WIP

.cargo/config.toml

@@ -1,2 +0,0 @@
-[target.x86_64-unknown-linux-gnu]
-rustflags = ["-C", "link-arg=-fuse-ld=lld"]

.clang-format

@@ -1,32 +0,0 @@
-BasedOnStyle: LLVM
-
-Language: Cpp
-Standard: Cpp11
-
-AccessModifierOffset: -1
-AlignEscapedNewlines: Left
-AlwaysBreakAfterReturnType: None
-AlwaysBreakTemplateDeclarations: Yes
-AllowAllParametersOfDeclarationOnNextLine: false
-AllowShortFunctionsOnASingleLine: Inline
-BinPackParameters: false
-BreakBeforeBinaryOperators: NonAssignment
-BreakBeforeTernaryOperators: true
-BreakConstructorInitializers: AfterColon
-BreakInheritanceList: AfterColon
-ColumnLimit: 120
-ConstructorInitializerAllOnOneLineOrOnePerLine: true
-ContinuationIndentWidth: 4
-DerivePointerAlignment: false
-IndentCaseLabels: true
-IndentPPDirectives: None
-IndentWidth: 4
-MaxEmptyLinesToKeep: 1
-PointerAlignment: Left
-ReflowComments: true
-SortIncludes: false
-SortUsingDeclarations: true
-SpaceAfterTemplateKeyword: false
-SpacesBeforeTrailingComments: 2
-TabWidth: 4
-UseTab: Never

.gitignore

@@ -1,4 +1,3 @@
 __pycache__
 /target
-/nac3standalone/demo/linalg/target
 nix/windows/msys2

.pre-commit-config.yaml

@@ -1,24 +1,24 @@
 # See https://pre-commit.com for more information
 # See https://pre-commit.com/hooks.html for more hooks
 
-default_stages: [pre-commit]
+default_stages: [commit]
 
 repos:
   - repo: local
     hooks:
       - id: nac3-cargo-fmt
         name: nac3 cargo format
-        entry: nix
+        entry: cargo
         language: system
         types: [file, rust]
         pass_filenames: false
         description: Runs cargo fmt on the codebase.
-        args: [develop, -c, cargo, fmt, --all]
+        args: [fmt]
       - id: nac3-cargo-clippy
         name: nac3 cargo clippy
-        entry: nix
+        entry: cargo
         language: system
         types: [file, rust]
         pass_filenames: false
         description: Runs cargo clippy on the codebase.
-        args: [develop, -c, cargo, clippy, --tests]
+        args: [clippy, --tests]

Cargo.toml

@@ -4,7 +4,6 @@ members = [
   "nac3ast",
   "nac3parser",
   "nac3core",
-  "nac3core/nac3core_derive",
   "nac3standalone",
   "nac3artiq",
   "runkernel",

flake.lock

@@ -2,11 +2,11 @@
   "nodes": {
     "nixpkgs": {
       "locked": {
-        "lastModified": 1744098102,
-        "narHash": "sha256-tzCdyIJj9AjysC3OuKA+tMD/kDEDAF9mICPDU7ix0JA=",
+        "lastModified": 1720418205,
+        "narHash": "sha256-cPJoFPXU44GlhWg4pUk9oUPqurPlCFZ11ZQPk21GTPU=",
         "owner": "NixOS",
         "repo": "nixpkgs",
-        "rev": "c8cd81426f45942bb2906d5ed2fe21d2f19d95b7",
+        "rev": "655a58a72a6601292512670343087c2d75d859c1",
         "type": "github"
       },
       "original": {

flake.nix

@@ -6,32 +6,16 @@
   outputs = { self, nixpkgs }:
     let
       pkgs = import nixpkgs { system = "x86_64-linux"; };
-      pkgs32 = import nixpkgs { system = "i686-linux"; };
     in rec {
       packages.x86_64-linux = rec {
         llvm-nac3 = pkgs.callPackage ./nix/llvm {};
         llvm-tools-irrt = pkgs.runCommandNoCC "llvm-tools-irrt" {}
           ''
           mkdir -p $out/bin
-          ln -s ${pkgs.llvmPackages_16.clang-unwrapped}/bin/clang $out/bin/clang-irrt
-          ln -s ${pkgs.llvmPackages_16.llvm.out}/bin/llvm-as $out/bin/llvm-as-irrt
+          ln -s ${pkgs.llvmPackages_14.clang-unwrapped}/bin/clang $out/bin/clang-irrt
+          ln -s ${pkgs.llvmPackages_14.clang}/bin/clang $out/bin/clang-irrt-test
+          ln -s ${pkgs.llvmPackages_14.llvm.out}/bin/llvm-as $out/bin/llvm-as-irrt
           '';
-        demo-linalg-stub = pkgs.rustPlatform.buildRustPackage {
-          name = "demo-linalg-stub";
-          src = ./nac3standalone/demo/linalg;
-          cargoLock = {
-            lockFile = ./nac3standalone/demo/linalg/Cargo.lock;
-          };
-          doCheck = false;
-        };
-        demo-linalg-stub32 = pkgs32.rustPlatform.buildRustPackage {
-          name = "demo-linalg-stub32";
-          src = ./nac3standalone/demo/linalg;
-          cargoLock = {
-            lockFile = ./nac3standalone/demo/linalg/Cargo.lock;
-          };
-          doCheck = false;
-        };
         nac3artiq = pkgs.python3Packages.toPythonModule (
           pkgs.rustPlatform.buildRustPackage rec {
             name = "nac3artiq";
@@ -39,12 +23,10 @@
             src = self;
             cargoLock = {
               lockFile = ./Cargo.lock;
-              outputHashes = {
-                "inkwell-0.5.0" = "sha256-Qsqy/fhD/qK0rKBeXetE99vW9G9XAePxlXv0An3Yeuo=";
-              };
             };
+            cargoTestFlags = [ "--features" "test" ];
             passthru.cargoLock = cargoLock;
-            nativeBuildInputs = [ pkgs.python3 (pkgs.wrapClangMulti pkgs.llvmPackages_16.clang) llvm-tools-irrt pkgs.llvmPackages_16.llvm.out pkgs.llvmPackages_16.bintools llvm-nac3 ];
+            nativeBuildInputs = [ pkgs.python3 pkgs.llvmPackages_14.clang llvm-tools-irrt pkgs.llvmPackages_14.llvm.out llvm-nac3 ];
             buildInputs = [ pkgs.python3 llvm-nac3 ];
             checkInputs = [ (pkgs.python3.withPackages(ps: [ ps.numpy ps.scipy ])) ];
             checkPhase =
@@ -52,9 +34,7 @@
               echo "Checking nac3standalone demos..."
               pushd nac3standalone/demo
               patchShebangs .
-              export DEMO_LINALG_STUB=${demo-linalg-stub}/lib/liblinalg.a
-              export DEMO_LINALG_STUB32=${demo-linalg-stub32}/lib/liblinalg.a
-              ./check_demos.sh -i686
+              ./check_demos.sh
               popd
               echo "Running Cargo tests..."
               cargoCheckHook
@@ -80,7 +60,7 @@
 
         # LLVM PGO support
         llvm-nac3-instrumented = pkgs.callPackage ./nix/llvm {
-          stdenv = pkgs.llvmPackages_16.stdenv;
+          stdenv = pkgs.llvmPackages_14.stdenv;
           extraCmakeFlags = [ "-DLLVM_BUILD_INSTRUMENTED=IR" ];
         };
         nac3artiq-instrumented = pkgs.python3Packages.toPythonModule (
@@ -88,13 +68,13 @@
             name = "nac3artiq-instrumented";
             src = self;
             inherit (nac3artiq) cargoLock;
-            nativeBuildInputs = [ pkgs.python3 packages.x86_64-linux.llvm-tools-irrt pkgs.llvmPackages_16.bintools llvm-nac3-instrumented ];
+            nativeBuildInputs = [ pkgs.python3 packages.x86_64-linux.llvm-tools-irrt llvm-nac3-instrumented ];
             buildInputs = [ pkgs.python3 llvm-nac3-instrumented ];
             cargoBuildFlags = [ "--package" "nac3artiq" "--features" "init-llvm-profile" ];
             doCheck = false;
             configurePhase =
               ''
-              export CARGO_TARGET_X86_64_UNKNOWN_LINUX_GNU_RUSTFLAGS="-C link-arg=-L${pkgs.llvmPackages_16.compiler-rt}/lib/linux -C link-arg=-lclang_rt.profile-x86_64"
+              export CARGO_TARGET_X86_64_UNKNOWN_LINUX_GNU_RUSTFLAGS="-C link-arg=-L${pkgs.llvmPackages_14.compiler-rt}/lib/linux -C link-arg=-lclang_rt.profile-x86_64"
               '';
             installPhase =
               ''
@@ -110,20 +90,19 @@
             (pkgs.fetchFromGitHub {
               owner = "m-labs";
               repo = "sipyco";
-              rev = "094a6cd63ffa980ef63698920170e50dc9ba77fd";
-              sha256 = "sha256-PPnAyDedUQ7Og/Cby9x5OT9wMkNGTP8GS53V6N/dk4w=";
+              rev = "939f84f9b5eef7efbf7423c735d1834783b6140e";
+              sha256 = "sha256-15Nun4EY35j+6SPZkjzZtyH/ncxLS60KuGJjFh5kSTc=";
             })
             (pkgs.fetchFromGitHub {
               owner = "m-labs";
               repo = "artiq";
-              rev = "554b0749ca5985bf4d006c4f29a05e83de0a226d";
-              sha256 = "sha256-3eSNHTSlmdzLMcEMIspxqjmjrcQe4aIGqIfRgquUg18=";
+              rev = "923ca3377d42c815f979983134ec549dc39d3ca0";
+              sha256 = "sha256-oJoEeNEeNFSUyh6jXG8Tzp6qHVikeHS0CzfE+mODPgw=";
             })
           ];
           buildInputs = [
-            (python3-mimalloc.withPackages(ps: [ ps.numpy ps.scipy ps.jsonschema ps.lmdb ps.platformdirs nac3artiq-instrumented ]))
-            pkgs.llvmPackages_16.llvm.out
-            pkgs.llvmPackages_16.bintools
+            (python3-mimalloc.withPackages(ps: [ ps.numpy ps.scipy ps.jsonschema ps.lmdb nac3artiq-instrumented ]))
+            pkgs.llvmPackages_14.llvm.out
           ];
           phases = [ "buildPhase" "installPhase" ];
           buildPhase =
@@ -143,7 +122,7 @@
             '';
         };
         llvm-nac3-pgo = pkgs.callPackage ./nix/llvm {
-          stdenv = pkgs.llvmPackages_16.stdenv;
+          stdenv = pkgs.llvmPackages_14.stdenv;
           extraCmakeFlags = [ "-DLLVM_PROFDATA_FILE=${nac3artiq-profile}/llvm.profdata" ];
         };
         nac3artiq-pgo = pkgs.python3Packages.toPythonModule (
@@ -151,7 +130,7 @@
             name = "nac3artiq-pgo";
             src = self;
             inherit (nac3artiq) cargoLock;
-            nativeBuildInputs = [ pkgs.python3 packages.x86_64-linux.llvm-tools-irrt pkgs.llvmPackages_16.bintools llvm-nac3-pgo ];
+            nativeBuildInputs = [ pkgs.python3 packages.x86_64-linux.llvm-tools-irrt llvm-nac3-pgo ];
             buildInputs = [ pkgs.python3 llvm-nac3-pgo ];
             cargoBuildFlags = [ "--package" "nac3artiq" ];
             cargoTestFlags = [ "--package" "nac3ast" "--package" "nac3parser" "--package" "nac3core" "--package" "nac3artiq" ];
@@ -172,24 +151,22 @@
         buildInputs = with pkgs; [
           # build dependencies
           packages.x86_64-linux.llvm-nac3
-          (pkgs.wrapClangMulti llvmPackages_16.clang) llvmPackages_16.llvm.out llvmPackages_16.bintools  # for running nac3standalone demos
+          llvmPackages_14.clang llvmPackages_14.llvm.out  # for running nac3standalone demos
           packages.x86_64-linux.llvm-tools-irrt
           cargo
           rustc
           # runtime dependencies
-          lld_16                           # for running kernels on the host
+          lld_14                           # for running kernels on the host
           (packages.x86_64-linux.python3-mimalloc.withPackages(ps: [ ps.numpy ps.scipy ]))
           # development tools
           cargo-insta
           clippy
           pre-commit
           rustfmt
+          rust-analyzer
         ];
-        shellHook =
-          ''
-          export DEMO_LINALG_STUB=${packages.x86_64-linux.demo-linalg-stub}/lib/liblinalg.a
-          export DEMO_LINALG_STUB32=${packages.x86_64-linux.demo-linalg-stub32}/lib/liblinalg.a
-          '';
+        # https://nixos.wiki/wiki/Rust#Shell.nix_example
+        RUST_SRC_PATH = "${pkgs.rust.packages.stable.rustPlatform.rustLibSrc}";
       };
       devShells.x86_64-linux.msys2 = pkgs.mkShell {
         name = "nac3-dev-shell-msys2";

nac3artiq/Cargo.toml

@@ -2,21 +2,25 @@
 name = "nac3artiq"
 version = "0.1.0"
 authors = ["M-Labs"]
-edition = "2024"
+edition = "2021"
 
 [lib]
 name = "nac3artiq"
 crate-type = ["cdylib"]
 
 [dependencies]
-indexmap = "2.8"
-itertools = "0.14"
-pyo3 = { version = "0.24", features = ["extension-module"] }
+itertools = "0.13"
+pyo3 = { version = "0.21", features = ["extension-module", "gil-refs"] }
 parking_lot = "0.12"
-tempfile = "3.19"
+tempfile = "3.10"
+nac3parser = { path = "../nac3parser" }
 nac3core = { path = "../nac3core" }
 nac3ld = { path = "../nac3ld" }
 
+[dependencies.inkwell]
+version = "0.4"
+default-features = false
+features = ["llvm14-0", "target-x86", "target-arm", "target-riscv", "no-libffi-linking"]
+
 [features]
 init-llvm-profile = []
-no-escape-analysis = ["nac3core/no-escape-analysis"]

nac3artiq/demo/context_managers.py

@@ -1,84 +0,0 @@
-from numpy import int32
-from min_artiq import *
-
-@nac3
-class PassContextManager:
-    @kernel
-    def __init__(self):
-        pass
-
-    @kernel
-    def __enter__(self):
-        pass
-
-    @kernel
-    def __exit__(self):
-        pass
-
-@nac3
-class CustomDataContextManager:
-    a: Kernel[int32]
-    b: Kernel[int32]
-
-    @kernel
-    def __init__(self, a: int32, b: int32):
-        self.a = a
-        self.b = b
-
-    @kernel
-    def __enter__(self):
-        self.a += 1
-        print_int32(self.a)
-
-    @kernel
-    def __exit__(self):
-        self.b += 1
-        print_int32(self.b)
-
-@nac3
-class WithAsContextManager:
-    a: Kernel[int32]
-
-    @kernel
-    def __init__(self, val: int32):
-        self.a = val
-
-    @kernel
-    def __enter__(self) -> int32:
-        print_int32(self.a)
-        return self.a
-
-    @kernel
-    def __exit__(self):
-        print_int32(self.a)
-
-@nac3
-class CtxMgrTest:
-    core: KernelInvariant[Core]
-
-    def __init__(self):
-        self.core = Core()
-
-    @kernel
-    def run(self):
-        x = 0
-        a = PassContextManager()
-        with a:
-            x += 1
-        
-        h = CustomDataContextManager(1, 2)
-        with h:
-            x += h.a + h.b
-
-        with WithAsContextManager(15) as num:
-            x += num
-
-        with WithAsContextManager(10) as c, WithAsContextManager(5) as d:
-            e: int32 = c + d
-            x += e
-
-        print_int32(x)
-        return
-
-if __name__ == "__main__":
-    CtxMgrTest().run()

nac3artiq/demo/demo_import_mod.py

@@ -1,27 +0,0 @@
-from min_artiq import kernel, KernelInvariant, nac3
-import min_artiq as artiq
-
-
-@nac3
-class Demo:
-    core: KernelInvariant[artiq.Core]
-    led0: KernelInvariant[artiq.TTLOut]
-    led1: KernelInvariant[artiq.TTLOut]
-
-    def __init__(self):
-        self.core = artiq.Core()
-        self.led0 = artiq.TTLOut(self.core, 18)
-        self.led1 = artiq.TTLOut(self.core, 19)
-
-    @kernel
-    def run(self):
-        self.core.reset()
-        while True:
-            with artiq.parallel:
-                self.led0.pulse(100.*artiq.ms)
-                self.led1.pulse(100.*artiq.ms)
-            self.core.delay(100.*artiq.ms)
-
-
-if __name__ == "__main__":
-    Demo().run()

nac3artiq/demo/embedding_map.py

@@ -0,0 +1,66 @@
+class EmbeddingMap:
+    def __init__(self):
+        self.object_inverse_map = {}
+        self.object_map = {}
+        self.string_map = {}
+        self.string_reverse_map = {}
+        self.function_map = {}
+        self.attributes_writeback = []
+
+        # preallocate exception names
+        self.preallocate_runtime_exception_names(["RuntimeError",
+                                          "RTIOUnderflow",
+                                          "RTIOOverflow",
+                                          "RTIODestinationUnreachable",
+                                          "DMAError",
+                                          "I2CError",
+                                          "CacheError",
+                                          "SPIError",
+                                          "0:ZeroDivisionError",
+                                          "0:IndexError",
+                                          "0:ValueError",
+                                          "0:RuntimeError",
+                                          "0:AssertionError",
+                                          "0:KeyError",
+                                          "0:NotImplementedError",
+                                          "0:OverflowError",
+                                          "0:IOError",
+                                          "0:UnwrapNoneError"])
+
+    def preallocate_runtime_exception_names(self, names):
+        for i, name in enumerate(names):
+            if ":" not in name:
+                name = "0:artiq.coredevice.exceptions." + name
+            exn_id = self.store_str(name)
+            assert exn_id == i
+
+    def store_function(self, key, fun):
+        self.function_map[key] = fun
+        return key
+
+    def store_object(self, obj):
+        obj_id = id(obj)
+        if obj_id in self.object_inverse_map:
+            return self.object_inverse_map[obj_id]
+        key = len(self.object_map) + 1
+        self.object_map[key] = obj
+        self.object_inverse_map[obj_id] = key
+        return key
+
+    def store_str(self, s):
+        if s in self.string_reverse_map:
+            return self.string_reverse_map[s]
+        key = len(self.string_map)
+        self.string_map[key] = s
+        self.string_reverse_map[s] = key
+        return key
+
+    def retrieve_function(self, key):
+        return self.function_map[key]
+
+    def retrieve_object(self, key):
+        return self.object_map[key]
+
+    def retrieve_str(self, key):
+        return self.string_map[key]
+

nac3artiq/demo/min_artiq.py

@@ -1,11 +1,12 @@
 from inspect import getfullargspec
 from functools import wraps
+from types import SimpleNamespace
+from numpy import int32, int64
+from typing import Generic, TypeVar
 from math import floor, ceil
-from numpy import int32, int64, uint32, uint64, float64, bool_, str_, ndarray
-from types import GenericAlias, ModuleType, SimpleNamespace
-from typing import _GenericAlias, Generic, TypeVar
 
 import nac3artiq
+from embedding_map import EmbeddingMap
 
 
 __all__ = [
@@ -16,7 +17,7 @@ __all__ = [
     "rpc", "ms", "us", "ns",
     "print_int32", "print_int64",
     "Core", "TTLOut",
-    "parallel", "legacy_parallel", "sequential"
+    "parallel", "sequential"
 ]
 
 
@@ -40,10 +41,10 @@ class Option(Generic[T]):
 
     def is_none(self):
         return self._nac3_option is None
-
+    
     def is_some(self):
         return not self.is_none()
-
+    
     def unwrap(self):
         if self.is_none():
             raise UnwrapNoneError()
@@ -54,7 +55,7 @@ class Option(Generic[T]):
             return "none"
         else:
             return "Some({})".format(repr(self._nac3_option))
-
+    
     def __str__(self) -> str:
         if self.is_none():
             return "none"
@@ -85,46 +86,13 @@ def ceil64(x):
 import device_db
 core_arguments = device_db.device_db["core"]["arguments"]
 
-builtins = {
-    "int": int,
-    "float": float,
-    "bool": bool,
-    "str": str,
-    "list": list,
-    "tuple": tuple,
-    "Exception": Exception,
-
-    "types": {
-        "GenericAlias": GenericAlias,
-        "ModuleType": ModuleType,
-    },
-
-    "typing": {
-        "_GenericAlias": _GenericAlias,
-        "TypeVar": TypeVar,
-    },
-
-    "numpy": {
-        "int32": int32,
-        "int64": int64,
-        "uint32": uint32,
-        "uint64": uint64,
-        "float64": float64,
-        "bool_": bool_,
-        "str_": str_,
-        "ndarray": ndarray,
-    },
-
-    "artiq": {
-        "Kernel": Kernel,
-        "KernelInvariant": KernelInvariant,
-        "_ConstGenericMarker": _ConstGenericMarker,
-        "none": none,
-        "virtual": virtual,
-        "Option": Option,
-    },
+artiq_builtins = {
+    "none": none,
+    "virtual": virtual,
+    "_ConstGenericMarker": _ConstGenericMarker,
+    "Option": Option,
 }
-compiler = nac3artiq.NAC3(core_arguments["target"], builtins)
+compiler = nac3artiq.NAC3(core_arguments["target"], artiq_builtins)
 allow_registration = True
 # Delay NAC3 analysis until all referenced variables are supposed to exist on the CPython side.
 registered_functions = set()
@@ -144,15 +112,10 @@ def extern(function):
     register_function(function)
     return function
 
-
-def rpc(arg=None, flags={}):
-    """Decorates a function or method to be executed on the host interpreter."""
-    if arg is None:
-        def inner_decorator(function):
-            return rpc(function, flags)
-        return inner_decorator
-    register_function(arg)
-    return arg
+def rpc(function):
+    """Decorates a function declaration defined by the core device runtime."""
+    register_function(function)
+    return function
 
 def kernel(function_or_method):
     """Decorates a function or method to be executed on the core device."""
@@ -185,9 +148,9 @@ def nac3(cls):
     return cls
 
 
-ms: KernelInvariant[float] = 1e-3
-us: KernelInvariant[float] = 1e-6
-ns: KernelInvariant[float] = 1e-9
+ms = 1e-3
+us = 1e-6
+ns = 1e-9
 
 @extern
 def rtio_init():
@@ -225,46 +188,6 @@ def print_int64(x: int64):
     raise NotImplementedError("syscall not simulated")
 
 
-class EmbeddingMap:
-    def __init__(self):
-        self.object_inverse_map = {}
-        self.object_map = {}
-        self.string_map = {}
-        self.string_reverse_map = {}
-        self.function_map = {}
-        self.attributes_writeback = []
-
-    def store_function(self, key, fun):
-        self.function_map[key] = fun
-        return key
-
-    def store_object(self, obj):
-        obj_id = id(obj)
-        if obj_id in self.object_inverse_map:
-            return self.object_inverse_map[obj_id]
-        key = len(self.object_map) + 1
-        self.object_map[key] = obj
-        self.object_inverse_map[obj_id] = key
-        return key
-
-    def store_str(self, s):
-        if s in self.string_reverse_map:
-            return self.string_reverse_map[s]
-        key = len(self.string_map)
-        self.string_map[key] = s
-        self.string_reverse_map[s] = key
-        return key
-
-    def retrieve_function(self, key):
-        return self.function_map[key]
-
-    def retrieve_object(self, key):
-        return self.object_map[key]
-
-    def retrieve_str(self, key):
-        return self.string_map[key]
-
-
 @nac3
 class Core:
     ref_period: KernelInvariant[float]
@@ -278,7 +201,7 @@ class Core:
         embedding = EmbeddingMap()
 
         if allow_registration:
-            compiler.analyze(registered_functions, registered_classes, special_ids, set())
+            compiler.analyze(registered_functions, registered_classes)
             allow_registration = False
 
         if hasattr(method, "__self__"):
@@ -368,12 +291,5 @@ class UnwrapNoneError(Exception):
     """raised when unwrapping a none value"""
     artiq_builtin = True
 
