feat: add SystemVerilog backend generator (#1090)

This pull request adds support for generating a SystemVerilog package
containing RISC-V instruction and CSR definitions, and enhances the
instruction encoding logic to support a new schema format. The changes
include introducing a new generator script, updating the encoding
extraction logic to handle new formats, and integrating the
SystemVerilog generation into the build process.

**SystemVerilog Generation:**
- Added a new script `sverilog_generator.py` that generates a
SystemVerilog package (`riscv_decode_package.svh`) with instruction and
CSR encodings, including formatting and argument parsing for flexible
generation.
- Integrated a new Rake task (`gen:sverilog`) to invoke the
SystemVerilog generator, with options for configuration and output
directory.
- Created the `gen/sverilog` output directory as part of the build
process.

**Instruction Encoding Enhancements:**
- Added `build_match_from_format` function to extract encoding match
strings from the new `format` field in instruction definitions,
supporting opcodes with specific bit fields.
- Updated instruction loading logic to use the new
`build_match_from_format` when the legacy `encoding` field is missing,
improving compatibility with new schema definitions.
- Improved handling of RV32/RV64 instruction encodings to ensure both
variants are processed and named correctly.

**CSR Handling:**
- Simplified CSR naming logic by always using the uppercase name without
`.RV32` suffix, regardless of architecture.

---------

Signed-off-by: Afonso Oliveira <[email protected]>
Signed-off-by: Alieldin Alaa <[email protected]>
Co-authored-by: Afonso Oliveira <[email protected]>
Co-authored-by: Jordan Carlin <[email protected]>

Author: 3 people

.github/workflows/regress.yml

@@ -190,6 +190,17 @@ jobs:
         uses: ./.github/actions/singularity-setup
       - name: Generate c_header code
         run: ./do gen:c_header
+  regress-gen-sverilog:
+    runs-on: ubuntu-latest
+    env:
+      SINGULARITY: 1
+    steps:
+      - name: Clone Github Repo Action
+        uses: actions/checkout@v4
+      - name: singularity setup
+        uses: ./.github/actions/singularity-setup
+      - name: Generate sverilog_header code
+        run: ./do gen:sverilog
   regress-cpp-unit:
     runs-on: ubuntu-latest
     env:

backends/generators/c_header/generate_encoding.py

@@ -8,7 +8,6 @@
 import logging
 import argparse
 import yaml
-import json
 
 # Add parent directory to path to import generator.py
 parent_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
@@ -18,6 +17,7 @@
 from generator import (
     load_instructions,
     load_csrs,
+    load_exception_codes,
     parse_match,
     parse_extension_requirements,
 )
@@ -30,125 +30,6 @@ def calculate_mask(match_str):
     return int("".join("0" if c == "-" else "1" for c in match_str), 2)
 
