n = chunk.meta["size"]
-
-        if n > 0:
-            # Only copy data if it's not a hole
-            if chunk.meta["allocation"] == CH_DATA:
-                # Copy data from chunk to our buffer
-                memcpy(self.data + self.position + self.remaining, <const unsigned char*>PyBytes_AsString(chunk.data), n)
-            else:
-                # For holes, fill with zeros
-                memcpy(self.data + self.position + self.remaining, <const unsigned char*>PyBytes_AsString(zeros[:n]), n)
-
-            self.remaining += n
-            self.bytes_read += n
-        else:
-            self.eof = 1
-
-        return 1
-
-    cdef object process(self) except *:
-        """Process the chunker's buffer and return the next chunk."""
-        cdef uint32_t sum, chunk_mask = self.chunk_mask
-        cdef size_t n, old_last, min_size = self.min_size, window_size = self.window_size
-        cdef uint8_t* p
-        cdef uint8_t* stop_at
-        cdef size_t did_bytes
-
-        if self.done:
-            if self.bytes_read == self.bytes_yielded:
-                raise StopIteration
-            else:
-                raise Exception("chunkifier byte count mismatch")
-
-        while self.remaining < min_size + window_size + 1 and not self.eof:  # see assert in Chunker init
-            if not self.fill():
-                return None
-
-        # Here we either are at eof...
-        if self.eof:
-            self.done = 1
-            if self.remaining:
-                self.bytes_yielded += self.remaining
-                # Return a memory view of the remaining data
-                return memoryview((self.data + self.position)[:self.remaining])
-            else:
-                if self.bytes_read == self.bytes_yielded:
-                    raise StopIteration
-                else:
-                    raise Exception("chunkifier byte count mismatch")
-
-        # ... or we have at least min_size + window_size + 1 bytes remaining.
-        # We do not want to "cut" a chunk smaller than min_size and the hash
-        # window starts at the potential cutting place.
-        self.position += min_size
-        self.remaining -= min_size
-        sum = _buzhash(self.data + self.position, window_size, self.table)
-
-        while self.remaining > self.window_size and (sum & chunk_mask) and not (self.eof and self.remaining <= window_size):
-            p = self.data + self.position
-            stop_at = p + self.remaining - window_size
-
-            while p < stop_at and (sum & chunk_mask):
-                sum = _buzhash_update(sum, p[0], p[window_size], window_size, self.table)
-                p += 1
-
-            did_bytes = p - (self.data + self.position)
-            self.position += did_bytes
-            self.remaining -= did_bytes
-
-            if self.remaining <= window_size:
-                if not self.fill():
-                    return None
-
-        if self.remaining <= window_size:
-            self.position += self.remaining
-            self.remaining = 0
-
-        old_last = self.last
-        self.last = self.position
-        n = self.last - old_last
-        self.bytes_yielded += n
-
-        # Return a memory view of the chunk
-        return memoryview((self.data + old_last)[:n])
-
-    def chunkify(self, fd, fh=-1, fmap=None):
-        """
-        Cut a file into chunks.
-
-        :param fd: Python file object
-        :param fh: OS-level file handle (if available),
-                   defaults to -1 which means not to use OS-level fd.
-        :param fmap: a file map, same format as generated by sparsemap
-        """
-        self._fd = fd
-        self.fh = fh
-        self.file_reader = FileReader(fd=fd, fh=fh, read_size=self.reader_block_size, sparse=self.sparse, fmap=fmap)
-        self.done = 0
-        self.remaining = 0
-        self.bytes_read = 0
-        self.bytes_yielded = 0
-        self.position = 0
-        self.last = 0
-        self.eof = 0
-        return self
-
-    def __iter__(self):
-        return self
-
-    def __next__(self):
-        started_chunking = time.monotonic()
-        data = self.process()
-        got = len(data)
-        # we do not have SEEK_DATA/SEEK_HOLE support in chunker_process C code,
-        # but we can just check if data was all-zero (and either came from a hole
-        # or from stored zeros - we can not detect that here).
-        if zeros.startswith(data):
-            data = None
-            allocation = CH_ALLOC
-        else:
-            allocation = CH_DATA
-        self.chunking_time += time.monotonic() - started_chunking
-        return Chunk(data, size=got, allocation=allocation)
-
-
-def buzhash(data, unsigned long seed):
-    cdef uint32_t *table
-    cdef uint32_t sum
-    table = buzhash_init_table(seed & 0xffffffff)
-    sum = _buzhash(<const unsigned char *> data, len(data), table)
-    free(table)
-    return sum
-
-
-def buzhash_update(uint32_t sum, unsigned char remove, unsigned char add, size_t len, unsigned long seed):
-    cdef uint32_t *table
-    table = buzhash_init_table(seed & 0xffffffff)
-    sum = _buzhash_update(sum, remove, add, len, table)
-    free(table)
-    return sum