-parallel: KernelInvariant[KernelContextManager] = KernelContextManager()
-legacy_parallel: KernelInvariant[KernelContextManager] = KernelContextManager()
-sequential: KernelInvariant[KernelContextManager] = KernelContextManager()
-
-special_ids = {
-    "parallel": id(parallel),
-    "legacy_parallel": id(legacy_parallel),
-    "sequential": id(sequential),
-}
+parallel = KernelContextManager()
+sequential = KernelContextManager()

nac3artiq/demo/module.py

@@ -1,26 +0,0 @@
-from min_artiq import *
-from numpy import int32
-
-# Global Variable Definition
-X: Kernel[int32] = 1
-
-# TopLevelFunction Defintion
-@kernel
-def display_X():
-    print_int32(X)
-
-# TopLevel Class Definition
-@nac3
-class A:    
-    @kernel
-    def __init__(self):
-        self.set_x(1)
-    
-    @kernel
-    def set_x(self, new_val: int32):
-        global X
-        X = new_val
-    
-    @kernel
-    def get_X(self) -> int32:
-        return X

nac3artiq/demo/module_support.py

@@ -1,26 +0,0 @@
-from min_artiq import *
-import module as module_definition
-
-@nac3
-class TestModuleSupport:
-    core: KernelInvariant[Core]
-
-    def __init__(self):
-        self.core = Core()
-
-    @kernel
-    def run(self):
-        # Accessing classes
-        obj = module_definition.A()
-        obj.get_X()
-        obj.set_x(2)
-        
-        # Calling functions
-        module_definition.display_X()
-
-        # Updating global variables
-        module_definition.X = 9
-        module_definition.display_X()
-
-if __name__ == "__main__":
-    TestModuleSupport().run()

nac3artiq/demo/numpy_primitives_decay.py

@@ -1,29 +0,0 @@
-from min_artiq import *
-import numpy
-from numpy import int32
-
-
-@nac3
-class NumpyBoolDecay:
-    core: KernelInvariant[Core]
-    np_true: KernelInvariant[bool]
-    np_false: KernelInvariant[bool]
-    np_int: KernelInvariant[int32]
-    np_float: KernelInvariant[float]
-    np_str: KernelInvariant[str]
-
-    def __init__(self):
-        self.core = Core()
-        self.np_true = numpy.True_
-        self.np_false = numpy.False_
-        self.np_int = numpy.int32(0)
-        self.np_float = numpy.float64(0.0)
-        self.np_str = numpy.str_("")
-
-    @kernel
-    def run(self):
-        pass
-
-
-if __name__ == "__main__":
-    NumpyBoolDecay().run()

nac3artiq/demo/str_abi.py

@@ -1,26 +0,0 @@
-from min_artiq import *
-from numpy import ndarray, zeros as np_zeros
-
-
-@nac3
-class StrFail:
-    core: KernelInvariant[Core]
-
-    def __init__(self):
-        self.core = Core()
-
-    @kernel
-    def hello(self, arg: str):
-        pass
-
-    @kernel
-    def consume_ndarray(self, arg: ndarray[str, 1]):
-        pass
-
-    def run(self):
-        self.hello("world")
-        self.consume_ndarray(np_zeros([10], dtype=str))
-
-
-if __name__ == "__main__":
-    StrFail().run()

nac3artiq/demo/string_attribute_issue337.py

@@ -3,22 +3,22 @@ from numpy import int32
 
 
 @nac3
-class EmptyList:
+class Demo:
     core: KernelInvariant[Core]
+    attr1: KernelInvariant[str]
+    attr2: KernelInvariant[int32]
+
 
     def __init__(self):
         self.core = Core()
-
-    @rpc
-    def get_empty(self) -> list[int32]:
-        return []
+        self.attr2 = 32
+        self.attr1 = "SAMPLE"
 
     @kernel
     def run(self):
-        a: list[int32] = self.get_empty()
-        if a != []:
-            raise ValueError
+        print_int32(self.attr2)
+        self.attr1
 
 
 if __name__ == "__main__":
-    EmptyList().run()
+    Demo().run()

nac3artiq/demo/support_class_attr_issue102.py

@@ -0,0 +1,40 @@
+from min_artiq import *
+from numpy import int32
+
+
+@nac3
+class Demo:
+    attr1: KernelInvariant[int32] = 2
+    attr2: int32 = 4
+    attr3: Kernel[int32]
+
+    @kernel
+    def __init__(self):
+        self.attr3 = 8
+
+
+@nac3
+class NAC3Devices:
+    core: KernelInvariant[Core]
+    attr4: KernelInvariant[int32] = 16
+    
+    def __init__(self):
+        self.core = Core()
+
+    @kernel
+    def run(self):
+        Demo.attr1  # Supported
+        # Demo.attr2  # Field not accessible on Kernel
+        # Demo.attr3  # Only attributes can be accessed in this way
+        # Demo.attr1 = 2  # Attributes are immutable
+
+        self.attr4  # Attributes can be accessed within class
+
+        obj = Demo()
+        obj.attr1  # Attributes can be accessed by class objects
+
+        NAC3Devices.attr4  # Attributes accessible for classes without __init__
+
+
+if __name__ == "__main__":
+    NAC3Devices().run()

nac3artiq/src/debug.rs

@@ -1,273 +0,0 @@
-use itertools::Itertools;
-
-use nac3core::{toplevel::TopLevelDef, typecheck::typedef::Unifier};
-
-use super::{InnerResolver, symbol_resolver::PyValueHandle};
-
-impl InnerResolver {
-    pub fn debug_str(&self, tld: Option<&[TopLevelDef]>, unifier: &Option<&mut Unifier>) -> String {
-        fn fmt_elems(elems: &str) -> String {
-            if elems.is_empty() { String::new() } else { format!("\n{elems}\n\t") }
-        }
-        fn stringify_pyvalue_handle(handle: &PyValueHandle) -> String {
-            format!("(id: {}, value: {})", handle.0, handle.1)
-        }
-        fn stringify_tld(tld: &TopLevelDef) -> String {
-            match tld {
-                TopLevelDef::Module { name, .. } => {
-                    format!("TopLevelDef::Module {{ name: {name} }}")
-                }
-                TopLevelDef::Class { name, .. } => {
-                    format!("TopLevelDef::Class {{ name: {name} }}")
-                }
-                TopLevelDef::Function { name, .. } => {
-                    format!("TopLevelDef::Function {{ name: {name} }}")
-                }
-                TopLevelDef::Variable { name, .. } => {
-                    format!("TopLevelDef::Variable {{ name: {name} }}")
-                }
-            }
-        }
-
-        let mut str = String::new();
-        str.push_str("nac3artiq::InnerResolver {");
-
-        {
-            let id_to_type = self.id_to_type.read();
-            str.push_str(
-                format!(
-                    "\n\tid_to_type: {{{}}},",
-                    fmt_elems(
-                        id_to_type
-                            .iter()
-                            .sorted_by_cached_key(|(k, _)| k.to_string())
-                            .map(|(k, v)| {
-                                let ty_str = unifier.as_ref().map_or_else(
-                                    || format!("{v:?}"),
-                                    |unifier| unifier.stringify(*v),
-                                );
-                                format!("\t\t{k} -> {ty_str}")
-                            })
-                            .join(",\n")
-                            .as_str()
-                    ),
-                )
-                .as_str(),
-            );
-        }
-
-        {
-            let id_to_def = self.id_to_def.read();
-            str.push_str(
-                format!(
-                    "\n\tid_to_def: {{{}}},",
-                    fmt_elems(
-                        id_to_def
-                            .iter()
-                            .sorted_by_cached_key(|(k, _)| k.to_string())
-                            .map(|(k, v)| {
-                                let tld_str = tld.map_or_else(
-                                    || format!("{v:?}"),
-                                    |tlds| stringify_tld(&tlds[v.0]),
-                                );
-                                format!("\t\t{k} -> {tld_str}")
-                            })
-                            .join(",\n")
-                            .as_str()
-                    )
-                )
-                .as_str(),
-            );
-        }
-
-        {
-            let id_to_pyval = self.id_to_pyval.read();
-            str.push_str(
-                format!(
-                    "\n\tid_to_pyval: {{{}}},",
-                    fmt_elems(
-                        id_to_pyval
-                            .iter()
-                            .sorted_by_cached_key(|(k, _)| k.to_string())
-                            .map(|(k, v)| { format!("\t\t{k} -> {}", stringify_pyvalue_handle(v)) })
-                            .join(",\n")
-                            .as_str()
-                    )
-                )
-                .as_str(),
-            );
-        }
-
-        {
-            let id_to_primitive = self.id_to_primitive.read();
-            str.push_str(
-                format!(
-                    "\n\tid_to_primitive: {{{}}},",
-                    fmt_elems(
-                        id_to_primitive
-                            .iter()
-                            .sorted_by_key(|(k, _)| *k)
-                            .map(|(k, v)| { format!("\t\t{k} -> {v:?}") })
-                            .join(",\n")
-                            .as_str()
-                    )
-                )
-                .as_str(),
-            );
-        }
-
-        {
-            let field_to_val = self.field_to_val.read();
-            str.push_str(
-                format!(
-                    "\n\tfield_to_val: {{{}}},",
-                    fmt_elems(
-                        field_to_val
-                            .iter()
-                            .sorted_by_key(|((id, _), _)| *id)
-                            .map(|((id, name), pyval)| {
-                                format!(
-                                    "\t\t({id}, {name}) -> {}",
-                                    pyval.as_ref().map_or_else(
-                                        || String::from("None"),
-                                        |pyval| format!(
-                                            "Some({})",
-                                            stringify_pyvalue_handle(pyval)
-                                        )
-                                    )
-                                )
-                            })
-                            .join(",\n")
-                            .as_str()
-                    )
-                )
-                .as_str(),
-            );
-        }
-
-        {
-            let global_value_ids = self.global_value_ids.read();
-            str.push_str(
-                format!(
-                    "\n\tglobal_value_ids: {{{}}},",
-                    fmt_elems(
-                        global_value_ids
-                            .iter()
-                            .sorted_by_key(|(k, _)| *k)
-                            .map(|(k, v)| format!("\t\t{k} -> {v}"))
-                            .join(",\n")
-                            .as_str()
-                    )
-                )
-                .as_str(),
-            );
-        }
-
-        {
-            let pyid_to_def = self.pyid_to_def.read();
-            str.push_str(
-                format!(
-                    "\n\tpyid_to_def: {{{}}},",
-                    fmt_elems(
-                        pyid_to_def
-                            .iter()
-                            .sorted_by_key(|(k, _)| *k)
-                            .map(|(k, v)| {
-                                let tld_str = tld.map_or_else(
-                                    || format!("{v:?}"),
-                                    |tlds| stringify_tld(&tlds[v.0]),
-                                );
-                                format!("\t\t{k} -> {tld_str}")
-                            })
-                            .join(",\n")
-                            .as_str()
-                    )
-                )
-                .as_str(),
-            );
-        }
-
-        {
-            let pyid_to_type = self.pyid_to_type.read();
-            str.push_str(
-                format!(
-                    "\n\tpyid_to_type: {{{}}},",
-                    fmt_elems(
-                        pyid_to_type
-                            .iter()
-                            .sorted_by_key(|(k, _)| *k)
-                            .map(|(k, v)| {
-                                let ty_str = unifier.as_ref().map_or_else(
-                                    || format!("{v:?}"),
-                                    |unifier| unifier.stringify(*v),
-                                );
-                                format!("\t\t{k} -> {ty_str}")
-                            })
-                            .join(",\n")
-                            .as_str()
-                    )
-                )
-                .as_str(),
-            );
-        }
-
-        {
-            let string_store = self.string_store.read();
-            str.push_str(
-                format!(
-                    "\n\tstring_store: {{{}}},",
-                    fmt_elems(
-                        string_store
-                            .iter()
-                            .sorted_by_key(|(k, _)| *k)
-                            .map(|(k, v)| format!("\t\t{k} -> {v}"))
-                            .join(",\n")
-                            .as_str()
-                    )
-                )
-                .as_str(),
-            );
-        }
-
-        {
-            let exception_ids = self.exception_ids.read();
-            str.push_str(
-                format!(
-                    "\n\texception_ids: {{{}}},",
-                    fmt_elems(
-                        exception_ids
-                            .iter()
-                            .sorted_by_key(|(k, _)| *k)
-                            .map(|(k, v)| format!("\t\t{k} -> {v}"))
-                            .join(",\n")
-                            .as_str()
-                    )
-                )
-                .as_str(),
-            );
-        }
-
-        let name_to_pyid = &self.name_to_pyid;
-        str.push_str(
-            format!(
-                "\n\tname_to_pyid: {{{}}},",
-                fmt_elems(
-                    name_to_pyid
-                        .iter()
-                        .sorted_by_cached_key(|(k, _)| k.to_string())
-                        .map(|(k, v)| format!("\t\t{k} -> {v}"))
-                        .join(",\n")
-                        .as_str()
-                )
-            )
-            .as_str(),
-        );
-
-        let module = &self.module;
-        str.push_str(format!("\n\tmodule: {module}").as_str());
-
-        str.push_str("\n}");
-
-        str
-    }
-}

nac3artiq/src/timeline.rs

@@ -1,7 +1,9 @@
-use nac3core::{
-    codegen::{CodeGenContext, expr::infer_and_call_function},
-    inkwell::{AddressSpace, AtomicOrdering, values::BasicValueEnum},
+use inkwell::{
+    values::{BasicValueEnum, CallSiteValue},
+    AddressSpace, AtomicOrdering,
 };
+use itertools::Either;
+use nac3core::codegen::CodeGenContext;
 
 /// Functions for manipulating the timeline.
 pub trait TimeFns {
@@ -29,7 +31,7 @@ impl TimeFns for NowPinningTimeFns64 {
             .unwrap_or_else(|| ctx.module.add_global(i64_type, None, "now"));
         let now_hiptr = ctx
             .builder
-            .build_bit_cast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
+            .build_bitcast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
             .map(BasicValueEnum::into_pointer_value)
             .unwrap();
 
@@ -78,7 +80,7 @@ impl TimeFns for NowPinningTimeFns64 {
             .unwrap_or_else(|| ctx.module.add_global(i64_type, None, "now"));
         let now_hiptr = ctx
             .builder
-            .build_bit_cast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
+            .build_bitcast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
             .map(BasicValueEnum::into_pointer_value)
             .unwrap();
 
@@ -107,7 +109,7 @@ impl TimeFns for NowPinningTimeFns64 {
             .unwrap_or_else(|| ctx.module.add_global(i64_type, None, "now"));
         let now_hiptr = ctx
             .builder
-            .build_bit_cast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
+            .build_bitcast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
             .map(BasicValueEnum::into_pointer_value)
             .unwrap();
 
@@ -205,7 +207,7 @@ impl TimeFns for NowPinningTimeFns {
             .unwrap_or_else(|| ctx.module.add_global(i64_type, None, "now"));
         let now_hiptr = ctx
             .builder
-            .build_bit_cast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
+            .build_bitcast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
             .map(BasicValueEnum::into_pointer_value)
             .unwrap();
 
@@ -256,7 +258,7 @@ impl TimeFns for NowPinningTimeFns {
         let time_lo = ctx.builder.build_int_truncate(time, i32_type, "time.lo").unwrap();
         let now_hiptr = ctx
             .builder
-            .build_bit_cast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
+            .build_bitcast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
             .map(BasicValueEnum::into_pointer_value)
             .unwrap();
 
@@ -283,27 +285,36 @@ pub struct ExternTimeFns {}
 
 impl TimeFns for ExternTimeFns {
     fn emit_now_mu<'ctx>(&self, ctx: &mut CodeGenContext<'ctx, '_>) -> BasicValueEnum<'ctx> {
-        infer_and_call_function(
-            ctx,
-            "now_mu",
-            Some(ctx.ctx.i64_type().into()),
-            &[],
-            Some("now_mu"),
-            None,
-        )
-        .unwrap()
+        let now_mu = ctx.module.get_function("now_mu").unwrap_or_else(|| {
+            ctx.module.add_function("now_mu", ctx.ctx.i64_type().fn_type(&[], false), None)
+        });
+        ctx.builder
+            .build_call(now_mu, &[], "now_mu")
+            .map(CallSiteValue::try_as_basic_value)
+            .map(Either::unwrap_left)
+            .unwrap()
     }
 
     fn emit_at_mu<'ctx>(&self, ctx: &mut CodeGenContext<'ctx, '_>, t: BasicValueEnum<'ctx>) {
-        assert_eq!(t.get_type(), ctx.ctx.i64_type().into());
-
-        infer_and_call_function(ctx, "at_mu", None, &[t], Some("at_mu"), None);
+        let at_mu = ctx.module.get_function("at_mu").unwrap_or_else(|| {
+            ctx.module.add_function(
+                "at_mu",
+                ctx.ctx.void_type().fn_type(&[ctx.ctx.i64_type().into()], false),
+                None,
+            )
+        });
+        ctx.builder.build_call(at_mu, &[t.into()], "at_mu").unwrap();
     }
 
     fn emit_delay_mu<'ctx>(&self, ctx: &mut CodeGenContext<'ctx, '_>, dt: BasicValueEnum<'ctx>) {
-        assert_eq!(dt.get_type(), ctx.ctx.i64_type().into());
-
-        infer_and_call_function(ctx, "delay_mu", None, &[dt], Some("delay_mu"), None);
+        let delay_mu = ctx.module.get_function("delay_mu").unwrap_or_else(|| {
+            ctx.module.add_function(
+                "delay_mu",
+                ctx.ctx.void_type().fn_type(&[ctx.ctx.i64_type().into()], false),
+                None,
+            )
+        });
+        ctx.builder.build_call(delay_mu, &[dt.into()], "delay_mu").unwrap();
     }
 }
 

nac3ast/Cargo.toml

@@ -2,7 +2,7 @@
 name = "nac3ast"
 version = "0.1.0"
 authors = ["RustPython Team", "M-Labs"]
-edition = "2024"
+edition = "2021"
 
 [features]
 default = ["constant-optimization", "fold"]
@@ -10,6 +10,7 @@ constant-optimization = ["fold"]
 fold = []
 
 [dependencies]
+lazy_static = "1.5"
 parking_lot = "0.12"
-string-interner = "0.19"
+string-interner = "0.17"
 fxhash = "0.2"

nac3ast/src/ast_gen.rs

@@ -5,12 +5,14 @@ pub use crate::location::Location;
 
 use fxhash::FxBuildHasher;
 use parking_lot::{Mutex, MutexGuard};
-use std::{cell::RefCell, collections::HashMap, fmt, sync::LazyLock};
-use string_interner::{DefaultBackend, StringInterner, symbol::SymbolU32};
+use std::{cell::RefCell, collections::HashMap, fmt};
+use string_interner::{symbol::SymbolU32, DefaultBackend, StringInterner};
 
 pub type Interner = StringInterner<DefaultBackend, FxBuildHasher>;
-static INTERNER: LazyLock<Mutex<Interner>> =
-    LazyLock::new(|| Mutex::new(StringInterner::with_hasher(FxBuildHasher::default())));
+lazy_static! {
+    static ref INTERNER: Mutex<Interner> =
+        Mutex::new(StringInterner::with_hasher(FxBuildHasher::default()));
+}
 
 thread_local! {
     static LOCAL_INTERNER: RefCell<HashMap<String, StrRef>> = RefCell::default();

nac3ast/src/fold_helpers.rs

@@ -1,6 +1,6 @@
-use crate::StrRef;
 use crate::constant;
 use crate::fold::Fold;
+use crate::StrRef;
 
 pub(crate) trait Foldable<T, U> {
     type Mapped;

nac3ast/src/lib.rs

@@ -1,4 +1,10 @@
-#![deny(future_incompatible, let_underscore, nonstandard_style, clippy::all)]
+#![deny(
+    future_incompatible,
+    let_underscore,
+    nonstandard_style,
+    rust_2024_compatibility,
+    clippy::all
+)]
 #![warn(clippy::pedantic)]
 #![allow(
     clippy::missing_errors_doc,
@@ -8,6 +14,9 @@
     clippy::wildcard_imports
 )]
 
+#[macro_use]
+extern crate lazy_static;
+
 mod ast_gen;
 mod constant;
 #[cfg(feature = "fold")]

nac3core/Cargo.toml

@@ -1,35 +1,31 @@
+[features]
+test = []
+
 [package]
 name = "nac3core"
 version = "0.1.0"
 authors = ["M-Labs"]
-edition = "2024"
-
-[features]
-default = ["derive"]
-derive = ["dep:nac3core_derive"]
-no-escape-analysis = []
+edition = "2021"
 
 [dependencies]
-itertools = "0.14"
+itertools = "0.13"
 crossbeam = "0.8"
-indexmap = "2.8"
+indexmap = "2.2"
 parking_lot = "0.12"
-nac3core_derive = { path = "nac3core_derive", optional = true }
+rayon = "1.8"
 nac3parser = { path = "../nac3parser" }
-strum = "0.27"
-strum_macros = "0.27"
+strum = "0.26.2"
+strum_macros = "0.26.4"
 
 [dependencies.inkwell]
-git = "https://github.com/Derppening/inkwell"
-tag = "0.5.0_llvm15-typed-ptr"
+version = "0.4"
 default-features = false
-features = ["llvm16-0-prefer-dynamic", "target-x86", "target-arm", "target-riscv", "no-libffi-linking", "typed-pointers"]
+features = ["llvm14-0", "target-x86", "target-arm", "target-riscv", "no-libffi-linking"]
 
 [dev-dependencies]
-test-case = "3.3"
+test-case = "1.2.0"
 indoc = "2.0"
-insta = "1.42"
-function_name = "0.3"
+insta = "=1.11.0"
 
 [build-dependencies]
-regex = "1.11"
+regex = "1.10"

nac3core/build.rs

@@ -1,3 +1,4 @@
+use regex::Regex;
 use std::{
     env,
     fs::File,
@@ -6,60 +7,45 @@ use std::{
     process::{Command, Stdio},
 };
 
-use regex::Regex;
-
-fn main() {
-    let out_dir = env::var("OUT_DIR").unwrap();
-    let out_dir = Path::new(&out_dir);
-    let irrt_dir = Path::new("irrt");
-
+fn compile_irrt(irrt_dir: &Path, out_dir: &Path) {
     let irrt_cpp_path = irrt_dir.join("irrt.cpp");
 
     /*
      * HACK: Sadly, clang doesn't let us emit generic LLVM bitcode.
      * Compiling for WASM32 and filtering the output with regex is the closest we can get.
      */
-    let mut flags: Vec<&str> = vec![
+    let flags: &[&str] = &[
         "--target=wasm32",
+        irrt_cpp_path.to_str().unwrap(),
         "-x",
         "c++",
-        "-std=c++20",
         "-fno-discard-value-names",
         "-fno-exceptions",
         "-fno-rtti",
+        match env::var("PROFILE").as_deref() {
+            Ok("debug") => "-O0",
+            Ok("release") => "-O3",
+            flavor => panic!("Unknown or missing build flavor {flavor:?}"),
+        },
         "-emit-llvm",
-        "-Xclang",
-        "-no-opaque-pointers",
         "-S",
         "-Wall",
         "-Wextra",
-        "-o",
-        "-",
+        "-Werror=return-type",
         "-I",
         irrt_dir.to_str().unwrap(),
-        irrt_cpp_path.to_str().unwrap(),
+        "-o",
+        "-",
     ];
 
-    match env::var("PROFILE").as_deref() {
-        Ok("debug") => {
-            flags.push("-O0");
-            flags.push("-DIRRT_DEBUG_ASSERT");
-        }
-        Ok("release") => {
-            flags.push("-O3");
-        }
-        flavor => panic!("Unknown or missing build flavor {flavor:?}"),
-    }
+    println!("cargo:rerun-if-changed={}", out_dir.to_str().unwrap());
 
-    // Tell Cargo to rerun if any file under `irrt_dir` (recursive) changes
-    println!("cargo:rerun-if-changed={}", irrt_dir.to_str().unwrap());
-
-    // Compile IRRT and capture the LLVM IR output
     let output = Command::new("clang-irrt")
         .args(flags)
         .output()
-        .inspect(|o| {
+        .map(|o| {
             assert!(o.status.success(), "{}", std::str::from_utf8(&o.stderr).unwrap());
+            o
         })
         .unwrap();
 
@@ -67,17 +53,11 @@ fn main() {
     let output = std::str::from_utf8(&output.stdout).unwrap().replace("\r\n", "\n");
     let mut filtered_output = String::with_capacity(output.len());
 
-    // Filter out irrelevant IR
-    //
-    // Regex:
-    // - `(?ms:^define.*?\}$)` captures LLVM `define` blocks
-    // - `(?m:^declare.*?$)` captures LLVM `declare` lines
-    // - `(?m:^%.+?=\s*type\s*\{.+?\}$)` captures LLVM `type` declarations
-    // - `(?m:^@.+?=.+$)` captures global constants
-    let regex_filter = Regex::new(
-        r"(?ms:^define.*?\}$)|(?m:^declare.*?$)|(?m:^%.+?=\s*type\s*\{.+?\}$)|(?m:^@.+?=.+$)",
-    )
-    .unwrap();
+    // (?ms:^define.*?\}$) to capture `define` blocks
+    // (?m:^declare.*?$) to capture `declare` blocks
+    // (?m:^%.+?=\s*type\s*\{.+?\}$) to capture `type` declarations
+    let regex_filter =
+        Regex::new(r"(?ms:^define.*?\}$)|(?m:^declare.*?$)|(?m:^%.+?=\s*type\s*\{.+?\}$)").unwrap();
     for f in regex_filter.captures_iter(&output) {
         assert_eq!(f.len(), 1);
         filtered_output.push_str(&f[0]);
@@ -88,21 +68,16 @@ fn main() {
         .unwrap()
         .replace_all(&filtered_output, "");
 
