resolve linter errors

Signed-off-by: Wojciech Sipak <[email protected]>

Author: wsipak

xls/modules/zstd/README.md

@@ -28,18 +28,18 @@ decoder's `Memory Interface`.
 
 Once the decoding process is started, the decoder:
 
-1. Reads the configuration from the CSRs,
-2. Decodes the Frame Header,
-3. Decodes the Block Header,
-4. Decodes the Block Data with the correct processing unit picked based on the
-Block Type from the Block Header,
-5. Aggregates the processing unit results in the correct order into a stream
-and routes it to the history buffer,
-6. Assembles the data block outputs based on the history buffer contents and
-updates the history,
-7. Prepares the final output of the decoder and writes it to the memory,
-8. (Optional) Calculates checksum and compares it against the checksum read
-from the frame.[^2]
+1.  Reads the configuration from the CSRs,
+2.  Decodes the Frame Header,
+3.  Decodes the Block Header,
+4.  Decodes the Block Data with the correct processing unit picked based on the
+    Block Type from the Block Header,
+5.  Aggregates the processing unit results in the correct order into a stream
+    and routes it to the history buffer,
+6.  Assembles the data block outputs based on the history buffer contents and
+    updates the history,
+7.  Prepares the final output of the decoder and writes it to the memory,
+8.  (Optional) Calculates checksum and compares it against the checksum read
+    from the frame.[^2]
 
 ![brief data flow diagram of ZstdDecoder](img/ZSTD_decoder.png)
 
@@ -253,14 +253,14 @@ This stage receives data which is tagged either `literals` or `copy`. This stage
 will show the following behavior, depending on the tag:
 
 * `literals`
-    * Packet contents placed as newest in the history buffer,
-    * Packet contents copied to the decoder's output,
+    *   Packet contents placed as newest in the history buffer,
+    *   Packet contents copied to the decoder's output,
 * `copy`
-    * Wait for all previous writes to be completed,
-    * Copy `copy_length` literals starting `offset _length` from the newest in
-history buffer to the decoder's output,
-    * Copy `copy_length` literals starting `offset _length` from the newest in
-history buffer to the history buffer as the newest.
+    *   Wait for all previous writes to be completed,
+    *   Copy `copy_length` literals starting `offset _length` from the newest in
+        history buffer to the decoder's output,
+    *   Copy `copy_length` literals starting `offset _length` from the newest in
+        history buffer to the history buffer as the newest.
 
 ### Compressed block decoder architecture[^1] {#compressed-block-decoder-architecture1}
 
@@ -360,10 +360,10 @@ FSE decoder or to direct weight extraction.
 
 This stage performs multiple functions.
 
-1. It decodes and builds the FSE distribution table.
-2. It stores all remaining bitstream data.
-3. After receiving the last byte, it translates the bitstream to Huffman
-weights using 2 interleaved FSE streams.
+1.  It decodes and builds the FSE distribution table.
+2.  It stores all remaining bitstream data.
+3.  After receiving the last byte, it translates the bitstream to Huffman
+    weights using 2 interleaved FSE streams.
 
 ##### Direct weight decoder
 
@@ -468,8 +468,7 @@ Crossbar](https://github.com/alexforencich/verilog-axi).
 
 ![diagram of interfaces of decoder and its wrapper](img/ZSTD_decoder_wrapper.png)
 
-**Figure
-Zstd decoder wrapper connection diagram.**
+**Figure: Zstd decoder wrapper connection diagram.**
 
 Cocotb testbench interacts with the decoder with the help of a
 [cocotbext-axi](https://github.com/alexforencich/cocotbext-axi) extension that
@@ -545,19 +544,19 @@ calculated from current `WINDOW_LOG_MAX` (by default in Top Level Proc tests
 
 The `assert!()` or `fail!()` will occur in:
 
-* RawBlockDecoder
-  * Receive `BlockDataPacket` with `ID` different than the previous packet which
-    did not have the `last` flag set
-* DecoderMux
-  * At the beginning of the simulation or after receiving
-    `ExtendedBlockDataPacket` with `last` and `last_block` (decoding new ZSTD
-    frame) set receive on channels `raw_r`, `rle_r` and `cmp_r`
-    `ExtendedBlockDataPackets` without any of those having `ID==0`
-  * Receive `ExtendedBlockDataPacket` with a smaller `ID` than any of the
-    previously processed packets during the current ZSTD frame decoding
-* SequenceExecutor
-  * Receive `SequenceExecutorPacket` with `msg_type==SEQUENCE` and `content`
-    field with value: `0`
+*   RawBlockDecoder
+    *   Receive `BlockDataPacket` with `ID` different than the previous packet
+        which did not have the `last` flag set
+*   DecoderMux
+    *   At the beginning of the simulation or after receiving
+        `ExtendedBlockDataPacket` with `last` and `last_block` (decoding new ZSTD
+        frame) set receive on channels `raw_r`, `rle_r` and `cmp_r`
+        `ExtendedBlockDataPackets` without any of those having `ID==0`
+    *   Receive `ExtendedBlockDataPacket` with a smaller `ID` than any of the
+        previously processed packets during the current ZSTD frame decoding
+*   SequenceExecutor
+    *   Receive `SequenceExecutorPacket` with `msg_type==SEQUENCE` and `content`
+        field with value: `0`
 
