diff --git a/src/borg/chunker.pyi b/src/borg/chunker.pyi
index 1a4536d03..1ba388046 100644
--- a/src/borg/chunker.pyi
+++ b/src/borg/chunker.pyi
@@ -21,10 +21,33 @@ class ChunkerFailing:
     def __init__(self, block_size: int, map: str) -> None: ...
     def chunkify(self, fd: BinaryIO = None, fh: int = -1) -> Iterator: ...
 
+class FileFMAPReader:
+    def __init__(
+        self,
+        *,
+        fd: BinaryIO = None,
+        fh: int = -1,
+        read_size: int = 0,
+        header_size: int = 0,
+        sparse: bool = False,
+        fmap: List[fmap_entry] = None,
+    ) -> None: ...
+    def _build_fmap(self) -> List[fmap_entry]: ...
+    def blockify(self) -> Iterator: ...
+
 class FileReader:
-    def __init__(self, block_size: int, header_size: int = 0, sparse: bool = False) -> None: ...
-    def _build_fmap(self, fd: BinaryIO = None, fh: int = -1) -> List[fmap_entry]: ...
-    def blockify(self, fd: BinaryIO = None, fh: int = -1, fmap: List[fmap_entry] = None) -> Iterator: ...
+    def __init__(
+        self,
+        *,
+        fd: BinaryIO = None,
+        fh: int = -1,
+        read_size: int = 0,
+        header_size: int = 0,
+        sparse: bool = False,
+        fmap: List[fmap_entry] = None,
+    ) -> None: ...
+    def _fill_buffer(self) -> bool: ...
+    def read(self, size: int, return_chunk_info: bool = False) -> Any: ...
 
 class ChunkerFixed:
     def __init__(self, block_size: int, header_size: int = 0, sparse: bool = False) -> None: ...
diff --git a/src/borg/chunker.pyx b/src/borg/chunker.pyx
index 0c37ab7e1..ed3ae22db 100644
--- a/src/borg/chunker.pyx
+++ b/src/borg/chunker.pyx
@@ -165,7 +165,7 @@ class ChunkerFailing:
                 return
 
 
-class FileReader:
+class FileFMAPReader:
     """
     This is for reading blocks from a file.
 
@@ -180,29 +180,34 @@ class FileReader:
     Note: the last block of a data or hole range may be less than the block size,
           this is supported and not considered to be an error.
     """
-    def __init__(self, read_size, header_size=0, sparse=False):
-        self.read_size = read_size  # how much data we want to read at once
+    def __init__(self, *, fd=None, fh=-1, read_size=0, header_size=0, sparse=False, fmap=None):
+        assert fd is not None or fh >= 0
+        self.fd = fd
+        self.fh = fh
+        assert read_size > 0
         assert read_size <= len(zeros)
+        self.read_size = read_size  # how much data we want to read at once
+        assert header_size <= read_size
         self.header_size = header_size  # size of the first block
-        assert read_size >= header_size
         self.reading_time = 0.0  # time spent in reading/seeking
         # should borg try to do sparse input processing?
         # whether it actually can be done depends on the input file being seekable.
         self.try_sparse = sparse and has_seek_hole
+        self.fmap = fmap
 
-    def _build_fmap(self, fd=None, fh=-1):
+    def _build_fmap(self):
         started_fmap = time.monotonic()
         fmap = None
         if self.try_sparse:
             try:
                 if self.header_size > 0:
                     header_map = [(0, self.header_size, True), ]
-                    dseek(self.header_size, os.SEEK_SET, fd, fh)
-                    body_map = list(sparsemap(fd, fh))
-                    dseek(0, os.SEEK_SET, fd, fh)
+                    dseek(self.header_size, os.SEEK_SET, self.fd, self.fh)
+                    body_map = list(sparsemap(self.fd, self.fh))
+                    dseek(0, os.SEEK_SET, self.fd, self.fh)
                 else:
                     header_map = []
-                    body_map = list(sparsemap(fd, fh))
+                    body_map = list(sparsemap(self.fd, self.fh))
             except OSError as err:
                 # seeking did not work
                 pass
@@ -225,30 +230,27 @@ class FileReader:
         self.reading_time += time.monotonic() - started_fmap
         return fmap
 