-    // For debugging
-    // Doing `DEBUG_DUMP_IRRT=1 cargo build -p nac3core` dumps the LLVM IR generated
-    const DEBUG_DUMP_IRRT: &str = "DEBUG_DUMP_IRRT";
-    println!("cargo:rerun-if-env-changed={DEBUG_DUMP_IRRT}");
-    if env::var(DEBUG_DUMP_IRRT).is_ok() {
+    println!("cargo:rerun-if-env-changed=DEBUG_DUMP_IRRT");
+    if env::var("DEBUG_DUMP_IRRT").is_ok() {
         let mut file = File::create(out_dir.join("irrt.ll")).unwrap();
         file.write_all(output.as_bytes()).unwrap();
-
         let mut file = File::create(out_dir.join("irrt-filtered.ll")).unwrap();
         file.write_all(filtered_output.as_bytes()).unwrap();
     }
 
     let mut llvm_as = Command::new("llvm-as-irrt")
         .stdin(Stdio::piped())
-        .arg("-opaque-pointers=0")
         .arg("-o")
         .arg(out_dir.join("irrt.bc"))
         .spawn()
@@ -110,3 +85,50 @@ fn main() {
     llvm_as.stdin.as_mut().unwrap().write_all(filtered_output.as_bytes()).unwrap();
     assert!(llvm_as.wait().unwrap().success());
 }
+
+fn compile_irrt_test(irrt_dir: &Path, out_dir: &Path) {
+    let irrt_test_cpp_path = irrt_dir.join("irrt_test.cpp");
+    let exe_path = out_dir.join("irrt_test.out");
+
+    let flags: &[&str] = &[
+        irrt_test_cpp_path.to_str().unwrap(),
+        "-x",
+        "c++",
+        "-I",
+        irrt_dir.to_str().unwrap(),
+        "-g",
+        "-fno-discard-value-names",
+        "-O0",
+        "-Wall",
+        "-Wextra",
+        "-Werror=return-type",
+        "-lm", // for `tgamma()`, `lgamma()`
+        "-o",
+        exe_path.to_str().unwrap(),
+    ];
+
+    Command::new("clang-irrt-test")
+        .args(flags)
+        .output()
+        .map(|o| {
+            assert!(o.status.success(), "{}", std::str::from_utf8(&o.stderr).unwrap());
+            o
+        })
+        .unwrap();
+    println!("cargo:rerun-if-changed={}", out_dir.to_str().unwrap());
+}
+
+fn main() {
+    let out_dir = env::var("OUT_DIR").unwrap();
+    let out_dir = Path::new(&out_dir);
+
+    let irrt_dir = Path::new("./irrt");
+
+    compile_irrt(irrt_dir, out_dir);
+
+    // https://github.com/rust-lang/cargo/issues/2549
+    // `cargo test -F test` to also build `irrt_test.cpp
+    if cfg!(feature = "test") {
+        compile_irrt_test(irrt_dir, out_dir);
+    }
+}

nac3core/irrt/irrt.cpp

@@ -1,16 +1,5 @@
-#include "irrt/cc-builtins.hpp"
-#include "irrt/exception.hpp"
-#include "irrt/list.hpp"
-#include "irrt/math.hpp"
-#include "irrt/range.hpp"
-#include "irrt/slice.hpp"
-#include "irrt/string.hpp"
-#include "irrt/ndarray/basic.hpp"
-#include "irrt/ndarray/def.hpp"
-#include "irrt/ndarray/iter.hpp"
-#include "irrt/ndarray/indexing.hpp"
-#include "irrt/ndarray/array.hpp"
-#include "irrt/ndarray/reshape.hpp"
-#include "irrt/ndarray/broadcast.hpp"
-#include "irrt/ndarray/transpose.hpp"
-#include "irrt/ndarray/matmul.hpp"
+#include "irrt_everything.hpp"
+
+/*
+    This file will be read by `clang-irrt` to conveniently produce LLVM IR for `nac3core/codegen`.
+*/

nac3core/irrt/irrt.hpp

@@ -0,0 +1,437 @@
+#ifndef IRRT_DONT_TYPEDEF_INTS
+typedef _BitInt(8) int8_t;
+typedef unsigned _BitInt(8) uint8_t;
+typedef _BitInt(32) int32_t;
+typedef unsigned _BitInt(32) uint32_t;
+typedef _BitInt(64) int64_t;
+typedef unsigned _BitInt(64) uint64_t;
+#endif
+
+// NDArray indices are always `uint32_t`.
+typedef uint32_t NDIndex;
+// The type of an index or a value describing the length of a range/slice is
+// always `int32_t`.
+typedef int32_t SliceIndex;
+
+template <typename T>
+static T max(T a, T b) {
+    return a > b ? a : b;
+}
+
+template <typename T>
+static T min(T a, T b) {
+    return a > b ? b : a;
+}
+
+// adapted from GNU Scientific Library: https://git.savannah.gnu.org/cgit/gsl.git/tree/sys/pow_int.c
+// need to make sure `exp >= 0` before calling this function
+template <typename T>
+static T __nac3_int_exp_impl(T base, T exp) {
+    T res = 1;
+    /* repeated squaring method */
+    do {
+        if (exp & 1) {
+            res *= base; /* for n odd */
+        }
+        exp >>= 1;
+        base *= base;
+    } while (exp);
+    return res;
+}
+
+template <typename SizeT>
+static SizeT __nac3_ndarray_calc_size_impl(
+    const SizeT *list_data,
+    SizeT list_len,
+    SizeT begin_idx,
+    SizeT end_idx
+) {
+    __builtin_assume(end_idx <= list_len);
+
+    SizeT num_elems = 1;
+    for (SizeT i = begin_idx; i < end_idx; ++i) {
+        SizeT val = list_data[i];
+        __builtin_assume(val > 0);
+        num_elems *= val;
+    }
+    return num_elems;
+}
+
+template <typename SizeT>
+static void __nac3_ndarray_calc_nd_indices_impl(
+    SizeT index,
+    const SizeT *dims,
+    SizeT num_dims,
+    NDIndex *idxs
+) {
+    SizeT stride = 1;
+    for (SizeT dim = 0; dim < num_dims; dim++) {
+        SizeT i = num_dims - dim - 1;
+        __builtin_assume(dims[i] > 0);
+        idxs[i] = (index / stride) % dims[i];
+        stride *= dims[i];
+    }
+}
+
+template <typename SizeT>
+static SizeT __nac3_ndarray_flatten_index_impl(
+    const SizeT *dims,
+    SizeT num_dims,
+    const NDIndex *indices,
+    SizeT num_indices
+) {
+    SizeT idx = 0;
+    SizeT stride = 1;
+    for (SizeT i = 0; i < num_dims; ++i) {
+        SizeT ri = num_dims - i - 1;
+        if (ri < num_indices) {
+            idx += stride * indices[ri];
+        }
+
+        __builtin_assume(dims[i] > 0);
+        stride *= dims[ri];
+    }
+    return idx;
+}
+
+template <typename SizeT>
+static void __nac3_ndarray_calc_broadcast_impl(
+    const SizeT *lhs_dims,
+    SizeT lhs_ndims,
+    const SizeT *rhs_dims,
+    SizeT rhs_ndims,
+    SizeT *out_dims
+) {
+    SizeT max_ndims = lhs_ndims > rhs_ndims ? lhs_ndims : rhs_ndims;
+
+    for (SizeT i = 0; i < max_ndims; ++i) {
+        const SizeT *lhs_dim_sz = i < lhs_ndims ? &lhs_dims[lhs_ndims - i - 1] : nullptr;
+        const SizeT *rhs_dim_sz = i < rhs_ndims ? &rhs_dims[rhs_ndims - i - 1] : nullptr;
+        SizeT *out_dim = &out_dims[max_ndims - i - 1];
+
+        if (lhs_dim_sz == nullptr) {
+            *out_dim = *rhs_dim_sz;
+        } else if (rhs_dim_sz == nullptr) {
+            *out_dim = *lhs_dim_sz;
+        } else if (*lhs_dim_sz == 1) {
+            *out_dim = *rhs_dim_sz;
+        } else if (*rhs_dim_sz == 1) {
+            *out_dim = *lhs_dim_sz;
+        } else if (*lhs_dim_sz == *rhs_dim_sz) {
+            *out_dim = *lhs_dim_sz;
+        } else {
+            __builtin_unreachable();
+        }
+    }
+}
+
+template <typename SizeT>
+static void __nac3_ndarray_calc_broadcast_idx_impl(
+    const SizeT *src_dims,
+    SizeT src_ndims,
+    const NDIndex *in_idx,
+    NDIndex *out_idx
+) {
+    for (SizeT i = 0; i < src_ndims; ++i) {
+        SizeT src_i = src_ndims - i - 1;
+        out_idx[src_i] = src_dims[src_i] == 1 ? 0 : in_idx[src_i];
+    }
+}
+
+template<typename SizeT>
+static void __nac3_ndarray_strides_from_shape_impl(
+    SizeT ndims,
+    SizeT *shape,
+    SizeT *dst_strides
+) {
+    SizeT stride_product = 1;
+    for (SizeT i = 0; i < ndims; i++) {
+        int dim_i = ndims - i - 1;
+        dst_strides[dim_i] = stride_product;
+        stride_product *= shape[dim_i];
+    }
+}
+
+extern "C" {
+    #define DEF_nac3_int_exp_(T) \
+        T __nac3_int_exp_##T(T base, T exp) {\
+            return __nac3_int_exp_impl(base, exp);\
+        }
+
+    DEF_nac3_int_exp_(int32_t)
+    DEF_nac3_int_exp_(int64_t)
+    DEF_nac3_int_exp_(uint32_t)
+    DEF_nac3_int_exp_(uint64_t)
+
+    SliceIndex __nac3_slice_index_bound(SliceIndex i, const SliceIndex len) {
+        if (i < 0) {
+            i = len + i;
+        }
+        if (i < 0) {
+            return 0;
+        } else if (i > len) {
+            return len;
+        }
+        return i;
+    }
+
+    SliceIndex __nac3_range_slice_len(
+        const SliceIndex start,
+        const SliceIndex end,
+        const SliceIndex step
+    ) {
+        SliceIndex diff = end - start;
+        if (diff > 0 && step > 0) {
+            return ((diff - 1) / step) + 1;
+        } else if (diff < 0 && step < 0) {
+            return ((diff + 1) / step) + 1;
+        } else {
+            return 0;
+        }
+    }
+
+    // Handle list assignment and dropping part of the list when
+    // both dest_step and src_step are +1.
+    // - All the index must *not* be out-of-bound or negative,
+    // - The end index is *inclusive*,
+    // - The length of src and dest slice size should already
+    // be checked: if dest.step == 1 then len(src) <= len(dest) else len(src) == len(dest)
+    SliceIndex __nac3_list_slice_assign_var_size(
+        SliceIndex dest_start,
+        SliceIndex dest_end,
+        SliceIndex dest_step,
+        uint8_t *dest_arr,
+        SliceIndex dest_arr_len,
+        SliceIndex src_start,
+        SliceIndex src_end,
+        SliceIndex src_step,
+        uint8_t *src_arr,
+        SliceIndex src_arr_len,
+        const SliceIndex size
+    ) {
+        /* if dest_arr_len == 0, do nothing since we do not support extending list */
+        if (dest_arr_len == 0) return dest_arr_len;
+        /* if both step is 1, memmove directly, handle the dropping of the list, and shrink size */
+        if (src_step == dest_step && dest_step == 1) {
+            const SliceIndex src_len = (src_end >= src_start) ? (src_end - src_start + 1) : 0;
+            const SliceIndex dest_len = (dest_end >= dest_start) ? (dest_end - dest_start + 1) : 0;
+            if (src_len > 0) {
+                __builtin_memmove(
+                    dest_arr + dest_start * size,
+                    src_arr + src_start * size,
+                    src_len * size
+                );
+            }
+            if (dest_len > 0) {
+                /* dropping */
+                __builtin_memmove(
+                    dest_arr + (dest_start + src_len) * size,
+                    dest_arr + (dest_end + 1) * size,
+                    (dest_arr_len - dest_end - 1) * size
+                );
+            }
+            /* shrink size */
+            return dest_arr_len - (dest_len - src_len);
+        }
+        /* if two range overlaps, need alloca */
+        uint8_t need_alloca =
+            (dest_arr == src_arr)
+            && !(
+                max(dest_start, dest_end) < min(src_start, src_end)
+                || max(src_start, src_end) < min(dest_start, dest_end)
+            );
+        if (need_alloca) {
+            uint8_t *tmp = reinterpret_cast<uint8_t *>(__builtin_alloca(src_arr_len * size));
+            __builtin_memcpy(tmp, src_arr, src_arr_len * size);
+            src_arr = tmp;
+        }
+        SliceIndex src_ind = src_start;
+        SliceIndex dest_ind = dest_start;
+        for (;
+            (src_step > 0) ? (src_ind <= src_end) : (src_ind >= src_end);
+            src_ind += src_step, dest_ind += dest_step
+        ) {
+            /* for constant optimization */
+            if (size == 1) {
+                __builtin_memcpy(dest_arr + dest_ind, src_arr + src_ind, 1);
+            } else if (size == 4) {
+                __builtin_memcpy(dest_arr + dest_ind * 4, src_arr + src_ind * 4, 4);
+            } else if (size == 8) {
+                __builtin_memcpy(dest_arr + dest_ind * 8, src_arr + src_ind * 8, 8);
+            } else {
+                /* memcpy for var size, cannot overlap after previous alloca */
+                __builtin_memcpy(dest_arr + dest_ind * size, src_arr + src_ind * size, size);
+            }
+        }
+        /* only dest_step == 1 can we shrink the dest list. */
+        /* size should be ensured prior to calling this function */
+        if (dest_step == 1 && dest_end >= dest_start) {
+            __builtin_memmove(
+                dest_arr + dest_ind * size,
+                dest_arr + (dest_end + 1) * size,
+                (dest_arr_len - dest_end - 1) * size
+            );
+            return dest_arr_len - (dest_end - dest_ind) - 1;
+        }
+        return dest_arr_len;
+    }
+
+    int32_t __nac3_isinf(double x) {
+        return __builtin_isinf(x);
+    }
+
+    int32_t __nac3_isnan(double x) {
+        return __builtin_isnan(x);
+    }
+
+    double tgamma(double arg);
+
+    double __nac3_gamma(double z) {
+        // Handling for denormals
+        //     | x                 | Python gamma(x) | C tgamma(x) |
+        // --- | ----------------- | --------------- | ----------- |
+        // (1) | nan               | nan             | nan         |
+        // (2) | -inf              | -inf            | inf         |
+        // (3) | inf               | inf             | inf         |
+        // (4) | 0.0               | inf             | inf         |
+        // (5) | {-1.0, -2.0, ...} | inf             | nan         |
+
+        // (1)-(3)
+        if (__builtin_isinf(z) || __builtin_isnan(z)) {
+            return z;
+        }
+
+        double v = tgamma(z);
+
+        // (4)-(5)
+        return __builtin_isinf(v) || __builtin_isnan(v) ? __builtin_inf() : v;
+    }
+
+    double lgamma(double arg);
+
+    double __nac3_gammaln(double x) {
+        // libm's handling of value overflows differs from scipy:
+        // - scipy: gammaln(-inf) -> -inf
+        // - libm : lgamma(-inf) -> inf
+
+        if (__builtin_isinf(x)) {
+            return x;
+        }
+
+        return lgamma(x);
+    }
+
+    double j0(double x);
+
+    double __nac3_j0(double x) {
+        // libm's handling of value overflows differs from scipy:
+        // - scipy: j0(inf) -> nan
+        // - libm : j0(inf) -> 0.0
+
+        if (__builtin_isinf(x)) {
+            return __builtin_nan("");
+        }
+
+        return j0(x);
+    }
+
+    uint32_t __nac3_ndarray_calc_size(
+        const uint32_t *list_data,
+        uint32_t list_len,
+        uint32_t begin_idx,
+        uint32_t end_idx
+    ) {
+        return __nac3_ndarray_calc_size_impl(list_data, list_len, begin_idx, end_idx);
+    }
+
+    uint64_t __nac3_ndarray_calc_size64(
+        const uint64_t *list_data,
+        uint64_t list_len,
+        uint64_t begin_idx,
+        uint64_t end_idx
+    ) {
+        return __nac3_ndarray_calc_size_impl(list_data, list_len, begin_idx, end_idx);
+    }
+
+    void __nac3_ndarray_calc_nd_indices(
+        uint32_t index,
+        const uint32_t* dims,
+        uint32_t num_dims,
+        NDIndex* idxs
+    ) {
+        __nac3_ndarray_calc_nd_indices_impl(index, dims, num_dims, idxs);
+    }
+
+    void __nac3_ndarray_calc_nd_indices64(
+        uint64_t index,
+        const uint64_t* dims,
+        uint64_t num_dims,
+        NDIndex* idxs
+    ) {
+        __nac3_ndarray_calc_nd_indices_impl(index, dims, num_dims, idxs);
+    }
+
+    uint32_t __nac3_ndarray_flatten_index(
+        const uint32_t* dims,
+        uint32_t num_dims,
+        const NDIndex* indices,
+        uint32_t num_indices
+    ) {
+        return __nac3_ndarray_flatten_index_impl(dims, num_dims, indices, num_indices);
+    }
+
+    uint64_t __nac3_ndarray_flatten_index64(
+        const uint64_t* dims,
+        uint64_t num_dims,
+        const NDIndex* indices,
+        uint64_t num_indices
+    ) {
+        return __nac3_ndarray_flatten_index_impl(dims, num_dims, indices, num_indices);
+    }
+
+    void __nac3_ndarray_calc_broadcast(
+        const uint32_t *lhs_dims,
+        uint32_t lhs_ndims,
+        const uint32_t *rhs_dims,
+        uint32_t rhs_ndims,
+        uint32_t *out_dims
+    ) {
+        return __nac3_ndarray_calc_broadcast_impl(lhs_dims, lhs_ndims, rhs_dims, rhs_ndims, out_dims);
+    }
+
+    void __nac3_ndarray_calc_broadcast64(
+        const uint64_t *lhs_dims,
+        uint64_t lhs_ndims,
+        const uint64_t *rhs_dims,
+        uint64_t rhs_ndims,
+        uint64_t *out_dims
+    ) {
+        return __nac3_ndarray_calc_broadcast_impl(lhs_dims, lhs_ndims, rhs_dims, rhs_ndims, out_dims);
+    }
+
+    void __nac3_ndarray_calc_broadcast_idx(
+        const uint32_t *src_dims,
+        uint32_t src_ndims,
+        const NDIndex *in_idx,
+        NDIndex *out_idx
+    ) {
+        __nac3_ndarray_calc_broadcast_idx_impl(src_dims, src_ndims, in_idx, out_idx);
+    }
+
+    void __nac3_ndarray_calc_broadcast_idx64(
+        const uint64_t *src_dims,
+        uint64_t src_ndims,
+        const NDIndex *in_idx,
+        NDIndex *out_idx
+    ) {
+        __nac3_ndarray_calc_broadcast_idx_impl(src_dims, src_ndims, in_idx, out_idx);
+    }
+
+    void __nac3_ndarray_strides_from_shape(uint32_t ndims, uint32_t* shape, uint32_t* dst_strides) {
+        __nac3_ndarray_strides_from_shape_impl(ndims, shape, dst_strides);
+    }
+
+    void __nac3_ndarray_strides_from_shape64(uint64_t ndims, uint64_t* shape, uint64_t* dst_strides) {
+        __nac3_ndarray_strides_from_shape_impl(ndims, shape, dst_strides);
+    }
+}

nac3core/irrt/irrt/cc-builtins.hpp

@@ -1,34 +0,0 @@
-#pragma once
-
-extern "C" {
-#define DEF_builtin_unary(RET, NAME, TY) \
-    RET __nac3_##NAME(TY v) { return __builtin_##NAME(v); }
-#define DEF_builtin_binary(RET, NAME, TY1, TY2) \
-    RET __nac3_##NAME(TY1 v1, TY2 v2) { return __builtin_##NAME(v1, v2); }
-
-DEF_builtin_unary(bool, isinf, double);
-DEF_builtin_unary(bool, isnan, double);
-DEF_builtin_unary(double, tan, double);
-DEF_builtin_unary(double, asin, double);
-DEF_builtin_unary(double, acos, double);
-DEF_builtin_unary(double, atan, double);
-DEF_builtin_unary(double, sinh, double);
-DEF_builtin_unary(double, cosh, double);
-DEF_builtin_unary(double, tanh, double);
-DEF_builtin_unary(double, asinh, double);
-DEF_builtin_unary(double, acosh, double);
-DEF_builtin_unary(double, atanh, double);
-DEF_builtin_unary(double, expm1, double);
-DEF_builtin_unary(double, cbrt, double);
-DEF_builtin_unary(double, erf, double);
-DEF_builtin_unary(double, erfc, double);
-
-#define __builtin_gamma __builtin_tgamma
-DEF_builtin_unary(double, gamma, double);
-#undef __builtin_gamma
-
-DEF_builtin_binary(double, atan2, double, double);
-DEF_builtin_binary(double, hypot, double, double);
-DEF_builtin_binary(double, nextafter, double, double);
-DEF_builtin_binary(double, ldexp, double, int);
-}

nac3core/irrt/irrt/cslice.hpp

@@ -1,9 +0,0 @@
-#pragma once
-
-#include "irrt/int_types.hpp"
-
-template<typename SizeT>
-struct CSlice {
-    void* base;
-    SizeT len;
-};

nac3core/irrt/irrt/debug.hpp

@@ -1,25 +0,0 @@
-#pragma once
-
-// Set in nac3core/build.rs
-#ifdef IRRT_DEBUG_ASSERT
-#define IRRT_DEBUG_ASSERT_BOOL true
-#else
-#define IRRT_DEBUG_ASSERT_BOOL false
-#endif
-
-#define raise_debug_assert(SizeT, msg, param1, param2, param3) \
-    raise_exception(SizeT, EXN_ASSERTION_ERROR, "IRRT debug assert failed: " msg, param1, param2, param3)
-
-#define debug_assert_eq(SizeT, lhs, rhs)                                           \
-    if constexpr (IRRT_DEBUG_ASSERT_BOOL) {                                        \
-        if ((lhs) != (rhs)) {                                                      \
-            raise_debug_assert(SizeT, "LHS = {0}. RHS = {1}", lhs, rhs, NO_PARAM); \
-        }                                                                          \
-    }
-
-#define debug_assert(SizeT, expr)                                                  \
-    if constexpr (IRRT_DEBUG_ASSERT_BOOL) {                                        \
-        if (!(expr)) {                                                             \
-            raise_debug_assert(SizeT, "Got false.", NO_PARAM, NO_PARAM, NO_PARAM); \
-        }                                                                          \
-    }

nac3core/irrt/irrt/exception.hpp

@@ -1,85 +0,0 @@
-#pragma once
-
-#include "irrt/cslice.hpp"
-#include "irrt/int_types.hpp"
-
-/**
- * @brief The int type of ARTIQ exception IDs.
- */
-using ExceptionId = int32_t;
-
-/*
- * Set of exceptions C++ IRRT can use.
- * Must be synchronized with `setup_irrt_exceptions` in `nac3core/src/codegen/irrt/mod.rs`.
- */
-extern "C" {
-ExceptionId EXN_INDEX_ERROR;
-ExceptionId EXN_VALUE_ERROR;
-ExceptionId EXN_ASSERTION_ERROR;
-ExceptionId EXN_TYPE_ERROR;
-}
-
-/**
- * @brief Extern function to `__nac3_raise`
- *
- * The parameter `err` could be `Exception<int32_t>` or `Exception<int64_t>`. The caller
- * must make sure to pass `Exception`s with the correct `SizeT` depending on the `size_t` of the runtime.
- */
-extern "C" void __nac3_raise(void* err);
-
-namespace {
-/**
- * @brief NAC3's Exception struct
- */
-template<typename SizeT>
-struct Exception {
-    ExceptionId id;
-    CSlice<SizeT> filename;
-    int32_t line;
-    int32_t column;
-    CSlice<SizeT> function;
-    CSlice<SizeT> msg;
-    int64_t params[3];
-};
-
-constexpr int64_t NO_PARAM = 0;
-
-template<typename SizeT>
-void _raise_exception_helper(ExceptionId id,
-                             const char* filename,
-                             int32_t line,
-                             const char* function,
-                             const char* msg,
-                             int64_t param0,
-                             int64_t param1,
-                             int64_t param2) {
-    Exception<SizeT> e = {
-        .id = id,
-        .filename = {.base = reinterpret_cast<void*>(const_cast<char*>(filename)),
-                     .len = static_cast<SizeT>(__builtin_strlen(filename))},
-        .line = line,
-        .column = 0,
-        .function = {.base = reinterpret_cast<void*>(const_cast<char*>(function)),
-                     .len = static_cast<SizeT>(__builtin_strlen(function))},
-        .msg = {.base = reinterpret_cast<void*>(const_cast<char*>(msg)),
-                .len = static_cast<SizeT>(__builtin_strlen(msg))},
-    };
-    e.params[0] = param0;
-    e.params[1] = param1;
-    e.params[2] = param2;
-    __nac3_raise(reinterpret_cast<void*>(&e));
-    __builtin_unreachable();
-}
-}  // namespace
-
-/**
- * @brief Raise an exception with location details (location in the IRRT source files).
- * @param SizeT The runtime `size_t` type.
- * @param id The ID of the exception to raise.
- * @param msg A global constant C-string of the error message.
- *
- * `param0` to `param2` are optional format arguments of `msg`. They should be set to
- * `NO_PARAM` to indicate they are unused.
- */
-#define raise_exception(SizeT, id, msg, param0, param1, param2) \
-    _raise_exception_helper<SizeT>(id, __FILE__, __LINE__, __FUNCTION__, msg, param0, param1, param2)