 
-def load_exception_codes(
-    ext_dir, enabled_extensions=None, include_all=False, resolved_codes_file=None
-):
-    """Load exception codes from extension YAML files or pre-resolved JSON file."""
-    exception_codes = []
-    found_extensions = 0
-    found_files = 0
-
-    if enabled_extensions is None:
-        enabled_extensions = []
-
-    # If we have a resolved codes file, use it instead of processing YAML files
-    if resolved_codes_file and os.path.exists(resolved_codes_file):
-        try:
-            with open(resolved_codes_file, encoding="utf-8") as f:
-                resolved_codes = json.load(f)
-
-            for code in resolved_codes:
-                num = code.get("num")
-                name = code.get("name")
-                if num is not None and name is not None:
-                    sanitized_name = (
-                        name.lower()
-                        .replace(" ", "_")
-                        .replace("/", "_")
-                        .replace("-", "_")
-                    )
-                    exception_codes.append((num, sanitized_name))
-
-            logging.info(
-                f"Loaded {len(exception_codes)} pre-resolved exception codes from {resolved_codes_file}"
-            )
-
-            # Sort by exception code number and deduplicate
-            seen_nums = set()
-            unique_codes = []
-            for num, name in sorted(exception_codes, key=lambda x: x[0]):
-                if num not in seen_nums:
-                    seen_nums.add(num)
-                    unique_codes.append((num, name))
-
-            return unique_codes
-
-        except Exception as e:
-            logging.error(
-                f"Error loading resolved codes file {resolved_codes_file}: {e}"
-            )
-            # Fall back to processing YAML files
-
-    for dirpath, _, filenames in os.walk(ext_dir):
-        for fname in filenames:
-            if not fname.endswith(".yaml"):
-                continue
-
-            found_files += 1
-            path = os.path.join(dirpath, fname)
-
-            try:
-                with open(path, encoding="utf-8") as f:
-                    data = yaml.safe_load(f)
-
-                if not isinstance(data, dict) or data.get("kind") != "extension":
-                    continue
-
-                found_extensions += 1
-                ext_name = data.get("name", "unnamed")
-
-                # Skip extension filtering if include_all is True
-                if not include_all:
-                    # Filter by extension requirements
-                    definedBy = data.get("definedBy")
-                    if definedBy:
-                        meets_req = parse_extension_requirements(definedBy)
-                        if not meets_req(enabled_extensions):
-                            continue
-
-                    # Check if excluded
-                    excludedBy = data.get("excludedBy")
-                    if excludedBy:
-                        exclusion_check = parse_extension_requirements(excludedBy)
-                        if exclusion_check(enabled_extensions):
-                            continue
-
-                # Get exception codes
-                for code in data.get("exception_codes", []):
-                    num = code.get("num")
-                    name = code.get("name")
-
-                    if num is not None and name is not None:
-                        sanitized_name = (
-                            name.lower()
-                            .replace(" ", "_")
-                            .replace("/", "_")
-                            .replace("-", "_")
-                        )
-                        exception_codes.append((num, sanitized_name))
-
-            except Exception as e:
-                logging.error(f"Error processing file {path}: {e}")
-
-    if found_extensions > 0:
-        logging.info(
-            f"Found {found_extensions} extension definitions in {found_files} files"
-        )
-        logging.info(f"Added {len(exception_codes)} exception codes to the output")
-    else:
-        logging.warning(f"No extension definitions found in {ext_dir}")
-
-    # Sort by exception code number and deduplicate
-    seen_nums = set()
-    unique_codes = []
-    for num, name in sorted(exception_codes, key=lambda x: x[0]):
-        if num not in seen_nums:
-            seen_nums.add(num)
-            unique_codes.append((num, name))
-
-    return unique_codes
-
-
 def extract_instruction_fields(instructions):
     """Extract field names and their positions from instruction definitions."""
     field_dict = {}

backends/generators/generator.py

@@ -3,6 +3,7 @@
 import yaml
 import logging
 import pprint
+import json
 
 pp = pprint.PrettyPrinter(indent=2)
 logging.basicConfig(level=logging.INFO, format="%(levelname)s:: %(message)s")
@@ -326,14 +327,15 @@ def load_instructions(
 
                         # Process RV64 encoding
                         rv64_match = rv64_encoding.get("match")
+                        rv32_match = rv32_encoding.get("match")
+
                         if rv64_match:
                             instr_dict[name] = {
                                 "match": rv64_match
                             }  # RV64 gets the default name
 
-                        # Process RV32 encoding with a _rv32 suffix
-                        rv32_match = rv32_encoding.get("match")
-                        if rv32_match:
+                        if rv32_match and rv32_match != rv64_match:
+                            # Process RV32 encoding with a _rv32 suffix
                             instr_dict[f"{name}_rv32"] = {"match": rv32_match}
 
                         continue  # Skip the rest of the loop as we've already added the encodings
@@ -491,11 +493,7 @@ def load_csrs(csr_root, enabled_extensions, include_all=False, target_arch="RV64
                 else:
                     addr_int = int(addr_to_use, 0)
 
-                # For BOTH architecture, add suffix to RV32-specific CSRs
-                if target_arch == "BOTH" and base == 32:
-                    csrs[addr_int] = f"{name.upper()}.RV32"
-                else:
-                    csrs[addr_int] = name.upper()
+                csrs[addr_int] = name.upper()
             except Exception as e:
                 logging.error(f"Error parsing address {addr_to_use} in {path}: {e}")
                 address_errors += 1
@@ -518,6 +516,124 @@ def load_csrs(csr_root, enabled_extensions, include_all=False, target_arch="RV64
     return csrs
 