-    def blockify(self, fd=None, fh=-1, fmap=None):
+    def blockify(self):
         """
         Read <read_size> sized blocks from a file, optionally supporting a differently sized header block.
-
-        :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
         """
-        fmap =self._build_fmap(fd, fh) if fmap is None else fmap
+        if self.fmap is None:
+            self.fmap = self._build_fmap()
+
         offset = 0
         # note: the optional header block is implemented via the first fmap entry
-        for range_start, range_size, is_data in fmap:
+        for range_start, range_size, is_data in self.fmap:
             if range_start != offset:
                 # this is for the case when the fmap does not cover the file completely,
                 # e.g. it could be without the ranges of holes or of unchanged data.
                 offset = range_start
-                dseek(offset, os.SEEK_SET, fd, fh)
+                dseek(offset, os.SEEK_SET, self.fd, self.fh)
             while range_size:
                 started_reading = time.monotonic()
                 wanted = min(range_size, self.read_size)
                 if is_data:
                     # read block from the range
-                    data = dread(offset, wanted, fd, fh)
+                    data = dread(offset, wanted, self.fd, self.fh)
                     got = len(data)
                     if zeros.startswith(data):
                         data = None
@@ -257,20 +259,164 @@ class FileReader:
                         allocation = CH_DATA
                 else:  # hole
                     # seek over block from the range
-                    pos = dseek(wanted, os.SEEK_CUR, fd, fh)
+                    pos = dseek(wanted, os.SEEK_CUR, self.fd, self.fh)
                     got = pos - offset
                     data = None
                     allocation = CH_HOLE
+                self.reading_time += time.monotonic() - started_reading
                 if got > 0:
                     offset += got
                     range_size -= got
-                    self.reading_time += time.monotonic() - started_reading
                     yield Chunk(data, size=got, allocation=allocation)
                 if got < wanted:
                     # we did not get enough data, looks like EOF.
                     return
 
 
+class FileReader:
+    """
+    This is a buffered reader for file data.
+
+    It maintains a buffer that is filled by using FileFMAPReader.blockify generator when needed.
+    The data in that buffer is consumed by clients calling FileReader.read.
+    """
+    def __init__(self, *, fd=None, fh=-1, read_size=0, header_size=0, sparse=False, fmap=None):
+        self.reader = FileFMAPReader(fd=fd, fh=fh, read_size=read_size, header_size=header_size, sparse=sparse, fmap=fmap)
+        self.buffer = []  # list of (data, meta) tuples
+        self.offset = 0  # offset into the first buffer object's data
+        self.remaining_bytes = 0  # total bytes available in buffer
+        self.blockify_gen = None  # generator from FileFMAPReader.blockify
+        self.fd = fd
+        self.fh = fh
+        self.fmap = fmap
+
+    def _fill_buffer(self):
+        """
+        Fill the buffer with more data from the blockify generator.
+        Returns True if more data was added, False if EOF.
+        """
+        if self.blockify_gen is None:
+            return False
+
+        try:
+            chunk = next(self.blockify_gen)
+            # Store both data and metadata in the buffer
+            self.buffer.append((chunk.data, chunk.meta))
+            self.remaining_bytes += chunk.meta["size"]
+            return True
+        except StopIteration:
+            self.blockify_gen = None
+            return False
+
+    def read(self, size, return_chunk_info=False):
+        """
+        Read up to 'size' bytes from the file.
+
+        :param size: Number of bytes to read
+        :param return_chunk_info: if True, return a tuple (data, allocation, size) instead of just data
+        :return: Bytes object containing the read data, or None if no data is available.
+                 If return_chunk_info is True, returns a tuple (data, allocation, size).
+        """
+        # Initialize if not already done
+        if self.blockify_gen is None:
+            self.buffer = []
+            self.offset = 0
+            self.remaining_bytes = 0
+            self.blockify_gen = self.reader.blockify()
+
+        # If we don't have enough data in the buffer, try to fill it
+        while self.remaining_bytes < size:
+            if not self._fill_buffer():
+                # No more data available, return what we have
+                break
+
+        # If we have no data at all, return None
+        if not self.buffer:
+            return None if not return_chunk_info else (None, None, 0)
+
+        # Get the first chunk from the buffer
+        data, meta = self.buffer[0]
+        chunk_size = meta["size"]
+        allocation = meta["allocation"]
+
+        # If we're returning chunk info and this is a non-data chunk, handle it specially
+        if return_chunk_info and (allocation != CH_DATA or data is None):
+            # For non-data chunks, we return the allocation type and size
+            size_to_return = min(size, chunk_size - self.offset)
+
+            # Update buffer state
+            if size_to_return == chunk_size - self.offset:
+                self.buffer.pop(0)
+                self.offset = 0
+            else:
+                self.offset += size_to_return
+
+            self.remaining_bytes -= size_to_return
+
+            return (None, allocation, size_to_return)
+
+        # For data chunks or when not returning chunk info, proceed as before
+        # Prepare to collect the requested data
+        result = bytearray()
+        bytes_to_read = min(size, self.remaining_bytes)
+        bytes_read = 0
+
+        # Read data from the buffer
+        while bytes_read < bytes_to_read and self.buffer:
+            data, meta = self.buffer[0]
+            chunk_size = meta["size"]
+            allocation = meta["allocation"]
+
+            # Skip non-data chunks if not returning chunk info
+            if (allocation != CH_DATA or data is None) and not return_chunk_info:
+                self.buffer.pop(0)
+                self.remaining_bytes -= chunk_size
+                continue
+
+            # If this is a non-data chunk and we're returning chunk info, break to handle it
+            if (allocation != CH_DATA or data is None) and return_chunk_info:
+                if bytes_read > 0:
+                    # We've already read some data, so return that first
+                    break
+                else:
+                    # No data read yet, return info about this non-data chunk
+                    size_to_return = min(size, chunk_size - self.offset)
+
+                    # Update buffer state
+                    if size_to_return == chunk_size - self.offset:
+                        self.buffer.pop(0)
+                        self.offset = 0
+                    else:
+                        self.offset += size_to_return
+
+                    self.remaining_bytes -= size_to_return
+
+                    return (None, allocation, size_to_return)
+
+            # Calculate how much we can read from this chunk
+            available = chunk_size - self.offset
+            to_read = min(available, bytes_to_read - bytes_read)
+
+            # Read the data
+            if to_read > 0:
+                result.extend(data[self.offset:self.offset + to_read])
+                bytes_read += to_read
+
+                # Update offset or remove chunk if fully consumed
+                if to_read < available:
+                    self.offset += to_read
+                else:
+                    self.offset = 0
+                    self.buffer.pop(0)
+
+                self.remaining_bytes -= to_read
+
+        if return_chunk_info:
+            return (bytes(result) if result else None, CH_DATA, bytes_read)
+        else:
+            return bytes(result) if result else None
+
+
 class ChunkerFixed:
     """
     This is a simple chunker for input data with data usually staying at same
@@ -297,7 +443,8 @@ class ChunkerFixed:
         self.chunking_time = 0.0  # likely will stay close to zero - not much to do here.
         self.reader_block_size = self.block_size  # start simple
         assert self.reader_block_size % self.block_size == 0, "reader_block_size must be N * block_size"
-        self.reader = FileReader(self.reader_block_size, header_size=self.header_size, sparse=sparse)
+        self.reader = None
+        self.sparse = sparse
 