nac3core/irrt/irrt/int_types.hpp

@@ -1,25 +0,0 @@
-#pragma once
-
-#if __STDC_VERSION__ >= 202000
-using int8_t = _BitInt(8);
-using uint8_t = unsigned _BitInt(8);
-using int32_t = _BitInt(32);
-using uint32_t = unsigned _BitInt(32);
-using int64_t = _BitInt(64);
-using uint64_t = unsigned _BitInt(64);
-#else
-
-#pragma clang diagnostic push
-#pragma clang diagnostic ignored "-Wdeprecated-type"
-using int8_t = _ExtInt(8);
-using uint8_t = unsigned _ExtInt(8);
-using int32_t = _ExtInt(32);
-using uint32_t = unsigned _ExtInt(32);
-using int64_t = _ExtInt(64);
-using uint64_t = unsigned _ExtInt(64);
-#pragma clang diagnostic pop
-
-#endif
-
-// The type of an index or a value describing the length of a range/slice is always `int32_t`.
-using SliceIndex = int32_t;

nac3core/irrt/irrt/list.hpp

@@ -1,96 +0,0 @@
-#pragma once
-
-#include "irrt/int_types.hpp"
-#include "irrt/math_util.hpp"
-#include "irrt/slice.hpp"
-
-namespace {
-/**
- * @brief A list in NAC3.
- *
- * The `items` field is opaque. You must rely on external contexts to
- * know how to interpret it.
- */
-template<typename SizeT>
-struct List {
-    uint8_t* items;
-    SizeT len;
-};
-}  // namespace
-
-extern "C" {
-// Handle list assignment and dropping part of the list when
-// both dest_step and src_step are +1.
-// - All the index must *not* be out-of-bound or negative,
-// - The end index is *inclusive*,
-// - The length of src and dest slice size should already
-// be checked: if dest.step == 1 then len(src) <= len(dest) else len(src) == len(dest)
-SliceIndex __nac3_list_slice_assign_var_size(SliceIndex dest_start,
-                                             SliceIndex dest_end,
-                                             SliceIndex dest_step,
-                                             void* dest_arr,
-                                             SliceIndex dest_arr_len,
-                                             SliceIndex src_start,
-                                             SliceIndex src_end,
-                                             SliceIndex src_step,
-                                             void* src_arr,
-                                             SliceIndex src_arr_len,
-                                             const SliceIndex size) {
-    /* if dest_arr_len == 0, do nothing since we do not support extending list */
-    if (dest_arr_len == 0)
-        return dest_arr_len;
-    /* if both step is 1, memmove directly, handle the dropping of the list, and shrink size */
-    if (src_step == dest_step && dest_step == 1) {
-        const SliceIndex src_len = (src_end >= src_start) ? (src_end - src_start + 1) : 0;
-        const SliceIndex dest_len = (dest_end >= dest_start) ? (dest_end - dest_start + 1) : 0;
-        if (src_len > 0) {
-            __builtin_memmove(static_cast<uint8_t*>(dest_arr) + dest_start * size,
-                              static_cast<uint8_t*>(src_arr) + src_start * size, src_len * size);
-        }
-        if (dest_len > 0) {
-            /* dropping */
-            __builtin_memmove(static_cast<uint8_t*>(dest_arr) + (dest_start + src_len) * size,
-                              static_cast<uint8_t*>(dest_arr) + (dest_end + 1) * size,
-                              (dest_arr_len - dest_end - 1) * size);
-        }
-        /* shrink size */
-        return dest_arr_len - (dest_len - src_len);
-    }
-    /* if two range overlaps, need alloca */
-    uint8_t need_alloca = (dest_arr == src_arr)
-                          && !(max(dest_start, dest_end) < min(src_start, src_end)
-                               || max(src_start, src_end) < min(dest_start, dest_end));
-    if (need_alloca) {
-        void* tmp = __builtin_alloca(src_arr_len * size);
-        __builtin_memcpy(tmp, src_arr, src_arr_len * size);
-        src_arr = tmp;
-    }
-    SliceIndex src_ind = src_start;
-    SliceIndex dest_ind = dest_start;
-    for (; (src_step > 0) ? (src_ind <= src_end) : (src_ind >= src_end); src_ind += src_step, dest_ind += dest_step) {
-        /* for constant optimization */
-        if (size == 1) {
-            __builtin_memcpy(static_cast<uint8_t*>(dest_arr) + dest_ind, static_cast<uint8_t*>(src_arr) + src_ind, 1);
-        } else if (size == 4) {
-            __builtin_memcpy(static_cast<uint8_t*>(dest_arr) + dest_ind * 4,
-                             static_cast<uint8_t*>(src_arr) + src_ind * 4, 4);
-        } else if (size == 8) {
-            __builtin_memcpy(static_cast<uint8_t*>(dest_arr) + dest_ind * 8,
-                             static_cast<uint8_t*>(src_arr) + src_ind * 8, 8);
-        } else {
-            /* memcpy for var size, cannot overlap after previous alloca */
-            __builtin_memcpy(static_cast<uint8_t*>(dest_arr) + dest_ind * size,
-                             static_cast<uint8_t*>(src_arr) + src_ind * size, size);
-        }
-    }
-    /* only dest_step == 1 can we shrink the dest list. */
-    /* size should be ensured prior to calling this function */
-    if (dest_step == 1 && dest_end >= dest_start) {
-        __builtin_memmove(static_cast<uint8_t*>(dest_arr) + dest_ind * size,
-                          static_cast<uint8_t*>(dest_arr) + (dest_end + 1) * size,
-                          (dest_arr_len - dest_end - 1) * size);
-        return dest_arr_len - (dest_end - dest_ind) - 1;
-    }
-    return dest_arr_len;
-}
-}  // extern "C"

nac3core/irrt/irrt/math.hpp

@@ -1,62 +0,0 @@
-#pragma once
-
-#include "irrt/int_types.hpp"
-
-namespace {
-// adapted from GNU Scientific Library: https://git.savannah.gnu.org/cgit/gsl.git/tree/sys/pow_int.c
-// need to make sure `exp >= 0` before calling this function
-template<typename T>
-T __nac3_int_exp_impl(T base, T exp) {
-    T res = 1;
-    /* repeated squaring method */
-    do {
-        if (exp & 1) {
-            res *= base; /* for n odd */
-        }
-        exp >>= 1;
-        base *= base;
-    } while (exp);
-    return res;
-}
-}  // namespace
-
-#define DEF_nac3_int_exp_(T)                   \
-    T __nac3_int_exp_##T(T base, T exp) {      \
-        return __nac3_int_exp_impl(base, exp); \
-    }
-
-extern "C" {
-
-// Putting semicolons here to make clang-format not reformat this into
-// a stair shape.
-DEF_nac3_int_exp_(int32_t);
-DEF_nac3_int_exp_(int64_t);
-DEF_nac3_int_exp_(uint32_t);
-DEF_nac3_int_exp_(uint64_t);
-
-double __nac3_gammaln(double x) {
-    // libm's handling of value overflows differs from scipy:
-    // - scipy: gammaln(-inf) -> -inf
-    // - libm : lgamma(-inf) -> inf
-
-    if (__builtin_isinf(x)) {
-        return x;
-    }
-
-    return __builtin_lgamma(x);
-}
-
-double j0(double x);
-
-double __nac3_j0(double x) {
-    // libm's handling of value overflows differs from scipy:
-    // - scipy: j0(inf) -> nan
-    // - libm : j0(inf) -> 0.0
-
-    if (__builtin_isinf(x)) {
-        return __builtin_nan("");
-    }
-
-    return j0(x);
-}
-}  // namespace

nac3core/irrt/irrt/math_util.hpp

@@ -1,13 +0,0 @@
-#pragma once
-
-namespace {
-template<typename T>
-const T& max(const T& a, const T& b) {
-    return a > b ? a : b;
-}
-
-template<typename T>
-const T& min(const T& a, const T& b) {
-    return a > b ? b : a;
-}
-}  // namespace

nac3core/irrt/irrt/ndarray/array.hpp

@@ -1,132 +0,0 @@
-#pragma once
-
-#include "irrt/debug.hpp"
-#include "irrt/exception.hpp"
-#include "irrt/int_types.hpp"
-#include "irrt/list.hpp"
-#include "irrt/ndarray/basic.hpp"
-#include "irrt/ndarray/def.hpp"
-
-namespace {
-namespace ndarray::array {
-/**
- * @brief In the context of `np.array(<list>)`, deduce the ndarray's shape produced by `<list>` and raise
- * an exception if there is anything wrong with `<shape>` (e.g., inconsistent dimensions `np.array([[1.0, 2.0],
- * [3.0]])`)
- *
- * If this function finds no issues with `<list>`, the deduced shape is written to `shape`. The caller has the
- * responsibility to allocate `[SizeT; ndims]` for `shape`. The caller must also initialize `shape` with `-1`s because
- * of implementation details.
- */
-template<typename SizeT>
-void set_and_validate_list_shape_helper(SizeT axis, List<SizeT>* list, SizeT ndims, SizeT* shape) {
-    if (shape[axis] == -1) {
-        // Dimension is unspecified. Set it.
-        shape[axis] = list->len;
-    } else {
-        // Dimension is specified. Check.
-        if (shape[axis] != list->len) {
-            // Mismatch, throw an error.
-            // NOTE: NumPy's error message is more complex and needs more PARAMS to display.
-            raise_exception(SizeT, EXN_VALUE_ERROR,
-                            "The requested array has an inhomogenous shape "
-                            "after {0} dimension(s).",
-                            axis, shape[axis], list->len);
-        }
-    }
-
-    if (axis + 1 == ndims) {
-        // `list` has type `list[ItemType]`
-        // Do nothing
-    } else {
-        // `list` has type `list[list[...]]`
-        List<SizeT>** lists = (List<SizeT>**)(list->items);
-        for (SizeT i = 0; i < list->len; i++) {
-            set_and_validate_list_shape_helper<SizeT>(axis + 1, lists[i], ndims, shape);
-        }
-    }
-}
-
-/**
- * @brief See `set_and_validate_list_shape_helper`.
- */
-template<typename SizeT>
-void set_and_validate_list_shape(List<SizeT>* list, SizeT ndims, SizeT* shape) {
-    for (SizeT axis = 0; axis < ndims; axis++) {
-        shape[axis] = -1;  // Sentinel to say this dimension is unspecified.
-    }
-    set_and_validate_list_shape_helper<SizeT>(0, list, ndims, shape);
-}
-
-/**
- * @brief In the context of `np.array(<list>)`, copied the contents stored in `list` to `ndarray`.
- *
- * `list` is assumed to be "legal". (i.e., no inconsistent dimensions)
- *
- * # Notes on `ndarray`
- * The caller is responsible for allocating space for `ndarray`.
- * Here is what this function expects from `ndarray` when called:
- *   - `ndarray->data` has to be allocated, contiguous, and may contain uninitialized values.
- *   - `ndarray->itemsize` has to be initialized.
- *   - `ndarray->ndims` has to be initialized.
- *   - `ndarray->shape` has to be initialized.
- *   - `ndarray->strides` is ignored, but note that `ndarray->data` is contiguous.
- * When this function call ends:
- *   - `ndarray->data` is written with contents from `<list>`.
- */
-template<typename SizeT>
-void write_list_to_array_helper(SizeT axis, SizeT* index, List<SizeT>* list, NDArray<SizeT>* ndarray) {
-    debug_assert_eq(SizeT, list->len, ndarray->shape[axis]);
-    if (IRRT_DEBUG_ASSERT_BOOL) {
-        if (!ndarray::basic::is_c_contiguous(ndarray)) {
-            raise_debug_assert(SizeT, "ndarray is not C-contiguous", ndarray->strides[0], ndarray->strides[1],
-                               NO_PARAM);
-        }
-    }
-
-    if (axis + 1 == ndarray->ndims) {
-        // `list` has type `list[scalar]`
-        // `ndarray` is contiguous, so we can do this, and this is fast.
-        uint8_t* dst = static_cast<uint8_t*>(ndarray->data) + (ndarray->itemsize * (*index));
-        __builtin_memcpy(dst, list->items, ndarray->itemsize * list->len);
-        *index += list->len;
-    } else {
-        // `list` has type `list[list[...]]`
-        List<SizeT>** lists = (List<SizeT>**)(list->items);
-
-        for (SizeT i = 0; i < list->len; i++) {
-            write_list_to_array_helper<SizeT>(axis + 1, index, lists[i], ndarray);
-        }
-    }
-}
-
-/**
- * @brief See `write_list_to_array_helper`.
- */
-template<typename SizeT>
-void write_list_to_array(List<SizeT>* list, NDArray<SizeT>* ndarray) {
-    SizeT index = 0;
-    write_list_to_array_helper<SizeT>((SizeT)0, &index, list, ndarray);
-}
-}  // namespace ndarray::array
-}  // namespace
-
-extern "C" {
-using namespace ndarray::array;
-
-void __nac3_ndarray_array_set_and_validate_list_shape(List<int32_t>* list, int32_t ndims, int32_t* shape) {
-    set_and_validate_list_shape(list, ndims, shape);
-}
-
-void __nac3_ndarray_array_set_and_validate_list_shape64(List<int64_t>* list, int64_t ndims, int64_t* shape) {
-    set_and_validate_list_shape(list, ndims, shape);
-}
-
-void __nac3_ndarray_array_write_list_to_array(List<int32_t>* list, NDArray<int32_t>* ndarray) {
-    write_list_to_array(list, ndarray);
-}
-
-void __nac3_ndarray_array_write_list_to_array64(List<int64_t>* list, NDArray<int64_t>* ndarray) {
-    write_list_to_array(list, ndarray);
-}
-}

nac3core/irrt/irrt/ndarray/basic.hpp

@@ -1,340 +0,0 @@
-#pragma once
-
-#include "irrt/debug.hpp"
-#include "irrt/exception.hpp"
-#include "irrt/int_types.hpp"
-#include "irrt/ndarray/def.hpp"
-
-namespace {
-namespace ndarray::basic {
-/**
- * @brief Assert that `shape` does not contain negative dimensions.
- *
- * @param ndims Number of dimensions in `shape`
- * @param shape The shape to check on
- */
-template<typename SizeT>
-void assert_shape_no_negative(SizeT ndims, const SizeT* shape) {
-    for (SizeT axis = 0; axis < ndims; axis++) {
-        if (shape[axis] < 0) {
-            raise_exception(SizeT, EXN_VALUE_ERROR,
-                            "negative dimensions are not allowed; axis {0} "
-                            "has dimension {1}",
-                            axis, shape[axis], NO_PARAM);
-        }
-    }
-}
-
-/**
- * @brief Assert that two shapes are the same in the context of writing output to an ndarray.
- */
-template<typename SizeT>
-void assert_output_shape_same(SizeT ndarray_ndims,
-                              const SizeT* ndarray_shape,
-                              SizeT output_ndims,
-                              const SizeT* output_shape) {
-    if (ndarray_ndims != output_ndims) {
-        // There is no corresponding NumPy error message like this.
-        raise_exception(SizeT, EXN_VALUE_ERROR, "Cannot write output of ndims {0} to an ndarray with ndims {1}",
-                        output_ndims, ndarray_ndims, NO_PARAM);
-    }
-
-    for (SizeT axis = 0; axis < ndarray_ndims; axis++) {
-        if (ndarray_shape[axis] != output_shape[axis]) {
-            // There is no corresponding NumPy error message like this.
-            raise_exception(SizeT, EXN_VALUE_ERROR,
-                            "Mismatched dimensions on axis {0}, output has "
-                            "dimension {1}, but destination ndarray has dimension {2}.",
-                            axis, output_shape[axis], ndarray_shape[axis]);
-        }
-    }
-}
-
-/**
- * @brief Return the number of elements of an ndarray given its shape.
- *
- * @param ndims Number of dimensions in `shape`
- * @param shape The shape of the ndarray
- */
-template<typename SizeT>
-SizeT calc_size_from_shape(SizeT ndims, const SizeT* shape) {
-    SizeT size = 1;
-    for (SizeT axis = 0; axis < ndims; axis++)
-        size *= shape[axis];
-    return size;
-}
-
-/**
- * @brief Compute the array indices of the `nth` (0-based) element of an ndarray given only its shape.
- *
- * @param ndims Number of elements in `shape` and `indices`
- * @param shape The shape of the ndarray
- * @param indices The returned indices indexing the ndarray with shape `shape`.
- * @param nth The index of the element of interest.
- */
-template<typename SizeT>
-void set_indices_by_nth(SizeT ndims, const SizeT* shape, SizeT* indices, SizeT nth) {
-    for (SizeT i = 0; i < ndims; i++) {
-        SizeT axis = ndims - i - 1;
-        SizeT dim = shape[axis];
-
-        indices[axis] = nth % dim;
-        nth /= dim;
-    }
-}
-
-/**
- * @brief Return the number of elements of an `ndarray`
- *
- * This function corresponds to `<an_ndarray>.size`
- */
-template<typename SizeT>
-SizeT size(const NDArray<SizeT>* ndarray) {
-    return calc_size_from_shape(ndarray->ndims, ndarray->shape);
-}
-
-/**
- * @brief Return of the number of its content of an `ndarray`.
- *
- * This function corresponds to `<an_ndarray>.nbytes`.
- */
-template<typename SizeT>
-SizeT nbytes(const NDArray<SizeT>* ndarray) {
-    return size(ndarray) * ndarray->itemsize;
-}
-
-/**
- * @brief Get the `len()` of an ndarray, and asserts that `ndarray` is a sized object.
- *
- * This function corresponds to `<an_ndarray>.__len__`.
- *
- * @param dst_length The length.
- */
-template<typename SizeT>
-SizeT len(const NDArray<SizeT>* ndarray) {
-    if (ndarray->ndims != 0) {
-        return ndarray->shape[0];
-    }
-
-    // numpy prohibits `__len__` on unsized objects
-    raise_exception(SizeT, EXN_TYPE_ERROR, "len() of unsized object", NO_PARAM, NO_PARAM, NO_PARAM);
-    __builtin_unreachable();
-}
-
-/**
- * @brief Return a boolean indicating if `ndarray` is (C-)contiguous.
- *
- * You may want to see ndarray's rules for C-contiguity:
- * https://github.com/numpy/numpy/blob/df256d0d2f3bc6833699529824781c58f9c6e697/numpy/core/src/multiarray/flagsobject.c#L95C1-L99C45
- */
-template<typename SizeT>
-bool is_c_contiguous(const NDArray<SizeT>* ndarray) {
-    // References:
-    // - tinynumpy's implementation:
-    // https://github.com/wadetb/tinynumpy/blob/0d23d22e07062ffab2afa287374c7b366eebdda1/tinynumpy/tinynumpy.py#L102
-    // - ndarray's flags["C_CONTIGUOUS"]:
-    // https://numpy.org/doc/stable/reference/generated/numpy.ndarray.flags.html#numpy.ndarray.flags
-    // - ndarray's rules for C-contiguity:
-    // https://github.com/numpy/numpy/blob/df256d0d2f3bc6833699529824781c58f9c6e697/numpy/core/src/multiarray/flagsobject.c#L95C1-L99C45
-
-    // From
-    // https://github.com/numpy/numpy/blob/df256d0d2f3bc6833699529824781c58f9c6e697/numpy/core/src/multiarray/flagsobject.c#L95C1-L99C45:
-    //
-    // The traditional rule is that for an array to be flagged as C contiguous,
-    // the following must hold:
-    //
-    // strides[-1] == itemsize
-    // strides[i] == shape[i+1] * strides[i + 1]
-    // [...]
-    // According to these rules, a 0- or 1-dimensional array is either both
-    // C- and F-contiguous, or neither; and an array with 2+ dimensions
-    // can be C- or F- contiguous, or neither, but not both. Though there
-    // there are exceptions for arrays with zero or one item, in the first
-    // case the check is relaxed up to and including the first dimension
-    // with shape[i] == 0. In the second case `strides == itemsize` will
-    // can be true for all dimensions and both flags are set.
-
-    if (ndarray->ndims == 0) {
-        return true;
-    }
-
-    if (ndarray->strides[ndarray->ndims - 1] != ndarray->itemsize) {
-        return false;
-    }
-
-    for (SizeT i = 1; i < ndarray->ndims; i++) {
-        SizeT axis_i = ndarray->ndims - i - 1;
-        if (ndarray->strides[axis_i] != ndarray->shape[axis_i + 1] * ndarray->strides[axis_i + 1]) {
-            return false;
-        }
-    }
-
-    return true;
-}
-
-/**
- * @brief Return the pointer to the element indexed by `indices` along the ndarray's axes.
- *
- * This function does no bound check.
- */
-template<typename SizeT>
-void* get_pelement_by_indices(const NDArray<SizeT>* ndarray, const SizeT* indices) {
-    void* element = ndarray->data;
-    for (SizeT dim_i = 0; dim_i < ndarray->ndims; dim_i++)
-        element = static_cast<uint8_t*>(element) + indices[dim_i] * ndarray->strides[dim_i];
-    return element;
-}
-
-/**
- * @brief Return the pointer to the nth (0-based) element of `ndarray` in flattened view.
- *
- * This function does no bound check.
- */
-template<typename SizeT>
-void* get_nth_pelement(const NDArray<SizeT>* ndarray, SizeT nth) {
-    void* element = ndarray->data;
-    for (SizeT i = 0; i < ndarray->ndims; i++) {
-        SizeT axis = ndarray->ndims - i - 1;
-        SizeT dim = ndarray->shape[axis];
-        element = static_cast<uint8_t*>(element) + ndarray->strides[axis] * (nth % dim);
-        nth /= dim;
-    }
-    return element;
-}
-
-/**
- * @brief Update the strides of an ndarray given an ndarray `shape` to be contiguous.
- *
- * You might want to read https://ajcr.net/stride-guide-part-1/.
- */
-template<typename SizeT>
-void set_strides_by_shape(NDArray<SizeT>* ndarray) {
-    SizeT stride_product = 1;
-    for (SizeT i = 0; i < ndarray->ndims; i++) {
-        SizeT axis = ndarray->ndims - i - 1;
-        ndarray->strides[axis] = stride_product * ndarray->itemsize;
-        stride_product *= ndarray->shape[axis];
-    }
-}
-
-/**
- * @brief Set an element in `ndarray`.
- *
- * @param pelement Pointer to the element in `ndarray` to be set.
- * @param pvalue Pointer to the value `pelement` will be set to.
- */
-template<typename SizeT>
-void set_pelement_value(NDArray<SizeT>* ndarray, void* pelement, const void* pvalue) {
-    __builtin_memcpy(pelement, pvalue, ndarray->itemsize);
-}
-
-/**
- * @brief Copy data from one ndarray to another of the exact same size and itemsize.
- *
- * Both ndarrays will be viewed in their flatten views when copying the elements.
- */
-template<typename SizeT>
-void copy_data(const NDArray<SizeT>* src_ndarray, NDArray<SizeT>* dst_ndarray) {
-    // TODO: Make this faster with memcpy when we see a contiguous segment.
-    // TODO: Handle overlapping.
-
-    debug_assert_eq(SizeT, src_ndarray->itemsize, dst_ndarray->itemsize);
-
-    for (SizeT i = 0; i < size(src_ndarray); i++) {
-        auto src_element = ndarray::basic::get_nth_pelement(src_ndarray, i);
-        auto dst_element = ndarray::basic::get_nth_pelement(dst_ndarray, i);
-        ndarray::basic::set_pelement_value(dst_ndarray, dst_element, src_element);
-    }
-}
-}  // namespace ndarray::basic
-}  // namespace
-
-extern "C" {
-using namespace ndarray::basic;
-
-void __nac3_ndarray_util_assert_shape_no_negative(int32_t ndims, int32_t* shape) {
-    assert_shape_no_negative(ndims, shape);
-}
-
-void __nac3_ndarray_util_assert_shape_no_negative64(int64_t ndims, int64_t* shape) {
-    assert_shape_no_negative(ndims, shape);
-}
-
-void __nac3_ndarray_util_assert_output_shape_same(int32_t ndarray_ndims,
-                                                  const int32_t* ndarray_shape,
-                                                  int32_t output_ndims,
-                                                  const int32_t* output_shape) {
-    assert_output_shape_same(ndarray_ndims, ndarray_shape, output_ndims, output_shape);
-}
-
-void __nac3_ndarray_util_assert_output_shape_same64(int64_t ndarray_ndims,
-                                                    const int64_t* ndarray_shape,
-                                                    int64_t output_ndims,
-                                                    const int64_t* output_shape) {
-    assert_output_shape_same(ndarray_ndims, ndarray_shape, output_ndims, output_shape);
-}
-
-uint32_t __nac3_ndarray_size(NDArray<int32_t>* ndarray) {
-    return size(ndarray);
-}
-
-uint64_t __nac3_ndarray_size64(NDArray<int64_t>* ndarray) {
-    return size(ndarray);
-}
-
-uint32_t __nac3_ndarray_nbytes(NDArray<int32_t>* ndarray) {
-    return nbytes(ndarray);
-}
-
-uint64_t __nac3_ndarray_nbytes64(NDArray<int64_t>* ndarray) {
-    return nbytes(ndarray);
-}
-
-int32_t __nac3_ndarray_len(NDArray<int32_t>* ndarray) {
-    return len(ndarray);
-}
-
-int64_t __nac3_ndarray_len64(NDArray<int64_t>* ndarray) {
-    return len(ndarray);
-}
-
-bool __nac3_ndarray_is_c_contiguous(NDArray<int32_t>* ndarray) {
-    return is_c_contiguous(ndarray);
-}
-
-bool __nac3_ndarray_is_c_contiguous64(NDArray<int64_t>* ndarray) {
-    return is_c_contiguous(ndarray);
-}
-
-void* __nac3_ndarray_get_nth_pelement(const NDArray<int32_t>* ndarray, int32_t nth) {
-    return get_nth_pelement(ndarray, nth);
-}
-
-void* __nac3_ndarray_get_nth_pelement64(const NDArray<int64_t>* ndarray, int64_t nth) {
-    return get_nth_pelement(ndarray, nth);
-}
-
-void* __nac3_ndarray_get_pelement_by_indices(const NDArray<int32_t>* ndarray, int32_t* indices) {
-    return get_pelement_by_indices(ndarray, indices);
-}
-
-void* __nac3_ndarray_get_pelement_by_indices64(const NDArray<int64_t>* ndarray, int64_t* indices) {
-    return get_pelement_by_indices(ndarray, indices);
-}
-
-void __nac3_ndarray_set_strides_by_shape(NDArray<int32_t>* ndarray) {
-    set_strides_by_shape(ndarray);
-}
-
-void __nac3_ndarray_set_strides_by_shape64(NDArray<int64_t>* ndarray) {
-    set_strides_by_shape(ndarray);
-}
-
-void __nac3_ndarray_copy_data(NDArray<int32_t>* src_ndarray, NDArray<int32_t>* dst_ndarray) {
-    copy_data(src_ndarray, dst_ndarray);
-}
-
-void __nac3_ndarray_copy_data64(NDArray<int64_t>* src_ndarray, NDArray<int64_t>* dst_ndarray) {
-    copy_data(src_ndarray, dst_ndarray);
-}
-}