 There are also several `impossible cases` covered by `fail!()`.
 Those are mostly enforced by the type checker for the `match` expressions to

xls/modules/zstd/cocotb/channel.py

@@ -37,7 +37,9 @@ class XLSChannel(Bus):
   _optional_signals = XLS_CHANNEL_OPTIONAL_SIGNALS
 
   def __init__(self, entity, name, clk, *, start_now=False, **kwargs: Any):
-    super().__init__(entity, name, self._signals, self._optional_signals, **kwargs)
+    super().__init__(
+      entity, name, self._signals, self._optional_signals, **kwargs
+    )
     self.clk = clk
     if start_now:
         self.start_recv_loop()
@@ -58,17 +60,30 @@ class XLSChannelDriver(BusDriver):
   _signals = XLS_CHANNEL_SIGNALS
   _optional_signals = XLS_CHANNEL_OPTIONAL_SIGNALS
 
-  def __init__(self, entity: SimHandleBase, name: str, clock: SimHandleBase, **kwargs: Any):
+  def __init__(
+    self,
+    entity: SimHandleBase,
+    name: str,
+    clock: SimHandleBase,
+    **kwargs: Any
+  ):
     BusDriver.__init__(self, entity, name, clock, **kwargs)
 
     self.bus.data.setimmediatevalue(0)
     self.bus.vld.setimmediatevalue(0)
 
-  async def _driver_send(self, transaction: Transaction, sync: bool = True, **kwargs: Any) -> None:
+  async def _driver_send(
+    self,
+    transaction: Transaction,
+    sync: bool = True,
+    **kwargs: Any
+  ) -> None:
     if sync:
       await RisingEdge(self.clock)
 
-    data_to_send = (transaction if isinstance(transaction, Sequence) else [transaction])
+    data_to_send = (
+      transaction if isinstance(transaction, Sequence) else [transaction]
+    )
 
     for word in data_to_send:
       self.bus.vld.value = 1
@@ -83,11 +98,18 @@ async def _driver_send(self, transaction: Transaction, sync: bool = True, **kwar
 
 
 class XLSChannelMonitor(BusMonitor):
-   """Monitors and decodes transactions on an XLS channel."""
+  """Monitors and decodes transactions on an XLS channel."""
   _signals = XLS_CHANNEL_SIGNALS
   _optional_signals = XLS_CHANNEL_OPTIONAL_SIGNALS
 
-  def __init__(self, entity: SimHandleBase, name: str, clock: SimHandleBase, struct: Type[XLSStruct], **kwargs: Any):
+  def __init__(
+    self,
+    entity: SimHandleBase,
+    name: str,
+    clock: SimHandleBase,
+    struct: Type[XLSStruct],
+    **kwargs: Any
+  ):
     BusMonitor.__init__(self, entity, name, clock, **kwargs)
     self.struct = struct
 

xls/modules/zstd/cocotb/data_generator.py

@@ -22,6 +22,8 @@
 import zstandard
 
 class BlockType(enum.Enum):
+  """Enum encoding of ZSTD block types."""
+
   RAW = 0
   RLE = 1
   COMPRESSED = 2
@@ -38,7 +40,9 @@ def from_string(s):
       raise ValueError(str(e)) from e
 
 def CallDecodecorpus(args):
-  decodecorpus = pathlib.Path(runfiles.get_path("decodecorpus", repository = "zstd"))
+  decodecorpus = pathlib.Path(
+    runfiles.get_path("decodecorpus", repository = "zstd")
+  )
   cmd = args
   cmd.insert(0, str(decodecorpus))
   cmd_concat = " ".join(cmd)

xls/modules/zstd/cocotb/memory.py

@@ -13,8 +13,7 @@
 # limitations under the License.
 
 """
-Extensions of the cocotb AXI RAM models that initialize
-memory contents from a binary file.
+Extensions of the cocotb AXI RAM models to memory contents from a binary file.
 """
 
 import os

xls/modules/zstd/cocotb/scoreboard.py

@@ -60,7 +60,7 @@ def _current_cycle(self):
   def _req_callback(self, transaction: XLSStruct):
     self.pending_req.put(LatencyQueueItem(transaction, self._current_cycle()))
 