     def chunkify(self, fd=None, fh=-1, fmap=None):
         """
@@ -308,35 +455,39 @@ class ChunkerFixed:
                    defaults to -1 which means not to use OS-level fd.
         :param fmap: a file map, same format as generated by sparsemap
         """
-        in_header = self.header_size > 0  # first block is header, if header size is given
-        for block in self.reader.blockify(fd, fh, fmap):
-            if in_header:
-                assert self.header_size == block.meta["size"]
-                yield block  # just pass through the header block we get from the reader
-                in_header = False
-                continue
-            # not much to do in here
-            if self.reader_block_size == self.block_size:
-                # trivial, the reader already did all the work
-                yield block  # just pass through, avoid creating new objects
-            else:
-                # reader block size is a multiple of our block size
-                read_size = block.meta["size"]
-                allocation = block.meta["allocation"]
-                start = 0
-                while read_size:
-                    started_chunking = time.monotonic()
-                    size = min(read_size, self.block_size)
-                    if allocation == CH_DATA:
-                        data = block.data[start:start+size]  # TODO memoryview?
-                    elif allocation in (CH_ALLOC, CH_HOLE):
-                        data = None
-                    else:
-                        raise ValueError("unsupported allocation")
-                    self.chunking_time += time.monotonic() - started_chunking
-                    yield Chunk(data, size=size, allocation=allocation)
-                    start += size
-                    read_size -= size
+        # Initialize the reader with the file descriptors
+        self.reader = FileReader(fd=fd, fh=fh, read_size=self.reader_block_size,
+                                header_size=self.header_size, sparse=self.sparse, fmap=fmap)
+
+        # Handle header if present
+        if self.header_size > 0:
+            # Read the header block using read
+            started_chunking = time.monotonic()
+            header_info = self.reader.read(self.header_size, return_chunk_info=True)
+            self.chunking_time += time.monotonic() - started_chunking
+
+            if header_info is not None and header_info[2] > 0:
+                # Unpack the header info
+                data, allocation, size = header_info
+                assert self.header_size == size
+                # Yield the header chunk
+                yield Chunk(data, size=size, allocation=allocation)
+
+        # Process the rest of the file using read
+        while True:
+            started_chunking = time.monotonic()
+            chunk_info = self.reader.read(self.block_size, return_chunk_info=True)
+            self.chunking_time += time.monotonic() - started_chunking
+
+            if chunk_info is None or chunk_info[2] == 0:
+                # End of file
+                break
+
+            # Unpack the chunk info
+            data, allocation, size = chunk_info
+
+            # Yield the chunk with the appropriate allocation type
+            yield Chunk(data, size=size, allocation=allocation)
 
 
 # Cyclic polynomial / buzhash