nac3core/irrt/irrt/ndarray/broadcast.hpp

@@ -1,165 +0,0 @@
-#pragma once
-
-#include "irrt/int_types.hpp"
-#include "irrt/ndarray/def.hpp"
-#include "irrt/slice.hpp"
-
-namespace {
-template<typename SizeT>
-struct ShapeEntry {
-    SizeT ndims;
-    SizeT* shape;
-};
-}  // namespace
-
-namespace {
-namespace ndarray::broadcast {
-/**
- * @brief Return true if `src_shape` can broadcast to `dst_shape`.
- *
- * See https://numpy.org/doc/stable/user/basics.broadcasting.html
- */
-template<typename SizeT>
-bool can_broadcast_shape_to(SizeT target_ndims, const SizeT* target_shape, SizeT src_ndims, const SizeT* src_shape) {
-    if (src_ndims > target_ndims) {
-        return false;
-    }
-
-    for (SizeT i = 0; i < src_ndims; i++) {
-        SizeT target_dim = target_shape[target_ndims - i - 1];
-        SizeT src_dim = src_shape[src_ndims - i - 1];
-        if (!(src_dim == 1 || target_dim == src_dim)) {
-            return false;
-        }
-    }
-    return true;
-}
-
-/**
- * @brief Performs `np.broadcast_shapes(<shapes>)`
- *
- * @param num_shapes Number of entries in `shapes`
- * @param shapes The list of shape to do `np.broadcast_shapes` on.
- * @param dst_ndims The length of `dst_shape`.
- *   `dst_ndims` must be `max([shape.ndims for shape in shapes])`, but the caller has to calculate it/provide it.
- *   for this function since they should already know in order to allocate `dst_shape` in the first place.
- * @param dst_shape The resulting shape. Must be pre-allocated by the caller. This function calculate the result
- *   of `np.broadcast_shapes` and write it here.
- */
-template<typename SizeT>
-void broadcast_shapes(SizeT num_shapes, const ShapeEntry<SizeT>* shapes, SizeT dst_ndims, SizeT* dst_shape) {
-    for (SizeT dst_axis = 0; dst_axis < dst_ndims; dst_axis++) {
-        dst_shape[dst_axis] = 1;
-    }
-
-#ifdef IRRT_DEBUG_ASSERT
-    SizeT max_ndims_found = 0;
-#endif
-
-    for (SizeT i = 0; i < num_shapes; i++) {
-        ShapeEntry<SizeT> entry = shapes[i];
-
-        // Check pre-condition: `dst_ndims` must be `max([shape.ndims for shape in shapes])`
-        debug_assert(SizeT, entry.ndims <= dst_ndims);
-
-#ifdef IRRT_DEBUG_ASSERT
-        max_ndims_found = max(max_ndims_found, entry.ndims);
-#endif
-
-        for (SizeT j = 0; j < entry.ndims; j++) {
-            SizeT entry_axis = entry.ndims - j - 1;
-            SizeT dst_axis = dst_ndims - j - 1;
-
-            SizeT entry_dim = entry.shape[entry_axis];
-            SizeT dst_dim = dst_shape[dst_axis];
-
-            if (dst_dim == 1) {
-                dst_shape[dst_axis] = entry_dim;
-            } else if (entry_dim == 1 || entry_dim == dst_dim) {
-                // Do nothing
-            } else {
-                raise_exception(SizeT, EXN_VALUE_ERROR,
-                                "shape mismatch: objects cannot be broadcast "
-                                "to a single shape.",
-                                NO_PARAM, NO_PARAM, NO_PARAM);
-            }
-        }
-    }
-
-#ifdef IRRT_DEBUG_ASSERT
-    // Check pre-condition: `dst_ndims` must be `max([shape.ndims for shape in shapes])`
-    debug_assert_eq(SizeT, max_ndims_found, dst_ndims);
-#endif
-}
-
-/**
- * @brief Perform `np.broadcast_to(<ndarray>, <target_shape>)` and appropriate assertions.
- *
- * This function attempts to broadcast `src_ndarray` to a new shape defined by `dst_ndarray.shape`,
- * and return the result by modifying `dst_ndarray`.
- *
- * # Notes on `dst_ndarray`
- * The caller is responsible for allocating space for the resulting ndarray.
- * Here is what this function expects from `dst_ndarray` when called:
- *   - `dst_ndarray->data` does not have to be initialized.
- *   - `dst_ndarray->itemsize` does not have to be initialized.
- *   - `dst_ndarray->ndims` must be initialized, determining the length of `dst_ndarray->shape`
- *   - `dst_ndarray->shape` must be allocated, and must contain the desired target broadcast shape.
- *   - `dst_ndarray->strides` must be allocated, through it can contain uninitialized values.
- * When this function call ends:
- *   - `dst_ndarray->data` is set to `src_ndarray->data` (`dst_ndarray` is just a view to `src_ndarray`)
- *   - `dst_ndarray->itemsize` is set to `src_ndarray->itemsize`
- *   - `dst_ndarray->ndims` is unchanged.
- *   - `dst_ndarray->shape` is unchanged.
- *   - `dst_ndarray->strides` is updated accordingly by how ndarray broadcast_to works.
- */
-template<typename SizeT>
-void broadcast_to(const NDArray<SizeT>* src_ndarray, NDArray<SizeT>* dst_ndarray) {
-    if (!ndarray::broadcast::can_broadcast_shape_to(dst_ndarray->ndims, dst_ndarray->shape, src_ndarray->ndims,
-                                                    src_ndarray->shape)) {
-        raise_exception(SizeT, EXN_VALUE_ERROR, "operands could not be broadcast together", NO_PARAM, NO_PARAM,
-                        NO_PARAM);
-    }
-
-    dst_ndarray->data = src_ndarray->data;
-    dst_ndarray->itemsize = src_ndarray->itemsize;
-
-    for (SizeT i = 0; i < dst_ndarray->ndims; i++) {
-        SizeT src_axis = src_ndarray->ndims - i - 1;
-        SizeT dst_axis = dst_ndarray->ndims - i - 1;
-        if (src_axis < 0 || (src_ndarray->shape[src_axis] == 1 && dst_ndarray->shape[dst_axis] != 1)) {
-            // Freeze the steps in-place
-            dst_ndarray->strides[dst_axis] = 0;
-        } else {
-            dst_ndarray->strides[dst_axis] = src_ndarray->strides[src_axis];
-        }
-    }
-}
-}  // namespace ndarray::broadcast
-}  // namespace
-
-extern "C" {
-using namespace ndarray::broadcast;
-
-void __nac3_ndarray_broadcast_to(NDArray<int32_t>* src_ndarray, NDArray<int32_t>* dst_ndarray) {
-    broadcast_to(src_ndarray, dst_ndarray);
-}
-
-void __nac3_ndarray_broadcast_to64(NDArray<int64_t>* src_ndarray, NDArray<int64_t>* dst_ndarray) {
-    broadcast_to(src_ndarray, dst_ndarray);
-}
-
-void __nac3_ndarray_broadcast_shapes(int32_t num_shapes,
-                                     const ShapeEntry<int32_t>* shapes,
-                                     int32_t dst_ndims,
-                                     int32_t* dst_shape) {
-    broadcast_shapes(num_shapes, shapes, dst_ndims, dst_shape);
-}
-
-void __nac3_ndarray_broadcast_shapes64(int64_t num_shapes,
-                                       const ShapeEntry<int64_t>* shapes,
-                                       int64_t dst_ndims,
-                                       int64_t* dst_shape) {
-    broadcast_shapes(num_shapes, shapes, dst_ndims, dst_shape);
-}
-}

nac3core/irrt/irrt/ndarray/def.hpp

@@ -1,51 +0,0 @@
-#pragma once
-
-#include "irrt/int_types.hpp"
-
-namespace {
-/**
- * @brief The NDArray object
- *
- * Official numpy implementation:
- * https://github.com/numpy/numpy/blob/735a477f0bc2b5b84d0e72d92f224bde78d4e069/doc/source/reference/c-api/types-and-structures.rst#pyarrayinterface
- *
- * Note that this implementation is based on `PyArrayInterface` rather of `PyArrayObject`. The
- * difference between `PyArrayInterface` and `PyArrayObject` (relevant to our implementation) is
- * that `PyArrayInterface` *has* `itemsize` and uses `void*` for its `data`, whereas `PyArrayObject`
- * does not require `itemsize` (probably using `strides[-1]` instead) and uses `char*` for its
- * `data`. There are also minor differences in the struct layout.
- */
-template<typename SizeT>
-struct NDArray {
-    /**
-     * @brief The number of bytes of a single element in `data`.
-     */
-    SizeT itemsize;
-
-    /**
-     * @brief The number of dimensions of this shape.
-     */
-    SizeT ndims;
-
-    /**
-     * @brief The NDArray shape, with length equal to `ndims`.
-     *
-     * Note that it may contain 0.
-     */
-    SizeT* shape;
-
-    /**
-     * @brief Array strides, with length equal to `ndims`
-     *
-     * The stride values are in units of bytes, not number of elements.
-     *
-     * Note that `strides` can have negative values or contain 0.
-     */
-    SizeT* strides;
-
-    /**
-     * @brief The underlying data this `ndarray` is pointing to.
-     */
-    void* data;
-};
-}  // namespace

nac3core/irrt/irrt/ndarray/indexing.hpp

@@ -1,219 +0,0 @@
-#pragma once
-
-#include "irrt/exception.hpp"
-#include "irrt/int_types.hpp"
-#include "irrt/ndarray/basic.hpp"
-#include "irrt/ndarray/def.hpp"
-#include "irrt/range.hpp"
-#include "irrt/slice.hpp"
-
-namespace {
-typedef uint8_t NDIndexType;
-
-/**
- * @brief A single element index
- *
- * `data` points to a `int32_t`.
- */
-const NDIndexType ND_INDEX_TYPE_SINGLE_ELEMENT = 0;
-
-/**
- * @brief A slice index
- *
- * `data` points to a `Slice<int32_t>`.
- */
-const NDIndexType ND_INDEX_TYPE_SLICE = 1;
-
-/**
- * @brief `np.newaxis` / `None`
- *
- * `data` is unused.
- */
-const NDIndexType ND_INDEX_TYPE_NEWAXIS = 2;
-
-/**
- * @brief `Ellipsis` / `...`
- *
- * `data` is unused.
- */
-const NDIndexType ND_INDEX_TYPE_ELLIPSIS = 3;
-
-/**
- * @brief An index used in ndarray indexing
- *
- * That is:
- * ```
- * my_ndarray[::-1, 3, ..., np.newaxis]
- *            ^^^^  ^  ^^^  ^^^^^^^^^^ each of these is represented by an NDIndex.
- * ```
- */
-struct NDIndex {
-    /**
-     * @brief Enum tag to specify the type of index.
-     *
-     * Please see the comment of each enum constant.
-     */
-    NDIndexType type;
-
-    /**
-     * @brief The accompanying data associated with `type`.
-     *
-     * Please see the comment of each enum constant.
-     */
-    uint8_t* data;
-};
-}  // namespace
-
-namespace {
-namespace ndarray::indexing {
-/**
- * @brief Perform ndarray "basic indexing" (https://numpy.org/doc/stable/user/basics.indexing.html#basic-indexing)
- *
- * This function is very similar to performing `dst_ndarray = src_ndarray[indices]` in Python.
- *
- * This function also does proper assertions on `indices` to check for out of bounds access and more.
- *
- * # Notes on `dst_ndarray`
- * The caller is responsible for allocating space for the resulting ndarray.
- * Here is what this function expects from `dst_ndarray` when called:
- *   - `dst_ndarray->data` does not have to be initialized.
- *   - `dst_ndarray->itemsize` does not have to be initialized.
- *   - `dst_ndarray->ndims` must be initialized, and it must be equal to the expected `ndims` of the `dst_ndarray` after
- *       indexing `src_ndarray` with `indices`.
- *   - `dst_ndarray->shape` must be allocated, through it can contain uninitialized values.
- *   - `dst_ndarray->strides` must be allocated, through it can contain uninitialized values.
- * When this function call ends:
- *   - `dst_ndarray->data` is set to `src_ndarray->data`.
- *   - `dst_ndarray->itemsize` is set to `src_ndarray->itemsize`.
- *   - `dst_ndarray->ndims` is unchanged.
- *   - `dst_ndarray->shape` is updated according to how `src_ndarray` is indexed.
- *   - `dst_ndarray->strides` is updated accordingly by how ndarray indexing works.
- *
- * @param indices indices to index `src_ndarray`, ordered in the same way you would write them in Python.
- * @param src_ndarray The NDArray to be indexed.
- * @param dst_ndarray The resulting NDArray after indexing. Further details in the comments above,
- */
-template<typename SizeT>
-void index(SizeT num_indices, const NDIndex* indices, const NDArray<SizeT>* src_ndarray, NDArray<SizeT>* dst_ndarray) {
-    // Validate `indices`.
-
-    // Expected value of `dst_ndarray->ndims`.
-    SizeT expected_dst_ndims = src_ndarray->ndims;
-    // To check for "too many indices for array: array is ?-dimensional, but ? were indexed"
-    SizeT num_indexed = 0;
-    // There may be ellipsis `...` in `indices`. There can only be 0 or 1 ellipsis.
-    SizeT num_ellipsis = 0;
-
-    for (SizeT i = 0; i < num_indices; i++) {
-        if (indices[i].type == ND_INDEX_TYPE_SINGLE_ELEMENT) {
-            expected_dst_ndims--;
-            num_indexed++;
-        } else if (indices[i].type == ND_INDEX_TYPE_SLICE) {
-            num_indexed++;
-        } else if (indices[i].type == ND_INDEX_TYPE_NEWAXIS) {
-            expected_dst_ndims++;
-        } else if (indices[i].type == ND_INDEX_TYPE_ELLIPSIS) {
-            num_ellipsis++;
-            if (num_ellipsis > 1) {
-                raise_exception(SizeT, EXN_INDEX_ERROR, "an index can only have a single ellipsis ('...')", NO_PARAM,
-                                NO_PARAM, NO_PARAM);
-            }
-        } else {
-            __builtin_unreachable();
-        }
-    }
-
-    debug_assert_eq(SizeT, expected_dst_ndims, dst_ndarray->ndims);
-
-    if (src_ndarray->ndims - num_indexed < 0) {
-        raise_exception(SizeT, EXN_INDEX_ERROR,
-                        "too many indices for array: array is {0}-dimensional, "
-                        "but {1} were indexed",
-                        src_ndarray->ndims, num_indices, NO_PARAM);
-    }
-
-    dst_ndarray->data = src_ndarray->data;
-    dst_ndarray->itemsize = src_ndarray->itemsize;
-
-    // Reference code:
-    // https://github.com/wadetb/tinynumpy/blob/0d23d22e07062ffab2afa287374c7b366eebdda1/tinynumpy/tinynumpy.py#L652
-    SizeT src_axis = 0;
-    SizeT dst_axis = 0;
-
-    for (int32_t i = 0; i < num_indices; i++) {
-        const NDIndex* index = &indices[i];
-        if (index->type == ND_INDEX_TYPE_SINGLE_ELEMENT) {
-            SizeT input = (SizeT) * ((int32_t*)index->data);
-
-            SizeT k = slice::resolve_index_in_length(src_ndarray->shape[src_axis], input);
-            if (k == -1) {
-                raise_exception(SizeT, EXN_INDEX_ERROR,
-                                "index {0} is out of bounds for axis {1} "
-                                "with size {2}",
-                                input, src_axis, src_ndarray->shape[src_axis]);
-            }
-
-            dst_ndarray->data = static_cast<uint8_t*>(dst_ndarray->data) + k * src_ndarray->strides[src_axis];
-
-            src_axis++;
-        } else if (index->type == ND_INDEX_TYPE_SLICE) {
-            Slice<int32_t>* slice = (Slice<int32_t>*)index->data;
-
-            Range<int32_t> range = slice->indices_checked<SizeT>(src_ndarray->shape[src_axis]);
-
-            dst_ndarray->data =
-                static_cast<uint8_t*>(dst_ndarray->data) + (SizeT)range.start * src_ndarray->strides[src_axis];
-            dst_ndarray->strides[dst_axis] = ((SizeT)range.step) * src_ndarray->strides[src_axis];
-            dst_ndarray->shape[dst_axis] = (SizeT)range.len<SizeT>();
-
-            dst_axis++;
-            src_axis++;
-        } else if (index->type == ND_INDEX_TYPE_NEWAXIS) {
-            dst_ndarray->strides[dst_axis] = 0;
-            dst_ndarray->shape[dst_axis] = 1;
-
-            dst_axis++;
-        } else if (index->type == ND_INDEX_TYPE_ELLIPSIS) {
-            // The number of ':' entries this '...' implies.
-            SizeT ellipsis_size = src_ndarray->ndims - num_indexed;
-
-            for (SizeT j = 0; j < ellipsis_size; j++) {
-                dst_ndarray->strides[dst_axis] = src_ndarray->strides[src_axis];
-                dst_ndarray->shape[dst_axis] = src_ndarray->shape[src_axis];
-
-                dst_axis++;
-                src_axis++;
-            }
-        } else {
-            __builtin_unreachable();
-        }
-    }
-
-    for (; dst_axis < dst_ndarray->ndims; dst_axis++, src_axis++) {
-        dst_ndarray->shape[dst_axis] = src_ndarray->shape[src_axis];
-        dst_ndarray->strides[dst_axis] = src_ndarray->strides[src_axis];
-    }
-
-    debug_assert_eq(SizeT, src_ndarray->ndims, src_axis);
-    debug_assert_eq(SizeT, dst_ndarray->ndims, dst_axis);
-}
-}  // namespace ndarray::indexing
-}  // namespace
-
-extern "C" {
-using namespace ndarray::indexing;
-
-void __nac3_ndarray_index(int32_t num_indices,
-                          NDIndex* indices,
-                          NDArray<int32_t>* src_ndarray,
-                          NDArray<int32_t>* dst_ndarray) {
-    index(num_indices, indices, src_ndarray, dst_ndarray);
-}
-
-void __nac3_ndarray_index64(int64_t num_indices,
-                            NDIndex* indices,
-                            NDArray<int64_t>* src_ndarray,
-                            NDArray<int64_t>* dst_ndarray) {
-    index(num_indices, indices, src_ndarray, dst_ndarray);
-}
-}

nac3core/irrt/irrt/ndarray/iter.hpp

@@ -1,146 +0,0 @@
-#pragma once
-
-#include "irrt/int_types.hpp"
-#include "irrt/ndarray/def.hpp"
-
-namespace {
-/**
- * @brief Helper struct to enumerate through an ndarray *efficiently*.
- *
- * Example usage (in pseudo-code):
- * ```
- * // Suppose my_ndarray has been initialized, with shape [2, 3] and dtype `double`
- * NDIter nditer;
- * nditer.initialize(my_ndarray);
- * while (nditer.has_element()) {
- *    // This body is run 6 (= my_ndarray.size) times.
- *
- *    // [0, 0] -> [0, 1] -> [0, 2] -> [1, 0] -> [1, 1] -> [1, 2] -> end
- *    print(nditer.indices);
- *
- *    // 0 -> 1 -> 2 -> 3 -> 4 -> 5
- *    print(nditer.nth);
- *
- *    // <1st element> -> <2nd element> -> ... -> <6th element> -> end
- *    print(*((double *) nditer.element))
- *
- *    nditer.next(); // Go to next element.
- * }
- * ```
- *
- * Interesting cases:
- * - If `my_ndarray.ndims` == 0, there is one iteration.
- * - If `my_ndarray.shape` contains zeroes, there are no iterations.
- */
-template<typename SizeT>
-struct NDIter {
-    // Information about the ndarray being iterated over.
-    SizeT ndims;
-    SizeT* shape;
-    SizeT* strides;
-
-    /**
-     * @brief The current indices.
-     *
-     * Must be allocated by the caller.
-     */
-    SizeT* indices;
-
-    /**
-     * @brief The nth (0-based) index of the current indices.
-     *
-     * Initially this is 0.
-     */
-    SizeT nth;
-
-    /**
-     * @brief Pointer to the current element.
-     *
-     * Initially this points to first element of the ndarray.
-     */
-    void* element;
-
-    /**
-     * @brief Cache for the product of shape.
-     *
-     * Could be 0 if `shape` has 0s in it.
-     */
-    SizeT size;
-
-    void initialize(SizeT ndims, SizeT* shape, SizeT* strides, void* element, SizeT* indices) {
-        this->ndims = ndims;
-        this->shape = shape;
-        this->strides = strides;
-
-        this->indices = indices;
-        this->element = element;
-
-        // Compute size
-        this->size = 1;
-        for (SizeT i = 0; i < ndims; i++) {
-            this->size *= shape[i];
-        }
-
-        // `indices` starts on all 0s.
-        for (SizeT axis = 0; axis < ndims; axis++)
-            indices[axis] = 0;
-        nth = 0;
-    }
-
-    void initialize_by_ndarray(NDArray<SizeT>* ndarray, SizeT* indices) {
-        // NOTE: ndarray->data is pointing to the first element, and `NDIter`'s `element` should also point to the first
-        // element as well.
-        this->initialize(ndarray->ndims, ndarray->shape, ndarray->strides, ndarray->data, indices);
-    }
-
-    // Is the current iteration valid?
-    // If true, then `element`, `indices` and `nth` contain details about the current element.
-    bool has_element() { return nth < size; }
-
-    // Go to the next element.
-    void next() {
-        for (SizeT i = 0; i < ndims; i++) {
-            SizeT axis = ndims - i - 1;
-            indices[axis]++;
-            if (indices[axis] >= shape[axis]) {
-                indices[axis] = 0;
-
-                // TODO: There is something called backstrides to speedup iteration.
-                // See https://ajcr.net/stride-guide-part-1/, and
-                // https://docs.scipy.org/doc/numpy-1.13.0/reference/c-api.types-and-structures.html#c.PyArrayIterObject.PyArrayIterObject.backstrides.
-                element = static_cast<void*>(reinterpret_cast<uint8_t*>(element) - strides[axis] * (shape[axis] - 1));
-            } else {
-                element = static_cast<void*>(reinterpret_cast<uint8_t*>(element) + strides[axis]);
-                break;
-            }
-        }
-        nth++;
-    }
-};
-}  // namespace
-
-extern "C" {
-void __nac3_nditer_initialize(NDIter<int32_t>* iter, NDArray<int32_t>* ndarray, int32_t* indices) {
-    iter->initialize_by_ndarray(ndarray, indices);
-}
-
-void __nac3_nditer_initialize64(NDIter<int64_t>* iter, NDArray<int64_t>* ndarray, int64_t* indices) {
-    iter->initialize_by_ndarray(ndarray, indices);
-}
-
-bool __nac3_nditer_has_element(NDIter<int32_t>* iter) {
-    return iter->has_element();
-}
-
-bool __nac3_nditer_has_element64(NDIter<int64_t>* iter) {
-    return iter->has_element();
-}
-
-void __nac3_nditer_next(NDIter<int32_t>* iter) {
-    iter->next();
-}
-
-void __nac3_nditer_next64(NDIter<int64_t>* iter) {
-    iter->next();
-}
-}