 
+def load_exception_codes(
+    ext_dir, enabled_extensions=None, include_all=False, resolved_codes_file=None
+):
+    """Load exception codes from extension YAML files or pre-resolved JSON file."""
+    exception_codes = []
+    found_extensions = 0
+    found_files = 0
+
+    if enabled_extensions is None:
+        enabled_extensions = []
+    # If we have a resolved codes file, use it instead of processing YAML files
+    if resolved_codes_file and os.path.exists(resolved_codes_file):
+        try:
+            with open(resolved_codes_file, encoding="utf-8") as f:
+                resolved_codes = json.load(f)
+
+            for code in resolved_codes:
+                num = code.get("num")
+                name = code.get("name")
+                if num is not None and name is not None:
+                    sanitized_name = (
+                        name.lower()
+                        .replace(" ", "_")
+                        .replace("/", "_")
+                        .replace("-", "_")
+                    )
+                    exception_codes.append((num, sanitized_name))
+
+            logging.info(
+                f"Loaded {len(exception_codes)} pre-resolved exception codes from {resolved_codes_file}"
+            )
+
+            # Sort by exception code number and deduplicate
+            seen_nums = set()
+            unique_codes = []
+            for num, name in sorted(exception_codes, key=lambda x: x[0]):
+                if num not in seen_nums:
+                    seen_nums.add(num)
+                    unique_codes.append((num, name))
+
+            return unique_codes
+
+        except Exception as e:
+            logging.error(
+                f"Error loading resolved codes file {resolved_codes_file}: {e}"
+            )
+            # Fall back to processing YAML files
+
+    for dirpath, _, filenames in os.walk(ext_dir):
+        for fname in filenames:
+            if not fname.endswith(".yaml"):
+                continue
+
+            found_files += 1
+            path = os.path.join(dirpath, fname)
+
+            try:
+                with open(path, encoding="utf-8") as f:
+                    data = yaml.safe_load(f)
+
+                if not isinstance(data, dict) or data.get("kind") != "extension":
+                    continue
+
+                found_extensions += 1
+                ext_name = data.get("name", "unnamed")
+
+                # Skip extension filtering if include_all is True
+                if not include_all:
+                    # Filter by extension requirements
+                    definedBy = data.get("definedBy")
+                    if definedBy:
+                        meets_req = parse_extension_requirements(definedBy)
+                        if not meets_req(enabled_extensions):
+                            continue
+
+                    # Check if excluded
+                    excludedBy = data.get("excludedBy")
+                    if excludedBy:
+                        exclusion_check = parse_extension_requirements(excludedBy)
+                        if exclusion_check(enabled_extensions):
+                            continue
+
+                # Get exception codes
+                for code in data.get("exception_codes", []):
+                    num = code.get("num")
+                    name = code.get("name")
+
+                    if num is not None and name is not None:
+                        sanitized_name = (
+                            name.lower()
+                            .replace(" ", "_")
+                            .replace("/", "_")
+                            .replace("-", "_")
+                        )
+                        exception_codes.append((num, sanitized_name))
+
+            except Exception as e:
+                logging.error(f"Error processing file {path}: {e}")
+
+    if found_extensions > 0:
+        logging.info(
+            f"Found {found_extensions} extension definitions in {found_files} files"
+        )
+        logging.info(f"Added {len(exception_codes)} exception codes to the output")
+    else:
+        logging.warning(f"No extension definitions found in {ext_dir}")
+
+    # Sort by exception code number and deduplicate
+    seen_nums = set()
+    unique_codes = []
+    for num, name in sorted(exception_codes, key=lambda x: x[0]):
+        if num not in seen_nums:
+            seen_nums.add(num)
+            unique_codes.append((num, name))
+
+    return unique_codes
+
+
 def parse_match(match_str):
     """
     Convert the bit pattern string to an integer.