-  def _resp_callback(self, _transaction: XLSStruct):
+  def _resp_callback(self, unused_transaction: XLSStruct):
     latency_item = self.pending_req.get()
     self.results.append(self._current_cycle() - latency_item.timestamp)
 

xls/modules/zstd/cocotb/utils.py

@@ -26,8 +26,12 @@
 
 
 def setup_com_iverilog():
-  iverilog_path = pathlib.Path(runfiles.get_path("iverilog", repository = "com_icarus_iverilog"))
-  vvp_path = pathlib.Path(runfiles.get_path("vvp", repository = "com_icarus_iverilog"))
+  iverilog_path = pathlib.Path(
+    runfiles.get_path("iverilog", repository = "com_icarus_iverilog")
+  )
+  vvp_path = pathlib.Path(
+    runfiles.get_path("vvp", repository = "com_icarus_iverilog")
+  )
   os.environ["PATH"] += os.pathsep + str(iverilog_path.parent)
   os.environ["PATH"] += os.pathsep + str(vvp_path.parent)
   build_dir = pathlib.Path(os.environ['BUILD_WORKING_DIRECTORY'], "sim_build")

xls/modules/zstd/cocotb/xlsstruct.py

@@ -17,10 +17,7 @@
 """
 
 import random
-from dataclasses import asdict
-from dataclasses import astuple
-from dataclasses import dataclass
-from dataclasses import fields
+import dataclasses
 
 from cocotb.binary import BinaryValue
 
@@ -42,21 +39,23 @@ class MyStruct(XLSStruct):
   Returns:
     type: The dataclass-decorated class with repr disabled.
   """
-  return dataclass(cls, repr=False)
+  return dataclasses.dataclass(cls, repr=False)
 
-@dataclass
+@dataclasses.dataclass
 class XLSStruct:
-  """
-  Represents XLS struct on the Python side, allowing serialization/deserialization
-  to/from common formats and usage with XLS{Driver, Monitor}.
+  """Represents XLS struct on the Python side.
+
+  Allows serialization/deserialization to/from common formats and usage with
+  XLS{Driver, Monitor}.
 
-  The intended way to use this class is to inherit from it, specify the fields with
-  <field>: <width> [= <default value>] syntax and decorate the inheriting class with
-  @XLSDataclass. Objects of this class can be instantiated and used like usual
-  dataclass objects, with a few extra methods and properties available. They can also
-  be passed as arguments to XLSChannelDriver.send and will be serialized to expected
-  bit vector. Class can be passed to XLSChannelMonitor ``struct`` constructor argument
-  to automatically deserialize all transfers to the provided struct.
+  The intended way to use this class is to inherit from it, specify the fields
+  with <field>: <width> [= <default value>] syntax and decorate the inheriting
+  class with @XLSDataclass. Objects of this class can be instantiated and used
+  like usual dataclass objects, with a few extra methods and properties
+  available. They can also be passed as arguments to XLSChannelDriver.send and
+  will be serialized to expected bit vector. Class can be passed to
+  XLSChannelMonitor ``struct`` constructor argument to automatically
+  deserialize all transfers to the provided struct.
 
   Example:
 
@@ -88,7 +87,7 @@ def _masks(cls):
     returns [2'b11, 3'b111, 4'b1111].
     """
     masks = []
-    for field in fields(cls):
+    for field in dataclasses.fields(cls):
       width = field.type
       masks += [(1 << width) - 1]
     return masks
@@ -101,7 +100,7 @@ def _positions(cls):
     returns [20, 18, 15, 11, 6, 0]
     """
     positions = []
-    for i, field in enumerate(fields(cls)):
+    for i, field in enumerate(dataclasses.fields(cls)):
       width = field.type
       if i == 0:
         positions += [cls.total_width - width]
@@ -113,17 +112,18 @@ def _positions(cls):
   @property
   def total_width(cls):
     """Returns total bit width of the struct."""
-    return sum(field.type for field in fields(cls))
+    return sum(field.type for field in dataclasses.fields(cls))
 
   @property
   def value(self):
     """Returns struct's value as a Python integer."""
     value = 0
     masks = self._masks()
     positions = self._positions()
-    for field_val, mask, pos in zip(astuple(self), masks, positions):
+    for field_val, mask, pos in zip(
+        dataclasses.astuple(self), masks, positions):
       if field_val > mask:
-        raise TruncationError(f"Signal value is wider than its bit width")
+        raise TruncationError("Signal value is wider than its bit width")
       value |= (field_val & mask) << pos
     return value
 
@@ -151,15 +151,15 @@ def from_int(cls, value):
     instance = {}
     masks = cls._masks()
     positions = cls._positions()
-    for field, mask, pos in zip(fields(cls), masks, positions):
+    for field, mask, pos in zip(dataclasses.fields(cls), masks, positions):
       instance[field.name] = (value >> pos) & mask
     return cls(**instance)
 
   @classmethod
   def randomize(cls):
     """Returns an instance of the struct with all fields' values randomized."""
     instance = {}
-    for field in fields(cls):
+    for field in dataclasses.fields(cls):
       instance[field.name] = random.randrange(0, 2**field.type)
     return cls(**instance)
 
@@ -168,5 +168,7 @@ def __str__(self):
 
   def __repr__(self):
     classname = self.__class__.__name__
-    class_fields = [f"{name}={hex(value)}" for name, value in asdict(self).items()]
+    class_fields = [
+      f"{name}={hex(value)}" for name, value in dataclasses.asdict(self).items()
+    ]
     return f"{classname}({', '.join(class_fields)})"