nac3core/irrt/irrt/ndarray/matmul.hpp

@@ -1,98 +0,0 @@
-#pragma once
-
-#include "irrt/debug.hpp"
-#include "irrt/exception.hpp"
-#include "irrt/int_types.hpp"
-#include "irrt/ndarray/basic.hpp"
-#include "irrt/ndarray/broadcast.hpp"
-#include "irrt/ndarray/iter.hpp"
-
-// NOTE: Everything would be much easier and elegant if einsum is implemented.
-
-namespace {
-namespace ndarray::matmul {
-
-/**
- * @brief Perform the broadcast in `np.einsum("...ij,...jk->...ik", a, b)`.
- *
- * Example:
- *   Suppose      `a_shape ==      [1, 97, 4, 2]`
- *       and      `b_shape == [99, 98,  1, 2, 5]`,
- *
- *   ...then  `new_a_shape == [99, 98, 97, 4, 2]`,
- *            `new_b_shape == [99, 98, 97, 2, 5]`,
- *      and     `dst_shape == [99, 98, 97, 4, 5]`.
- *                             ^^^^^^^^^^  ^^^^
- *                          (broadcasted)  (4x2 @ 2x5 => 4x5)
- *
- * @param a_ndims Length of `a_shape`.
- * @param a_shape Shape of `a`.
- * @param b_ndims Length of `b_shape`.
- * @param b_shape Shape of `b`.
- * @param final_ndims Should be equal to `max(a_ndims, b_ndims)`. This is the length of `new_a_shape`,
- * `new_b_shape`, and `dst_shape` - the number of dimensions after broadcasting.
- */
-template<typename SizeT>
-void calculate_shapes(SizeT a_ndims,
-                      SizeT* a_shape,
-                      SizeT b_ndims,
-                      SizeT* b_shape,
-                      SizeT final_ndims,
-                      SizeT* new_a_shape,
-                      SizeT* new_b_shape,
-                      SizeT* dst_shape) {
-    debug_assert(SizeT, a_ndims >= 2);
-    debug_assert(SizeT, b_ndims >= 2);
-    debug_assert_eq(SizeT, max(a_ndims, b_ndims), final_ndims);
-
-    // Check that a and b are compatible for matmul
-    if (a_shape[a_ndims - 1] != b_shape[b_ndims - 2]) {
-        // This is a custom error message. Different from NumPy.
-        raise_exception(SizeT, EXN_VALUE_ERROR, "Cannot multiply LHS (shape ?x{0}) with RHS (shape {1}x?})",
-                        a_shape[a_ndims - 1], b_shape[b_ndims - 2], NO_PARAM);
-    }
-
-    const SizeT num_entries = 2;
-    ShapeEntry<SizeT> entries[num_entries] = {{.ndims = a_ndims - 2, .shape = a_shape},
-                                              {.ndims = b_ndims - 2, .shape = b_shape}};
-
-    // TODO: Optimize this
-    ndarray::broadcast::broadcast_shapes<SizeT>(num_entries, entries, final_ndims - 2, new_a_shape);
-    ndarray::broadcast::broadcast_shapes<SizeT>(num_entries, entries, final_ndims - 2, new_b_shape);
-    ndarray::broadcast::broadcast_shapes<SizeT>(num_entries, entries, final_ndims - 2, dst_shape);
-
-    new_a_shape[final_ndims - 2] = a_shape[a_ndims - 2];
-    new_a_shape[final_ndims - 1] = a_shape[a_ndims - 1];
-    new_b_shape[final_ndims - 2] = b_shape[b_ndims - 2];
-    new_b_shape[final_ndims - 1] = b_shape[b_ndims - 1];
-    dst_shape[final_ndims - 2] = a_shape[a_ndims - 2];
-    dst_shape[final_ndims - 1] = b_shape[b_ndims - 1];
-}
-}  // namespace ndarray::matmul
-}  // namespace
-
-extern "C" {
-using namespace ndarray::matmul;
-
-void __nac3_ndarray_matmul_calculate_shapes(int32_t a_ndims,
-                                            int32_t* a_shape,
-                                            int32_t b_ndims,
-                                            int32_t* b_shape,
-                                            int32_t final_ndims,
-                                            int32_t* new_a_shape,
-                                            int32_t* new_b_shape,
-                                            int32_t* dst_shape) {
-    calculate_shapes(a_ndims, a_shape, b_ndims, b_shape, final_ndims, new_a_shape, new_b_shape, dst_shape);
-}
-
-void __nac3_ndarray_matmul_calculate_shapes64(int64_t a_ndims,
-                                              int64_t* a_shape,
-                                              int64_t b_ndims,
-                                              int64_t* b_shape,
-                                              int64_t final_ndims,
-                                              int64_t* new_a_shape,
-                                              int64_t* new_b_shape,
-                                              int64_t* dst_shape) {
-    calculate_shapes(a_ndims, a_shape, b_ndims, b_shape, final_ndims, new_a_shape, new_b_shape, dst_shape);
-}
-}

nac3core/irrt/irrt/ndarray/reshape.hpp

@@ -1,97 +0,0 @@
-#pragma once
-
-#include "irrt/exception.hpp"
-#include "irrt/int_types.hpp"
-#include "irrt/ndarray/def.hpp"
-
-namespace {
-namespace ndarray::reshape {
-/**
- * @brief Perform assertions on and resolve unknown dimensions in `new_shape` in `np.reshape(<ndarray>, new_shape)`
- *
- * If `new_shape` indeed contains unknown dimensions (specified with `-1`, just like numpy), `new_shape` will be
- * modified to contain the resolved dimension.
- *
- * To perform assertions on and resolve unknown dimensions in `new_shape`, we don't need the actual
- * `<ndarray>` object itself, but only the `.size` of the `<ndarray>`.
- *
- * @param size The `.size` of `<ndarray>`
- * @param new_ndims Number of elements in `new_shape`
- * @param new_shape Target shape to reshape to
- */
-template<typename SizeT>
-void resolve_and_check_new_shape(SizeT size, SizeT new_ndims, SizeT* new_shape) {
-    // Is there a -1 in `new_shape`?
-    bool neg1_exists = false;
-    // Location of -1, only initialized if `neg1_exists` is true
-    SizeT neg1_axis_i;
-    // The computed ndarray size of `new_shape`
-    SizeT new_size = 1;
-
-    for (SizeT axis_i = 0; axis_i < new_ndims; axis_i++) {
-        SizeT dim = new_shape[axis_i];
-        if (dim < 0) {
-            if (dim == -1) {
-                if (neg1_exists) {
-                    // Multiple `-1` found. Throw an error.
-                    raise_exception(SizeT, EXN_VALUE_ERROR, "can only specify one unknown dimension", NO_PARAM,
-                                    NO_PARAM, NO_PARAM);
-                } else {
-                    neg1_exists = true;
-                    neg1_axis_i = axis_i;
-                }
-            } else {
-                // TODO: What? In `np.reshape` any negative dimensions is
-                // treated like its `-1`.
-                //
-                // Try running `np.zeros((3, 4)).reshape((-999, 2))`
-                //
-                // It is not documented by numpy.
-                // Throw an error for now...
-
-                raise_exception(SizeT, EXN_VALUE_ERROR, "Found non -1 negative dimension {0} on axis {1}", dim, axis_i,
-                                NO_PARAM);
-            }
-        } else {
-            new_size *= dim;
-        }
-    }
-
-    bool can_reshape;
-    if (neg1_exists) {
-        // Let `x` be the unknown dimension
-        // Solve `x * <new_size> = <size>`
-        if (new_size == 0 && size == 0) {
-            // `x` has infinitely many solutions
-            can_reshape = false;
-        } else if (new_size == 0 && size != 0) {
-            // `x` has no solutions
-            can_reshape = false;
-        } else if (size % new_size != 0) {
-            // `x` has no integer solutions
-            can_reshape = false;
-        } else {
-            can_reshape = true;
-            new_shape[neg1_axis_i] = size / new_size;  // Resolve dimension
-        }
-    } else {
-        can_reshape = (new_size == size);
-    }
-
-    if (!can_reshape) {
-        raise_exception(SizeT, EXN_VALUE_ERROR, "cannot reshape array of size {0} into given shape", size, NO_PARAM,
-                        NO_PARAM);
-    }
-}
-}  // namespace ndarray::reshape
-}  // namespace
-
-extern "C" {
-void __nac3_ndarray_reshape_resolve_and_check_new_shape(int32_t size, int32_t new_ndims, int32_t* new_shape) {
-    ndarray::reshape::resolve_and_check_new_shape(size, new_ndims, new_shape);
-}
-
-void __nac3_ndarray_reshape_resolve_and_check_new_shape64(int64_t size, int64_t new_ndims, int64_t* new_shape) {
-    ndarray::reshape::resolve_and_check_new_shape(size, new_ndims, new_shape);
-}
-}

nac3core/irrt/irrt/ndarray/transpose.hpp

@@ -1,143 +0,0 @@
-#pragma once
-
-#include "irrt/debug.hpp"
-#include "irrt/exception.hpp"
-#include "irrt/int_types.hpp"
-#include "irrt/ndarray/def.hpp"
-#include "irrt/slice.hpp"
-
-/*
- * Notes on `np.transpose(<array>, <axes>)`
- *
- * TODO: `axes`, if specified, can actually contain negative indices,
- * but it is not documented in numpy.
- *
- * Supporting it for now.
- */
-
-namespace {
-namespace ndarray::transpose {
-/**
- * @brief Do assertions on `<axes>` in `np.transpose(<array>, <axes>)`.
- *
- * Note that `np.transpose`'s `<axe>` argument is optional. If the argument
- * is specified but the user, use this function to do assertions on it.
- *
- * @param ndims The number of dimensions of `<array>`
- * @param num_axes Number of elements in `<axes>` as specified by the user.
- *   This should be equal to `ndims`. If not, a "ValueError: axes don't match array" is thrown.
- * @param axes The user specified `<axes>`.
- */
-template<typename SizeT>
-void assert_transpose_axes(SizeT ndims, SizeT num_axes, const SizeT* axes) {
-    if (ndims != num_axes) {
-        raise_exception(SizeT, EXN_VALUE_ERROR, "axes don't match array", NO_PARAM, NO_PARAM, NO_PARAM);
-    }
-
-    // TODO: Optimize this
-    bool* axe_specified = (bool*)__builtin_alloca(sizeof(bool) * ndims);
-    for (SizeT i = 0; i < ndims; i++)
-        axe_specified[i] = false;
-
-    for (SizeT i = 0; i < ndims; i++) {
-        SizeT axis = slice::resolve_index_in_length(ndims, axes[i]);
-        if (axis == -1) {
-            // TODO: numpy actually throws a `numpy.exceptions.AxisError`
-            raise_exception(SizeT, EXN_VALUE_ERROR, "axis {0} is out of bounds for array of dimension {1}", axis, ndims,
-                            NO_PARAM);
-        }
-
-        if (axe_specified[axis]) {
-            raise_exception(SizeT, EXN_VALUE_ERROR, "repeated axis in transpose", NO_PARAM, NO_PARAM, NO_PARAM);
-        }
-
-        axe_specified[axis] = true;
-    }
-}
-
-/**
- * @brief Create a transpose view of `src_ndarray` and perform proper assertions.
- *
- * This function is very similar to doing `dst_ndarray = np.transpose(src_ndarray, <axes>)`.
- * If `<axes>` is supposed to be `None`, caller can pass in a `nullptr` to `<axes>`.
- *
- * The transpose view created is returned by modifying `dst_ndarray`.
- *
- * The caller is responsible for setting up `dst_ndarray` before calling this function.
- * Here is what this function expects from `dst_ndarray` when called:
- *   - `dst_ndarray->data` does not have to be initialized.
- *   - `dst_ndarray->itemsize` does not have to be initialized.
- *   - `dst_ndarray->ndims` must be initialized, must be equal to `src_ndarray->ndims`.
- *   - `dst_ndarray->shape` must be allocated, through it can contain uninitialized values.
- *   - `dst_ndarray->strides` must be allocated, through it can contain uninitialized values.
- * When this function call ends:
- *   - `dst_ndarray->data` is set to `src_ndarray->data` (`dst_ndarray` is just a view to `src_ndarray`)
- *   - `dst_ndarray->itemsize` is set to `src_ndarray->itemsize`
- *   - `dst_ndarray->ndims` is unchanged
- *   - `dst_ndarray->shape` is updated according to how `np.transpose` works
- *   - `dst_ndarray->strides` is updated according to how `np.transpose` works
- *
- * @param src_ndarray The NDArray to build a transpose view on
- * @param dst_ndarray The resulting NDArray after transpose. Further details in the comments above,
- * @param num_axes Number of elements in axes. Unused if `axes` is nullptr.
- * @param axes Axes permutation. Set it to `nullptr` if `<axes>` is `None`.
- */
-template<typename SizeT>
-void transpose(const NDArray<SizeT>* src_ndarray, NDArray<SizeT>* dst_ndarray, SizeT num_axes, const SizeT* axes) {
-    debug_assert_eq(SizeT, src_ndarray->ndims, dst_ndarray->ndims);
-    const auto ndims = src_ndarray->ndims;
-
-    if (axes != nullptr)
-        assert_transpose_axes(ndims, num_axes, axes);
-
-    dst_ndarray->data = src_ndarray->data;
-    dst_ndarray->itemsize = src_ndarray->itemsize;
-
-    // Check out https://ajcr.net/stride-guide-part-2/ to see how `np.transpose` works behind the scenes.
-    if (axes == nullptr) {
-        // `np.transpose(<array>, axes=None)`
-
-        /*
-         * Minor note: `np.transpose(<array>, axes=None)` is equivalent to
-         * `np.transpose(<array>, axes=[N-1, N-2, ..., 0])` - basically it
-         * is reversing the order of strides and shape.
-         *
-         * This is a fast implementation to handle this special (but very common) case.
-         */
-
-        for (SizeT axis = 0; axis < ndims; axis++) {
-            dst_ndarray->shape[axis] = src_ndarray->shape[ndims - axis - 1];
-            dst_ndarray->strides[axis] = src_ndarray->strides[ndims - axis - 1];
-        }
-    } else {
-        // `np.transpose(<array>, <axes>)`
-
-        // Permute strides and shape according to `axes`, while resolving negative indices in `axes`
-        for (SizeT axis = 0; axis < ndims; axis++) {
-            // `i` cannot be OUT_OF_BOUNDS because of assertions
-            SizeT i = slice::resolve_index_in_length(ndims, axes[axis]);
-
-            dst_ndarray->shape[axis] = src_ndarray->shape[i];
-            dst_ndarray->strides[axis] = src_ndarray->strides[i];
-        }
-    }
-}
-}  // namespace ndarray::transpose
-}  // namespace
-
-extern "C" {
-using namespace ndarray::transpose;
-void __nac3_ndarray_transpose(const NDArray<int32_t>* src_ndarray,
-                              NDArray<int32_t>* dst_ndarray,
-                              int32_t num_axes,
-                              const int32_t* axes) {
-    transpose(src_ndarray, dst_ndarray, num_axes, axes);
-}
-
-void __nac3_ndarray_transpose64(const NDArray<int64_t>* src_ndarray,
-                                NDArray<int64_t>* dst_ndarray,
-                                int64_t num_axes,
-                                const int64_t* axes) {
-    transpose(src_ndarray, dst_ndarray, num_axes, axes);
-}
-}

nac3core/irrt/irrt/range.hpp

@@ -1,47 +0,0 @@
-#pragma once
-
-#include "irrt/debug.hpp"
-#include "irrt/int_types.hpp"
-
-namespace {
-namespace range {
-template<typename T>
-T len(T start, T stop, T step) {
-    // Reference:
-    // https://github.com/python/cpython/blob/9dbd12375561a393eaec4b21ee4ac568a407cdb0/Objects/rangeobject.c#L933
-    if (step > 0 && start < stop)
-        return 1 + (stop - 1 - start) / step;
-    else if (step < 0 && start > stop)
-        return 1 + (start - 1 - stop) / (-step);
-    else
-        return 0;
-}
-}  // namespace range
-
-/**
- * @brief A Python range.
- */
-template<typename T>
-struct Range {
-    T start;
-    T stop;
-    T step;
-
-    /**
-     * @brief Calculate the `len()` of this range.
-     */
-    template<typename SizeT>
-    T len() {
-        debug_assert(SizeT, step != 0);
-        return range::len(start, stop, step);
-    }
-};
-}  // namespace
-
-extern "C" {
-using namespace range;
-
-SliceIndex __nac3_range_slice_len(const SliceIndex start, const SliceIndex end, const SliceIndex step) {
-    return len(start, end, step);
-}
-}

nac3core/irrt/irrt/slice.hpp

@@ -1,156 +0,0 @@
-#pragma once
-
-#include "irrt/debug.hpp"
-#include "irrt/exception.hpp"
-#include "irrt/int_types.hpp"
-#include "irrt/math_util.hpp"
-#include "irrt/range.hpp"
-
-namespace {
-namespace slice {
-/**
- * @brief Resolve a possibly negative index in a list of a known length.
- *
- * Returns -1 if the resolved index is out of the list's bounds.
- */
-template<typename T>
-T resolve_index_in_length(T length, T index) {
-    T resolved = index < 0 ? length + index : index;
-    if (0 <= resolved && resolved < length) {
-        return resolved;
-    } else {
-        return -1;
-    }
-}
-
-/**
- * @brief Resolve a slice as a range.
- *
- * This is equivalent to `range(*slice(start, stop, step).indices(length))` in Python.
- */
-template<typename T>
-void indices(bool start_defined,
-             T start,
-             bool stop_defined,
-             T stop,
-             bool step_defined,
-             T step,
-             T length,
-             T* range_start,
-             T* range_stop,
-             T* range_step) {
-    // Reference: https://github.com/python/cpython/blob/main/Objects/sliceobject.c#L388
-    *range_step = step_defined ? step : 1;
-    bool step_is_negative = *range_step < 0;
-
-    T lower, upper;
-    if (step_is_negative) {
-        lower = -1;
-        upper = length - 1;
-    } else {
-        lower = 0;
-        upper = length;
-    }
-
-    if (start_defined) {
-        *range_start = start < 0 ? max(lower, start + length) : min(upper, start);
-    } else {
-        *range_start = step_is_negative ? upper : lower;
-    }
-
-    if (stop_defined) {
-        *range_stop = stop < 0 ? max(lower, stop + length) : min(upper, stop);
-    } else {
-        *range_stop = step_is_negative ? lower : upper;
-    }
-}
-}  // namespace slice
-
-/**
- * @brief A Python-like slice with **unresolved** indices.
- */
-template<typename T>
-struct Slice {
-    bool start_defined;
-    T start;
-
-    bool stop_defined;
-    T stop;
-
-    bool step_defined;
-    T step;
-
-    Slice() { this->reset(); }
-
-    void reset() {
-        this->start_defined = false;
-        this->stop_defined = false;
-        this->step_defined = false;
-    }
-
-    void set_start(T start) {
-        this->start_defined = true;
-        this->start = start;
-    }
-
-    void set_stop(T stop) {
-        this->stop_defined = true;
-        this->stop = stop;
-    }
-
-    void set_step(T step) {
-        this->step_defined = true;
-        this->step = step;
-    }
-
-    /**
-     * @brief Resolve this slice as a range.
-     *
-     * In Python, this would be `range(*slice(start, stop, step).indices(length))`.
-     */
-    template<typename SizeT>
-    Range<T> indices(T length) {
-        // Reference:
-        // https://github.com/python/cpython/blob/main/Objects/sliceobject.c#L388
-        debug_assert(SizeT, length >= 0);
-
-        Range<T> result;
-        slice::indices(start_defined, start, stop_defined, stop, step_defined, step, length, &result.start,
-                       &result.stop, &result.step);
-        return result;
-    }
-
-    /**
-     * @brief Like `.indices()` but with assertions.
-     */
-    template<typename SizeT>
-    Range<T> indices_checked(T length) {
-        // TODO: Switch to `SizeT length`
-
-        if (length < 0) {
-            raise_exception(SizeT, EXN_VALUE_ERROR, "length should not be negative, got {0}", length, NO_PARAM,
-                            NO_PARAM);
-        }
-
-        if (this->step_defined && this->step == 0) {
-            raise_exception(SizeT, EXN_VALUE_ERROR, "slice step cannot be zero", NO_PARAM, NO_PARAM, NO_PARAM);
-        }
-
-        return this->indices<SizeT>(length);
-    }
-};
-}  // namespace
-
-extern "C" {
-SliceIndex __nac3_slice_index_bound(SliceIndex i, const SliceIndex len) {
-    if (i < 0) {
-        i = len + i;
-    }
-    if (i < 0) {
-        return 0;
-    } else if (i > len) {
-        return len;
-    }
-    return i;
-}
-}

nac3core/irrt/irrt/string.hpp

@@ -1,23 +0,0 @@
-#pragma once
-
-#include "irrt/int_types.hpp"
-
-namespace {
-template<typename SizeT>
-bool __nac3_str_eq_impl(const char* str1, SizeT len1, const char* str2, SizeT len2) {
-    if (len1 != len2) {
-        return 0;
-    }
-    return __builtin_memcmp(str1, str2, static_cast<SizeT>(len1)) == 0;
-}
-}  // namespace
-
-extern "C" {
-bool nac3_str_eq(const char* str1, uint32_t len1, const char* str2, uint32_t len2) {
-    return __nac3_str_eq_impl<uint32_t>(str1, len1, str2, len2);
-}
-
-bool nac3_str_eq64(const char* str1, uint64_t len1, const char* str2, uint64_t len2) {
-    return __nac3_str_eq_impl<uint64_t>(str1, len1, str2, len2);
-}
-}

nac3core/irrt/irrt_basic.hpp

@@ -0,0 +1,216 @@
+#pragma once
+
+#include "irrt_utils.hpp"
+#include "irrt_typedefs.hpp"
+
+/*
+    This header contains IRRT implementations
+    that do not deserved to be categorized (e.g., into numpy, etc.)
+
+    Check out other *.hpp files before including them here!!
+*/
+
+// The type of an index or a value describing the length of a range/slice is
+// always `int32_t`.
+
+namespace {
+    // adapted from GNU Scientific Library: https://git.savannah.gnu.org/cgit/gsl.git/tree/sys/pow_int.c
+    // need to make sure `exp >= 0` before calling this function
+    template <typename T>
+    T __nac3_int_exp_impl(T base, T exp) {
+        T res = 1;
+        /* repeated squaring method */
+        do {
+            if (exp & 1) {
+                res *= base; /* for n odd */
+            }
+            exp >>= 1;
+            base *= base;
+        } while (exp);
+        return res;
+    }
+}
+
+extern "C" {
+    #define DEF_nac3_int_exp_(T) \
+        T __nac3_int_exp_##T(T base, T exp) {\
+            return __nac3_int_exp_impl(base, exp);\
+        }
+
+    DEF_nac3_int_exp_(int32_t)
+    DEF_nac3_int_exp_(int64_t)
+    DEF_nac3_int_exp_(uint32_t)
+    DEF_nac3_int_exp_(uint64_t)
+
+    SliceIndex __nac3_slice_index_bound(SliceIndex i, const SliceIndex len) {
+        if (i < 0) {
+            i = len + i;
+        }
+        if (i < 0) {
+            return 0;
+        } else if (i > len) {
+            return len;
+        }
+        return i;
+    }
+
+    SliceIndex __nac3_range_slice_len(
+        const SliceIndex start,
+        const SliceIndex end,
+        const SliceIndex step
+    ) {
+        SliceIndex diff = end - start;
+        if (diff > 0 && step > 0) {
+            return ((diff - 1) / step) + 1;
+        } else if (diff < 0 && step < 0) {
+            return ((diff + 1) / step) + 1;
+        } else {
+            return 0;
+        }
+    }
+
+    // Handle list assignment and dropping part of the list when
+    // both dest_step and src_step are +1.
+    // - All the index must *not* be out-of-bound or negative,
+    // - The end index is *inclusive*,
+    // - The length of src and dest slice size should already
+    // be checked: if dest.step == 1 then len(src) <= len(dest) else len(src) == len(dest)
+    SliceIndex __nac3_list_slice_assign_var_size(
+        SliceIndex dest_start,
+        SliceIndex dest_end,
+        SliceIndex dest_step,
+        uint8_t *dest_arr,
+        SliceIndex dest_arr_len,
+        SliceIndex src_start,
+        SliceIndex src_end,
+        SliceIndex src_step,
+        uint8_t *src_arr,
+        SliceIndex src_arr_len,
+        const SliceIndex size
+    ) {
+        /* if dest_arr_len == 0, do nothing since we do not support extending list */
+        if (dest_arr_len == 0) return dest_arr_len;
+        /* if both step is 1, memmove directly, handle the dropping of the list, and shrink size */
+        if (src_step == dest_step && dest_step == 1) {
+            const SliceIndex src_len = (src_end >= src_start) ? (src_end - src_start + 1) : 0;
+            const SliceIndex dest_len = (dest_end >= dest_start) ? (dest_end - dest_start + 1) : 0;
+            if (src_len > 0) {
+                __builtin_memmove(
+                    dest_arr + dest_start * size,
+                    src_arr + src_start * size,
+                    src_len * size
+                );
+            }
+            if (dest_len > 0) {
+                /* dropping */
+                __builtin_memmove(
+                    dest_arr + (dest_start + src_len) * size,
+                    dest_arr + (dest_end + 1) * size,
+                    (dest_arr_len - dest_end - 1) * size
+                );
+            }
+            /* shrink size */
+            return dest_arr_len - (dest_len - src_len);
+        }
+        /* if two range overlaps, need alloca */
+        uint8_t need_alloca =
+            (dest_arr == src_arr)
+            && !(
+                max(dest_start, dest_end) < min(src_start, src_end)
+                || max(src_start, src_end) < min(dest_start, dest_end)
+            );
+        if (need_alloca) {
+            uint8_t *tmp = reinterpret_cast<uint8_t *>(__builtin_alloca(src_arr_len * size));
+            __builtin_memcpy(tmp, src_arr, src_arr_len * size);
+            src_arr = tmp;
+        }
+        SliceIndex src_ind = src_start;
+        SliceIndex dest_ind = dest_start;
+        for (;
+            (src_step > 0) ? (src_ind <= src_end) : (src_ind >= src_end);
+            src_ind += src_step, dest_ind += dest_step
+        ) {
+            /* for constant optimization */
+            if (size == 1) {
+                __builtin_memcpy(dest_arr + dest_ind, src_arr + src_ind, 1);
+            } else if (size == 4) {
+                __builtin_memcpy(dest_arr + dest_ind * 4, src_arr + src_ind * 4, 4);
+            } else if (size == 8) {
+                __builtin_memcpy(dest_arr + dest_ind * 8, src_arr + src_ind * 8, 8);
+            } else {
+                /* memcpy for var size, cannot overlap after previous alloca */
+                __builtin_memcpy(dest_arr + dest_ind * size, src_arr + src_ind * size, size);
+            }
+        }
+        /* only dest_step == 1 can we shrink the dest list. */
+        /* size should be ensured prior to calling this function */
+        if (dest_step == 1 && dest_end >= dest_start) {
+            __builtin_memmove(
+                dest_arr + dest_ind * size,
+                dest_arr + (dest_end + 1) * size,
+                (dest_arr_len - dest_end - 1) * size
+            );
+            return dest_arr_len - (dest_end - dest_ind) - 1;
+        }
+        return dest_arr_len;
+    }
+
+    int32_t __nac3_isinf(double x) {
+        return __builtin_isinf(x);
+    }
+
+    int32_t __nac3_isnan(double x) {
+        return __builtin_isnan(x);
+    }
+
+    double tgamma(double arg);
+
+    double __nac3_gamma(double z) {
+        // Handling for denormals
+        //     | x                 | Python gamma(x) | C tgamma(x) |
+        // --- | ----------------- | --------------- | ----------- |
+        // (1) | nan               | nan             | nan         |
+        // (2) | -inf              | -inf            | inf         |
+        // (3) | inf               | inf             | inf         |
+        // (4) | 0.0               | inf             | inf         |
+        // (5) | {-1.0, -2.0, ...} | inf             | nan         |
+
+        // (1)-(3)
+        if (__builtin_isinf(z) || __builtin_isnan(z)) {
+            return z;
+        }
+
+        double v = tgamma(z);
+
+        // (4)-(5)
+        return __builtin_isinf(v) || __builtin_isnan(v) ? __builtin_inf() : v;
+    }
+
+    double lgamma(double arg);
+
+    double __nac3_gammaln(double x) {
+        // libm's handling of value overflows differs from scipy:
+        // - scipy: gammaln(-inf) -> -inf
+        // - libm : lgamma(-inf) -> inf
+
+        if (__builtin_isinf(x)) {
+            return x;
+        }
+
+        return lgamma(x);
+    }
+
+    double j0(double x);
+
+    double __nac3_j0(double x) {
+        // libm's handling of value overflows differs from scipy:
+        // - scipy: j0(inf) -> nan
+        // - libm : j0(inf) -> 0.0
+
+        if (__builtin_isinf(x)) {
+            return __builtin_nan("");
+        }
+
+        return j0(x);
+    }
+}

nac3core/irrt/irrt_everything.hpp

@@ -0,0 +1,14 @@
+#pragma once
+
+#include "irrt_utils.hpp"
+#include "irrt_typedefs.hpp"
+#include "irrt_basic.hpp"
+#include "irrt_slice.hpp"
+#include "irrt_numpy_ndarray.hpp"
+
+/*
+    All IRRT implementations.
+
+    We don't have any pre-compiled objects, so we are writing all implementations in headers and
+    concatenate them with `#include` into one massive source file that contains all the IRRT stuff.
+*/

nac3core/irrt/irrt_numpy_ndarray.hpp

@@ -0,0 +1,466 @@
+#pragma once
+
+#include "irrt_utils.hpp"
+#include "irrt_typedefs.hpp"
+#include "irrt_slice.hpp"
+
+/*
+    NDArray-related implementations.
+`*/
+
+// NDArray indices are always `uint32_t`.
+using NDIndex = uint32_t;
+
+namespace {
+    namespace ndarray_util {
+        template <typename SizeT>
+        static void set_indices_by_nth(SizeT ndims, const SizeT* shape, SizeT* indices, SizeT nth) {
+            for (int32_t i = 0; i < ndims; i++) {
+                int32_t dim_i = ndims - i - 1;
+                int32_t dim = shape[dim_i];
+
+                indices[dim_i] = nth % dim;
+                nth /= dim;
+            }
+        }
+
+        // Compute the strides of an ndarray given an ndarray `shape`
+        // and assuming that the ndarray is *fully C-contagious*.
+        //
+        // You might want to read up on https://ajcr.net/stride-guide-part-1/.
+        template <typename SizeT>
+        static void set_strides_by_shape(SizeT itemsize, SizeT ndims, SizeT* dst_strides, const SizeT* shape) {
+            SizeT stride_product = 1;
+            for (SizeT i = 0; i < ndims; i++) {
+                int dim_i = ndims - i - 1;
+                dst_strides[dim_i] = stride_product * itemsize;
+                stride_product *= shape[dim_i];
+            }
+        }
+
+        // Compute the size/# of elements of an ndarray given its shape
+        template <typename SizeT>
+        static SizeT calc_size_from_shape(SizeT ndims, const SizeT* shape) {
+            SizeT size = 1;
+            for (SizeT dim_i = 0; dim_i < ndims; dim_i++) size *= shape[dim_i];
+            return size;
+        }
+
+        template <typename SizeT>
+        static bool can_broadcast_shape_to(
+            const SizeT target_ndims,
+            const SizeT *target_shape,
+            const SizeT src_ndims,
+            const SizeT *src_shape
+        ) {
+            /*
+                // See https://numpy.org/doc/stable/user/basics.broadcasting.html
+
+                This function handles this example:
+                ```
+                Image  (3d array): 256 x 256 x 3
+                Scale  (1d array):             3
+                Result (3d array): 256 x 256 x 3
+                ```
+
+                Other interesting examples to consider:
+                - `can_broadcast_shape_to([3], [1, 1, 1, 1, 3]) == true`
+                - `can_broadcast_shape_to([3], [3, 1]) == false`
+                - `can_broadcast_shape_to([256, 256, 3], [256, 1, 3]) == true`
+
+                In cases when the shapes contain zero(es):
+                - `can_broadcast_shape_to([0], [1]) == true`
+                - `can_broadcast_shape_to([0], [2]) == false`
+                - `can_broadcast_shape_to([0, 4, 0, 0], [1]) == true`
+                - `can_broadcast_shape_to([0, 4, 0, 0], [1, 1, 1, 1]) == true`
+                - `can_broadcast_shape_to([0, 4, 0, 0], [1, 4, 1, 1]) == true`
+                - `can_broadcast_shape_to([4, 3], [0, 3]) == false`
+                - `can_broadcast_shape_to([4, 3], [0, 0]) == false`
+            */
+
+            // This is essentially doing the following in Python:
+            // `for target_dim, src_dim in itertools.zip_longest(target_shape[::-1], src_shape[::-1], fillvalue=1)`
+            for (SizeT i = 0; i < max(target_ndims, src_ndims); i++) {
+                SizeT target_dim_i = target_ndims - i - 1;
+                SizeT src_dim_i = src_ndims - i - 1;
+
+                bool target_dim_exists = target_dim_i >= 0;
+                bool src_dim_exists = src_dim_i >= 0;
+
+                SizeT target_dim = target_dim_exists ? target_shape[target_dim_i] : 1;
+                SizeT src_dim = src_dim_exists ? src_shape[src_dim_i] : 1;
+
+                bool ok = src_dim == 1 || target_dim == src_dim;
+                if (!ok) return false;
+            }
+
+            return true;
+        }
+    }
+
+    typedef uint8_t NDSliceType;
+    extern "C" {
+        const NDSliceType INPUT_SLICE_TYPE_INDEX = 0;
+        const NDSliceType INPUT_SLICE_TYPE_SLICE = 1;
+    }
+
+    struct NDSlice {
+        // A poor-man's `std::variant<int, UserRange>`
+        NDSliceType type;
+
+        /*
+            if type == INPUT_SLICE_TYPE_INDEX => `slice` points to a single `SizeT`
+            if type == INPUT_SLICE_TYPE_SLICE => `slice` points to a single `UserRange`
+        */
+        uint8_t *slice;
+    };
+
+    namespace ndarray_util {
+        template<typename SizeT>
+        SizeT deduce_ndims_after_slicing(SizeT ndims, SizeT num_slices, const NDSlice *slices) {
+            irrt_assert(num_slices <= ndims);
+
+            SizeT final_ndims = ndims;
+            for (SizeT i = 0; i < num_slices; i++) {
+                if (slices[i].type == INPUT_SLICE_TYPE_INDEX) {
+                    final_ndims--; // An integer slice demotes the rank by 1
+                }
+            }
+            return final_ndims;
+        }
+    }
+
+    template <typename SizeT>
+    struct NDArrayIndicesIter {
+        SizeT ndims;
+        const SizeT *shape;
+        SizeT *indices;
+
+        void set_indices_zero() {
+            __builtin_memset(indices, 0, sizeof(SizeT) * ndims);
+        }
+
+        void next() {
+            for (SizeT i = 0; i < ndims; i++) {
+                SizeT dim_i = ndims - i - 1;
+
+                indices[dim_i]++;
+                if (indices[dim_i] < shape[dim_i]) {
+                    break;
+                } else {
+                    indices[dim_i] = 0;
+                }
+            }
+        }
+    };
+
+    // The NDArray object. `SizeT` is the *signed* size type of this ndarray.
+    //
+    // NOTE: The order of fields is IMPORTANT. DON'T TOUCH IT
+    //
+    // Some resources you might find helpful:
+    // - The official numpy implementations:
+    //   - https://github.com/numpy/numpy/blob/735a477f0bc2b5b84d0e72d92f224bde78d4e069/doc/source/reference/c-api/types-and-structures.rst
+    // - On strides (about reshaping, slicing, C-contagiousness, etc)
+    //   - https://ajcr.net/stride-guide-part-1/.
+    //   - https://ajcr.net/stride-guide-part-2/.
+    //   - https://ajcr.net/stride-guide-part-3/.
+    template <typename SizeT>
+    struct NDArray {
+        // The underlying data this `ndarray` is pointing to.
+        //
+        // NOTE: Formally this should be of type `void *`, but clang
+        // translates `void *` to `i8 *` when run with `-S -emit-llvm`,
+        // so we will put `uint8_t *` here for clarity.
+        uint8_t *data;
+
+        // The number of bytes of a single element in `data`.
+        //
+        // The `SizeT` is treated as `unsigned`.
+        SizeT itemsize;
+
+        // The number of dimensions of this shape.
+        //
+        // The `SizeT` is treated as `unsigned`.
+        SizeT ndims;
+
+        // Array shape, with length equal to `ndims`.
+        //
+        // The `SizeT` is treated as `unsigned`.
+        //
+        // NOTE: `shape` can contain 0.
+        //   (those appear when the user makes an out of bounds slice into an ndarray, e.g., `np.zeros((3, 3))[400:].shape == (0, 3)`)
+        SizeT *shape;
+
+        // Array strides (stride value is in number of bytes, NOT number of elements), with length equal to `ndims`.
+        //
+        // The `SizeT` is treated as `signed`.
+        //
+        // NOTE: `strides` can have negative numbers.
+        //   (those appear when there is a slice with a negative step, e.g., `my_array[::-1]`)
+        SizeT *strides;
+
+        // Calculate the size/# of elements of an `ndarray`.
+        // This function corresponds to `np.size(<ndarray>)` or `ndarray.size`
+        SizeT size() {
+            return ndarray_util::calc_size_from_shape(ndims, shape);
+        }
+
+        // Calculate the number of bytes of its content of an `ndarray` *in its view*.
+        // This function corresponds to `ndarray.nbytes`
+        SizeT nbytes() {
+            return this->size() * itemsize;
+        }
+
+        void set_value_at_pelement(uint8_t* pelement, const uint8_t* pvalue) {
+            __builtin_memcpy(pelement, pvalue, itemsize);
+        }
+
+        uint8_t* get_pelement(const SizeT *indices) {
+            uint8_t* element = data;
+            for (SizeT dim_i = 0; dim_i < ndims; dim_i++)
+                element += indices[dim_i] * strides[dim_i];
+            return element;
+        }
+
+        uint8_t* get_nth_pelement(SizeT nth) {
+            irrt_assert(0 <= nth);
+            irrt_assert(nth < this->size());
+
+            SizeT* indices = (SizeT*) __builtin_alloca(sizeof(SizeT) * this->ndims);
+            ndarray_util::set_indices_by_nth(this->ndims, this->shape, indices, nth);
+            return get_pelement(indices);
+        }
+
+        // Get pointer to the first element of this ndarray, assuming
+        // `this->size() > 0`, i.e., not "degenerate" due to zeroes in `this->shape`)
+        //
+        // This is particularly useful for when the ndarray is just containing a single scalar.
+        uint8_t* get_first_pelement() {
+            irrt_assert(this->size() > 0);
+            return this->data; // ...It is simply `this->data`
+        }
+
+        // Is the given `indices` valid/in-bounds?
+        bool in_bounds(const SizeT *indices) {
+            for (SizeT dim_i = 0; dim_i < ndims; dim_i++) {
+                bool dim_ok = indices[dim_i] < shape[dim_i];
+                if (!dim_ok) return false;
+            }
+            return true;
+        }
+
+        // Fill the ndarray with a value
+        void fill_generic(const uint8_t* pvalue) {
+            NDArrayIndicesIter<SizeT> iter;
+            iter.ndims = this->ndims;
+            iter.shape = this->shape;
+            iter.indices = (SizeT*) __builtin_alloca(sizeof(SizeT) * ndims);
+            iter.set_indices_zero();
+
+            for (SizeT i = 0; i < this->size(); i++, iter.next()) {
+                uint8_t* pelement = get_pelement(iter.indices);
+                set_value_at_pelement(pelement, pvalue);
+            }
+        }
+
+        // Set the strides of the ndarray with `ndarray_util::set_strides_by_shape`
+        void set_strides_by_shape() {
+            ndarray_util::set_strides_by_shape(itemsize, ndims, strides, shape);
+        }
+
+        // https://numpy.org/doc/stable/reference/generated/numpy.eye.html
+        void set_to_eye(SizeT k, const uint8_t* zero_pvalue, const uint8_t* one_pvalue) {
+            __builtin_assume(ndims == 2);
+
+            // TODO: Better implementation
+
+            fill_generic(zero_pvalue);
+            for (SizeT i = 0; i < min(shape[0], shape[1]); i++) {
+                SizeT row = i;
+                SizeT col = i + k;
+                SizeT indices[2] = { row, col };
+
+                if (!in_bounds(indices)) continue;
+
+                uint8_t* pelement = get_pelement(indices);
+                set_value_at_pelement(pelement, one_pvalue);
+            }
+        }
+
+        // To support numpy complex slices (e.g., `my_array[:50:2,4,:2:-1]`)
+        //
+        // Things assumed by this function:
+        // - `dst_ndarray` is allocated by the caller
+        // - `dst_ndarray.ndims` has the correct value (according to `ndarray_util::deduce_ndims_after_slicing`).
+        //   - ... and `dst_ndarray.shape` and `dst_ndarray.strides` have been allocated by the caller as well
+        //
+        // Other notes:
+        // - `dst_ndarray->data` does not have to be set, it will be derived.
+        // - `dst_ndarray->itemsize` does not have to be set, it will be set to `this->itemsize`
+        // - `dst_ndarray->shape` and `dst_ndarray.strides` can contain empty values
+        void slice(SizeT num_ndslices, NDSlice* ndslices, NDArray<SizeT>* dst_ndarray) {
+            // REFERENCE CODE (check out `_index_helper` in `__getitem__`):
+            //   https://github.com/wadetb/tinynumpy/blob/0d23d22e07062ffab2afa287374c7b366eebdda1/tinynumpy/tinynumpy.py#L652
+
+            irrt_assert(dst_ndarray->ndims == ndarray_util::deduce_ndims_after_slicing(this->ndims, num_ndslices, ndslices));
+
+            dst_ndarray->data = this->data;
+
+            SizeT this_axis = 0;
+            SizeT dst_axis = 0;
+
+            for (SizeT i = 0; i < num_ndslices; i++) {
+                NDSlice *ndslice = &ndslices[i];
+                if (ndslice->type == INPUT_SLICE_TYPE_INDEX) {
+                    // Handle when the ndslice is just a single (possibly negative) integer
+                    // e.g., `my_array[::2, -5, ::-1]`
+                    //                      ^^------ like this
+                    SizeT index_user = *((SizeT*) ndslice->slice);
+                    SizeT index = resolve_index_in_length(this->shape[this_axis], index_user);
+                    dst_ndarray->data += index * this->strides[this_axis]; // Add offset
+
+                    // Next
+                    this_axis++;
+                } else if (ndslice->type == INPUT_SLICE_TYPE_SLICE) {
+                    // Handle when the ndslice is a slice (represented by UserSlice in IRRT)
+                    // e.g., `my_array[::2, -5, ::-1]`
+                    //                 ^^^------^^^^----- like these
+                    UserSlice<SizeT>* user_slice = (UserSlice<SizeT>*) ndslice->slice;
+                    Slice<SizeT> slice = user_slice->indices(this->shape[this_axis]); // To resolve negative indices and other funny stuff written by the user
+
+                    // NOTE: There is no need to write special code to handle negative steps/strides.
+                    // This simple implementation meticulously handles both positive and negative steps/strides.
+                    // Check out the tinynumpy and IRRT's test cases if you are not convinced.
+                    dst_ndarray->data += slice.start * this->strides[this_axis]; // Add offset (NOTE: no need to `* itemsize`, strides count in # of bytes)
+                    dst_ndarray->strides[dst_axis] = slice.step * this->strides[this_axis]; // Determine stride
+                    dst_ndarray->shape[dst_axis] = slice.len(); // Determine shape dimension
+
+                    // Next
+                    dst_axis++;
+                    this_axis++;
+                } else {
+                    __builtin_unreachable();
+                }
+            }
+
+            irrt_assert(dst_axis == dst_ndarray->ndims); // Sanity check on the implementation
+        }
+
+        // Similar to `np.broadcast_to(<ndarray>, <target_shape>)`
+        // Assumptions:
+        //   - `this` has to be fully initialized.
+        //   - `dst_ndarray->ndims` has to be set.
+        //   - `dst_ndarray->shape` has to be set, this determines the shape `this` broadcasts to.
+        //
+        // Other notes:
+        //   - `dst_ndarray->data` does not have to be set, it will be set to `this->data`.
+        //   - `dst_ndarray->itemsize` does not have to be set, it will be set to `this->data`.
+        //   - `dst_ndarray->strides` does not have to be set, it will be overwritten.
+        //
+        // Cautions:
+        //   ```
+        //   xs = np.zeros((4,))
+        //   ys = np.zero((4, 1))
+        //   ys[:] = xs # ok
+        // 
+        //   xs = np.zeros((1, 4))
+        //   ys = np.zero((4,))
+        //   ys[:] = xs # allowed
+        //   # However `np.broadcast_to(xs, (4,))` would fails, as per numpy's broadcasting rule.
+        //   # and apparently numpy will "deprecate" this? SEE https://github.com/numpy/numpy/issues/21744
+        //   # This implementation will NOT support this assignment.
+        //   ```
+        void broadcast_to(NDArray<SizeT>* dst_ndarray) {
+            dst_ndarray->data = this->data;
+            dst_ndarray->itemsize = this->itemsize;
+
+            irrt_assert(
+                ndarray_util::can_broadcast_shape_to(
+                    dst_ndarray->ndims,
+                    dst_ndarray->shape,
+                    this->ndims,
+                    this->shape
+                )
+            );
+
+            SizeT stride_product = 1;
+            for (SizeT i = 0; i < max(this->ndims, dst_ndarray->ndims); i++) {
+                SizeT this_dim_i = this->ndims - i - 1;
+                SizeT dst_dim_i = dst_ndarray->ndims - i - 1;
+
+                bool this_dim_exists = this_dim_i >= 0;
+                bool dst_dim_exists = dst_dim_i >= 0;
+
+                // TODO: Explain how this works
+                bool c1 = this_dim_exists && this->shape[this_dim_i] == 1;
+                bool c2 = dst_dim_exists && dst_ndarray->shape[dst_dim_i] != 1;
+                if (!this_dim_exists || (c1 && c2)) {
+                    dst_ndarray->strides[dst_dim_i] = 0; // Freeze it in-place
+                } else {
+                    dst_ndarray->strides[dst_dim_i] = stride_product * this->itemsize;
+                    stride_product *= this->shape[this_dim_i]; // NOTE: this_dim_exist must be true here.
+                }
+            }
+        }
+
+        // Simulates `this_ndarray[:] = src_ndarray`, with automatic broadcasting.
+        // Caution on https://github.com/numpy/numpy/issues/21744
+        // Also see `NDArray::broadcast_to`
+        void assign_with(NDArray<SizeT>* src_ndarray) {
+            irrt_assert(
+                ndarray_util::can_broadcast_shape_to(
+                    this->ndims,
+                    this->shape,
+                    src_ndarray->ndims,
+                    src_ndarray->shape
+                )
+            );
+
+            // Broadcast the `src_ndarray` to make the reading process *much* easier
+            SizeT* broadcasted_src_ndarray_strides = __builtin_alloca(sizeof(SizeT) * this->ndims); // Remember to allocate strides beforehand
+            NDArray<SizeT> broadcasted_src_ndarray = {
+                .ndims = this->ndims,
+                .shape = this->shape,
+                .strides = broadcasted_src_ndarray_strides
+            };
+            src_ndarray->broadcast_to(&broadcasted_src_ndarray);
+
+            // Using iter instead of `get_nth_pelement` because it is slightly faster
+            SizeT* indices = __builtin_alloca(sizeof(SizeT) * this->ndims);
+            auto iter = NDArrayIndicesIter<SizeT> {
+                .ndims = this->ndims,
+                .shape = this->shape,
+                .indices = indices
+            };
+            const SizeT this_size = this->size();
+            for (SizeT i = 0; i < this_size; i++, iter.next()) {
+                uint8_t* src_pelement = broadcasted_src_ndarray_strides->get_pelement(indices);
+                uint8_t* this_pelement = this->get_pelement(indices);
+                this->set_value_at_pelement(src_pelement, src_pelement);
+            }
+        }
+    };
+}
+
+extern "C" {
+    uint32_t __nac3_ndarray_size(NDArray<int32_t>* ndarray) {
+        return ndarray->size();
+    }
+
+    uint64_t __nac3_ndarray_size64(NDArray<int64_t>* ndarray) {
+        return ndarray->size();
+    }
+
+    void __nac3_ndarray_fill_generic(NDArray<int32_t>* ndarray, uint8_t* pvalue) {
+        ndarray->fill_generic(pvalue);
+    }
+
+    void __nac3_ndarray_fill_generic64(NDArray<int64_t>* ndarray, uint8_t* pvalue) {
+        ndarray->fill_generic(pvalue);
+    }
+
+    // void __nac3_ndarray_slice(NDArray<int32_t>* ndarray, int32_t num_slices, NDSlice<int32_t> *slices, NDArray<int32_t> *dst_ndarray) {
+    //     // ndarray->slice(num_slices, slices, dst_ndarray);
+    // }
+}

nac3core/irrt/irrt_slice.hpp

@@ -0,0 +1,80 @@
+#pragma once
+
+#include "irrt_utils.hpp"
+#include "irrt_typedefs.hpp"
+
+namespace {
+    // A proper slice in IRRT, all negative indices have be resolved to absolute values.
+    // Even though nac3core's slices are always `int32_t`, we will template slice anyway
+    // since this struct is used as a general utility.
+    template <typename T>
+    struct Slice {
+        T start;
+        T stop;
+        T step;
+
+        // The length/The number of elements of the slice if it were a range,
+        // i.e., the value of `len(range(this->start, this->stop, this->end))`
+        T len() {
+            T diff = stop - start;
+            if (diff > 0 && step > 0) {
+                return ((diff - 1) / step) + 1;
+            } else if (diff < 0 && step < 0) {
+                return ((diff + 1) / step) + 1;
+            } else {
+                return 0;
+            }
+        }
+    };
+
+    template<typename T>
+    T resolve_index_in_length(T length, T index) {
+        irrt_assert(length >= 0);
+        if (index < 0) {
+            // Remember that index is negative, so do a plus here
+            return max(length + index, 0);
+        } else {
+            return min(length, index);
+        }
+    }
+
+    // NOTE: using a bitfield for the `*_defined` is better, at the
+    // cost of a more annoying implementation in nac3core inkwell
+    template <typename T>
+    struct UserSlice {
+        uint8_t start_defined;
+        T start;
+
+        uint8_t stop_defined;
+        T stop;
+
+        uint8_t step_defined;
+        T step;
+
+        // Like Python's `slice(start, stop, step).indices(length)`
+        Slice<T> indices(T length) {
+            // NOTE: This function implements Python's `slice.indices` *FAITHFULLY*.
+            // SEE: https://github.com/python/cpython/blob/f62161837e68c1c77961435f1b954412dd5c2b65/Objects/sliceobject.c#L546
+            irrt_assert(length >= 0);
+            irrt_assert(!step_defined || step != 0); // step_defined -> step != 0; step cannot be zero if specified by user
+
+            Slice<T> result;
+            result.step = step_defined ? step : 1;
+            bool step_is_negative = result.step < 0;
+
+            if (start_defined) {
+                result.start = resolve_index_in_length(length, start);
+            } else {
+                result.start = step_is_negative ? length - 1 : 0;
+            }
+            
+            if (stop_defined) {
+                result.stop = resolve_index_in_length(length, stop);
+            } else {
+                result.stop = step_is_negative ? -1 : length;
+            }
+
+            return result;
+        }
+    };
+}

nac3core/irrt/irrt_test.cpp

@@ -0,0 +1,658 @@
+// This file will be compiled like a real C++ program,
+// and we do have the luxury to use the standard libraries.
+// That is if the nix flakes do not have issues... especially on msys2...
+#include <cstdint>
+#include <cstdio>
+#include <cstdlib>
+
+// Set `IRRT_DONT_TYPEDEF_INTS` because `cstdint` defines them
+#define IRRT_DONT_TYPEDEF_INTS
+#include "irrt_everything.hpp"
+
+void test_fail() {
+    printf("[!] Test failed\n");
+    exit(1);
+}
+
+void __begin_test(const char* function_name, const char* file, int line) {
+    printf("######### Running %s @ %s:%d\n", function_name, file, line);
+}
+
+#define BEGIN_TEST() __begin_test(__FUNCTION__, __FILE__, __LINE__)
+
+template <typename T>
+void debug_print_array(const char* format, int len, T* as) {
+    printf("[");
+    for (int i = 0; i < len; i++) {
+        if (i != 0) printf(", ");
+        printf(format, as[i]);
+    }
+    printf("]");
+}
+
+template <typename T>
+void assert_arrays_match(const char* label, const char* format, int len, T* expected, T* got) {
+    if (!arrays_match(len, expected, got)) {
+        printf(">>>>>>> %s\n", label);
+        printf("    Expecting = ");
+        debug_print_array(format, len, expected);
+        printf("\n");
+        printf("          Got = ");
+        debug_print_array(format, len, got);
+        printf("\n");
+        test_fail();
+    }
+}
+
+template <typename T>
+void assert_values_match(const char* label, const char* format, T expected, T got) {
+    if (expected != got) {
+        printf(">>>>>>> %s\n", label);
+        printf("    Expecting = ");
+        printf(format, expected);
+        printf("\n");
+        printf("          Got = ");
+        printf(format, got);
+        printf("\n");
+        test_fail();
+    }
+}
+
+void print_repeated(const char *str, int count) {
+    for (int i = 0; i < count; i++) {
+        printf("%s", str);
+    }
+}
+
+template<typename SizeT, typename ElementT>
+void __print_ndarray_aux(const char *format, bool first, bool last, SizeT* cursor, SizeT depth, NDArray<SizeT>* ndarray) {
+    // A really lazy recursive implementation
+    
+    // Add left padding unless its the first entry (since there would be "[[[" before it)
+    if (!first) {
+        print_repeated(" ", depth);
+    }
+
+    const SizeT dim = ndarray->shape[depth];
+    if (depth + 1 == ndarray->ndims) {
+        // Recursed down to last dimension, print the values in a nice list
+        printf("[");
+
+        SizeT* indices = (SizeT*) __builtin_alloca(sizeof(SizeT) * ndarray->ndims);
+        for (SizeT i = 0; i < dim; i++) {
+            ndarray_util::set_indices_by_nth(ndarray->ndims, ndarray->shape, indices, *cursor);
+            ElementT* pelement = (ElementT*) ndarray->get_pelement(indices);
+            ElementT element = *pelement;
+
+            if (i != 0) printf(", "); // List delimiter
+            printf(format, element);
+            printf("(@");
+            debug_print_array("%d", ndarray->ndims, indices);
+            printf(")");
+
+            (*cursor)++;
+        }
+        printf("]");
+    } else {
+        printf("[");
+        for (SizeT i = 0; i < ndarray->shape[depth]; i++) {
+            __print_ndarray_aux<SizeT, ElementT>(
+                format,
+                i == 0, // first?
+                i + 1 == dim, // last?
+                cursor,
+                depth + 1,
+                ndarray
+            );
+        }
+        printf("]");
+    }
+
+    // Add newline unless its the last entry (since there will be "]]]" after it)
+    if (!last) {
+        print_repeated("\n", depth);
+    }
+}
+
+template<typename SizeT, typename ElementT>
+void print_ndarray(const char *format, NDArray<SizeT>* ndarray) {
+    if (ndarray->ndims == 0) {
+        printf("<empty ndarray>");
+    } else {
+        SizeT cursor = 0;
+        __print_ndarray_aux<SizeT, ElementT>(format, true, true, &cursor, 0, ndarray);
+    }
+    printf("\n");
+}
+
+void test_calc_size_from_shape_normal() {
+    // Test shapes with normal values
+    BEGIN_TEST();
+
+    int32_t shape[4] = { 2, 3, 5, 7 };
+    assert_values_match("size", "%d", 210, ndarray_util::calc_size_from_shape<int32_t>(4, shape));
+}
+
+void test_calc_size_from_shape_has_zero() {
+    // Test shapes with 0 in them
+    BEGIN_TEST();
+
+    int32_t shape[4] = { 2, 0, 5, 7 };
+    assert_values_match("size", "%d", 0, ndarray_util::calc_size_from_shape<int32_t>(4, shape));
+}
+
+void test_set_strides_by_shape() {
+    // Test `set_strides_by_shape()`
+    BEGIN_TEST();
+
+    int32_t shape[4] = { 99, 3, 5, 7 };
+    int32_t strides[4] = { 0 };
+    ndarray_util::set_strides_by_shape((int32_t) sizeof(int32_t), 4, strides, shape);
+
+    int32_t expected_strides[4] = {
+        105 * sizeof(int32_t),
+        35 * sizeof(int32_t),
+        7 * sizeof(int32_t),
+        1 * sizeof(int32_t)
+    };
+    assert_arrays_match("strides", "%u", 4u, expected_strides, strides);
+}
+
+void test_ndarray_indices_iter_normal() {
+    // Test NDArrayIndicesIter normal behavior
+    BEGIN_TEST();
+
+    int32_t shape[3] = { 1, 2, 3 };
+    int32_t indices[3] = { 0, 0, 0 };
+    auto iter = NDArrayIndicesIter<int32_t> {
+        .ndims = 3,
+        .shape = shape,
+        .indices = indices
+    };
+
+    assert_arrays_match("indices #0", "%u", 3u, iter.indices, (int32_t[3]) { 0, 0, 0 });
+    iter.next();
+    assert_arrays_match("indices #1", "%u", 3u, iter.indices, (int32_t[3]) { 0, 0, 1 });
+    iter.next();
+    assert_arrays_match("indices #2", "%u", 3u, iter.indices, (int32_t[3]) { 0, 0, 2 });
+    iter.next();
+    assert_arrays_match("indices #3", "%u", 3u, iter.indices, (int32_t[3]) { 0, 1, 0 });
+    iter.next();
+    assert_arrays_match("indices #4", "%u", 3u, iter.indices, (int32_t[3]) { 0, 1, 1 });
+    iter.next();
+    assert_arrays_match("indices #5", "%u", 3u, iter.indices, (int32_t[3]) { 0, 1, 2 });
+    iter.next();
+    assert_arrays_match("indices #6", "%u", 3u, iter.indices, (int32_t[3]) { 0, 0, 0 }); // Loops back
+    iter.next();
+    assert_arrays_match("indices #7", "%u", 3u, iter.indices, (int32_t[3]) { 0, 0, 1 });
+}
+
+void test_ndarray_fill_generic() {
+    // Test ndarray fill_generic
+    BEGIN_TEST();
+
+    // Choose a type that's neither int32_t nor uint64_t (candidates of SizeT) to spice it up
+    // Also make all the octets non-zero, to see if `memcpy` in `fill_generic` is working perfectly.
+    uint16_t fill_value = 0xFACE;
+
+    uint16_t in_data[6] = { 100, 101, 102, 103, 104, 105 }; // Fill `data` with values that != `999`
+    int32_t in_itemsize = sizeof(uint16_t);
+    const int32_t in_ndims = 2;
+    int32_t in_shape[in_ndims] = { 2, 3 };
+    int32_t in_strides[in_ndims] = {};
+    NDArray<int32_t> ndarray = {
+        .data = (uint8_t*) in_data,
+        .itemsize = in_itemsize,
+        .ndims = in_ndims,
+        .shape = in_shape,
+        .strides = in_strides,
+    };
+    ndarray.set_strides_by_shape();
+    ndarray.fill_generic((uint8_t*) &fill_value); // `fill_generic` here
+
+    uint16_t expected_data[6] = { fill_value, fill_value, fill_value, fill_value, fill_value, fill_value };
+    assert_arrays_match("data", "0x%hX", 6, expected_data, in_data);
+}
+
+void test_ndarray_set_to_eye() {
+    // Test `set_to_eye` behavior (helper function to implement `np.eye()`)
+    BEGIN_TEST();
+
+    double in_data[9] = { 99.0, 99.0, 99.0, 99.0, 99.0, 99.0, 99.0, 99.0, 99.0 };
+    int32_t in_itemsize = sizeof(double);
+    const int32_t in_ndims = 2;
+    int32_t in_shape[in_ndims] = { 3, 3 };
+    int32_t in_strides[in_ndims] = {};
+    NDArray<int32_t> ndarray = {
+        .data = (uint8_t*) in_data,
+        .itemsize = in_itemsize,
+        .ndims = in_ndims,
+        .shape = in_shape,
+        .strides = in_strides,
+    };
+    ndarray.set_strides_by_shape();
+
+    double zero = 0.0;
+    double one = 1.0;
+    ndarray.set_to_eye(1, (uint8_t*) &zero, (uint8_t*) &one);
+
+    assert_values_match("in_data[0]", "%f", 0.0, in_data[0]);
+    assert_values_match("in_data[1]", "%f", 1.0, in_data[1]);
+    assert_values_match("in_data[2]", "%f", 0.0, in_data[2]);
+    assert_values_match("in_data[3]", "%f", 0.0, in_data[3]);
+    assert_values_match("in_data[4]", "%f", 0.0, in_data[4]);
+    assert_values_match("in_data[5]", "%f", 1.0, in_data[5]);
+    assert_values_match("in_data[6]", "%f", 0.0, in_data[6]);
+    assert_values_match("in_data[7]", "%f", 0.0, in_data[7]);
+    assert_values_match("in_data[8]", "%f", 0.0, in_data[8]);
+}
+
+void test_slice_1() {
+    // Test `slice(5, None, None).indices(100) == slice(5, 100, 1)`
+    BEGIN_TEST();
+
+    UserSlice<int> user_slice = {
+        .start_defined = 1,
+        .start = 5,
+        .stop_defined = 0,
+        .step_defined = 0,
+    };
+
+    auto slice = user_slice.indices(100);
+    assert_values_match("start", "%d", 5, slice.start);
+    assert_values_match("stop", "%d", 100, slice.stop);
+    assert_values_match("step", "%d", 1, slice.step);
+}
+
+void test_slice_2() {
+    // Test `slice(400, 999, None).indices(100) == slice(100, 100, 1)`
+    BEGIN_TEST();
+
+    UserSlice<int> user_slice = {
+        .start_defined = 1,
+        .start = 400,
+        .stop_defined = 0,
+        .step_defined = 0,
+    };
+
+    auto slice = user_slice.indices(100);
+    assert_values_match("start", "%d", 100, slice.start);
+    assert_values_match("stop", "%d", 100, slice.stop);
+    assert_values_match("step", "%d", 1, slice.step);
+}
+
+void test_slice_3() {
+    // Test `slice(-10, -5, None).indices(100) == slice(90, 95, 1)`
+    BEGIN_TEST();
+
+    UserSlice<int> user_slice = {
+        .start_defined = 1,
+        .start = -10,
+        .stop_defined = 1,
+        .stop = -5,
+        .step_defined = 0,
+    };
+
+    auto slice = user_slice.indices(100);
+    assert_values_match("start", "%d", 90, slice.start);
+    assert_values_match("stop", "%d", 95, slice.stop);
+    assert_values_match("step", "%d", 1, slice.step);
+}
+
+void test_slice_4() {
+    // Test `slice(None, None, -5).indices(100) == (99, -1, -5)`
+    BEGIN_TEST();
+
+    UserSlice<int> user_slice = {
+        .start_defined = 0,
+        .stop_defined = 0,
+        .step_defined = 1,
+        .step = -5
+    };
+
+    auto slice = user_slice.indices(100);
+    assert_values_match("start", "%d", 99, slice.start);
+    assert_values_match("stop", "%d", -1, slice.stop);
+    assert_values_match("step", "%d", -5, slice.step);
+}
+
+void test_ndslice_1() {
+    /*
+        Reference Python code:
+        ```python
+        ndarray = np.arange(12, dtype=np.float64).reshape((3, 4));
+        # array([[ 0.,  1.,  2.,  3.],
+        #        [ 4.,  5.,  6.,  7.],
+        #        [ 8.,  9., 10., 11.]])
+
+        dst_ndarray = ndarray[-2:, 1::2]
+        # array([[ 5.,  7.],
+        #        [ 9., 11.]])
+
+        assert dst_ndarray.shape == (2, 2)
+        assert dst_ndarray.strides == (32, 16)
+        assert dst_ndarray[0, 0] == 5.0
+        assert dst_ndarray[0, 1] == 7.0
+        assert dst_ndarray[1, 0] == 9.0
+        assert dst_ndarray[1, 1] == 11.0
+        ```
+    */
+    BEGIN_TEST();
+
+    double in_data[12] = { 0.0, 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0 };
+    int32_t in_itemsize = sizeof(double);
+    const int32_t in_ndims = 2;
+    int32_t in_shape[in_ndims] = { 3, 4 };
+    int32_t in_strides[in_ndims] = {};
+    NDArray<int32_t> ndarray = {
+        .data = (uint8_t*) in_data,
+        .itemsize = in_itemsize,
+        .ndims = in_ndims,
+        .shape = in_shape,
+        .strides = in_strides
+    };
+    ndarray.set_strides_by_shape();
+
+    // Destination ndarray
+    // As documented, ndims and shape & strides must be allocated and determined by the caller.
+    const int32_t dst_ndims = 2;
+    int32_t dst_shape[dst_ndims] = {999, 999}; // Empty values
+    int32_t dst_strides[dst_ndims] = {999, 999}; // Empty values
+    NDArray<int32_t> dst_ndarray = {
+        .data = nullptr,
+        .ndims = dst_ndims,
+        .shape = dst_shape,
+        .strides = dst_strides
+    };
+
+    // Create the slice in `ndarray[-2::, 1::2]`
+    UserSlice<int32_t> user_slice_1 = {
+        .start_defined = 1,
+        .start = -2,
+        .stop_defined = 0,
+        .step_defined = 0
+    };
+
+    UserSlice<int32_t> user_slice_2 = {
+        .start_defined = 1,
+        .start = 1,
+        .stop_defined = 0,
+        .step_defined = 1,
+        .step = 2
+    };
+
+    const int32_t num_ndslices = 2;
+    NDSlice ndslices[num_ndslices] = {
+        { .type = INPUT_SLICE_TYPE_SLICE, .slice = (uint8_t*) &user_slice_1 },
+        { .type = INPUT_SLICE_TYPE_SLICE, .slice = (uint8_t*) &user_slice_2 }
+    };
+
+    ndarray.slice(num_ndslices, ndslices, &dst_ndarray);
+
+    int32_t expected_shape[dst_ndims] = { 2, 2 };
+    int32_t expected_strides[dst_ndims] = { 32, 16 };
+    assert_arrays_match("shape", "%d", dst_ndims, expected_shape, dst_ndarray.shape);
+    assert_arrays_match("strides", "%d", dst_ndims, expected_strides, dst_ndarray.strides);
+
+    assert_values_match("dst_ndarray[0, 0]", "%f", 5.0, *((double *) dst_ndarray.get_pelement((int32_t[dst_ndims]) { 0, 0 })));
+    assert_values_match("dst_ndarray[0, 1]", "%f", 7.0, *((double *) dst_ndarray.get_pelement((int32_t[dst_ndims]) { 0, 1 })));
+    assert_values_match("dst_ndarray[1, 0]", "%f", 9.0, *((double *) dst_ndarray.get_pelement((int32_t[dst_ndims]) { 1, 0 })));
+    assert_values_match("dst_ndarray[1, 1]", "%f", 11.0, *((double *) dst_ndarray.get_pelement((int32_t[dst_ndims]) { 1, 1 })));
+}
+
+void test_ndslice_2() {
+    /*
+        ```python
+        ndarray = np.arange(12, dtype=np.float64).reshape((3, 4))
+        # array([[ 0.,  1.,  2.,  3.],
+        #        [ 4.,  5.,  6.,  7.],
+        #        [ 8.,  9., 10., 11.]])
+
+        dst_ndarray = ndarray[2, ::-2]
+        # array([11.,  9.])
+
+        assert dst_ndarray.shape == (2,)
+        assert dst_ndarray.strides == (-16,)
+        assert dst_ndarray[0] == 11.0
+        assert dst_ndarray[1] == 9.0
+
+        dst_ndarray[1, 0] == 99 # If you write to `dst_ndarray`
+        assert ndarray[1, 3] == 99 # `ndarray` also updates!!
+        ```
+    */
+    BEGIN_TEST();
+
+    double in_data[12] = { 0.0, 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0 };
+    int32_t in_itemsize = sizeof(double);
+    const int32_t in_ndims = 2;
+    int32_t in_shape[in_ndims] = { 3, 4 };
+    int32_t in_strides[in_ndims] = {};
+    NDArray<int32_t> ndarray = {
+        .data = (uint8_t*) in_data,
+        .itemsize = in_itemsize,
+        .ndims = in_ndims,
+        .shape = in_shape,
+        .strides = in_strides
+    };
+    ndarray.set_strides_by_shape();
+
+    // Destination ndarray
+    // As documented, ndims and shape & strides must be allocated and determined by the caller.
+    const int32_t dst_ndims = 1;
+    int32_t dst_shape[dst_ndims] = {999}; // Empty values
+    int32_t dst_strides[dst_ndims] = {999}; // Empty values
+    NDArray<int32_t> dst_ndarray = {
+        .data = nullptr,
+        .ndims = dst_ndims,
+        .shape = dst_shape,
+        .strides = dst_strides
+    };
+
+    // Create the slice in `ndarray[2, ::-2]`
+    int32_t user_slice_1 = 2;
+    UserSlice<int32_t> user_slice_2 = {
+        .start_defined = 0,
+        .stop_defined = 0,
+        .step_defined = 1,
+        .step = -2
+    };
+
+    const int32_t num_ndslices = 2;
+    NDSlice ndslices[num_ndslices] = {
+        { .type = INPUT_SLICE_TYPE_INDEX, .slice = (uint8_t*) &user_slice_1 },
+        { .type = INPUT_SLICE_TYPE_SLICE, .slice = (uint8_t*) &user_slice_2 }
+    };
+
+    ndarray.slice(num_ndslices, ndslices, &dst_ndarray);
+
+    int32_t expected_shape[dst_ndims] = { 2 };
+    int32_t expected_strides[dst_ndims] = { -16 };
+    assert_arrays_match("shape", "%d", dst_ndims, expected_shape, dst_ndarray.shape);
+    assert_arrays_match("strides", "%d", dst_ndims, expected_strides, dst_ndarray.strides);
+
+    // [5.0, 3.0]
+    assert_values_match("dst_ndarray[0]", "%f", 11.0, *((double *) dst_ndarray.get_pelement((int32_t[dst_ndims]) { 0 })));
+    assert_values_match("dst_ndarray[1]", "%f", 9.0, *((double *) dst_ndarray.get_pelement((int32_t[dst_ndims]) { 1 })));
+}
+
+void test_can_broadcast_shape() {
+    BEGIN_TEST();
+
+    assert_values_match(
+        "can_broadcast_shape_to([3], [1, 1, 1, 1, 3]) == true",
+        "%d",
+        true,
+        ndarray_util::can_broadcast_shape_to(1, (int32_t[]) { 3 }, 5, (int32_t[]) { 1, 1, 1, 1, 3 })
+    );
+    assert_values_match(
+        "can_broadcast_shape_to([3], [3, 1]) == false",
+        "%d",
+        false,
+        ndarray_util::can_broadcast_shape_to(1, (int32_t[]) { 3 }, 2, (int32_t[]) { 3, 1 }));
+    assert_values_match(
+        "can_broadcast_shape_to([3], [3]) == true",
+        "%d",
+        true,
+        ndarray_util::can_broadcast_shape_to(1, (int32_t[]) { 3 }, 1, (int32_t[]) { 3 }));
+    assert_values_match(
+        "can_broadcast_shape_to([1], [3]) == false",
+        "%d",
+        false,
+        ndarray_util::can_broadcast_shape_to(1, (int32_t[]) { 1 }, 1, (int32_t[]) { 3 }));
+    assert_values_match(
+        "can_broadcast_shape_to([1], [1]) == true",
+        "%d",
+        true,
+        ndarray_util::can_broadcast_shape_to(1, (int32_t[]) { 1 }, 1, (int32_t[]) { 1 }));
+    assert_values_match(
+        "can_broadcast_shape_to([256, 256, 3], [256, 1, 3]) == true",
+        "%d",
+        true,
+        ndarray_util::can_broadcast_shape_to(3, (int32_t[]) { 256, 256, 3 }, 3, (int32_t[]) { 256, 1, 3 })
+    );
+    assert_values_match(
+        "can_broadcast_shape_to([256, 256, 3], [3]) == true",
+        "%d",
+        true,
+        ndarray_util::can_broadcast_shape_to(3, (int32_t[]) { 256, 256, 3 }, 1, (int32_t[]) { 3 })
+    );
+    assert_values_match(
+        "can_broadcast_shape_to([256, 256, 3], [2]) == false",
+        "%d",
+        false,
+        ndarray_util::can_broadcast_shape_to(3, (int32_t[]) { 256, 256, 3 }, 1, (int32_t[]) { 2 })
+    );
+    assert_values_match(
+        "can_broadcast_shape_to([256, 256, 3], [1]) == true",
+        "%d",
+        true,
+        ndarray_util::can_broadcast_shape_to(3, (int32_t[]) { 256, 256, 3 }, 1, (int32_t[]) { 1 })
+    );
+
+    // In cases when the shapes contain zero(es)
+    assert_values_match(
+        "can_broadcast_shape_to([0], [1]) == true",
+        "%d",
+        true,
+        ndarray_util::can_broadcast_shape_to(1, (int32_t[]) { 0 }, 1, (int32_t[]) { 1 })
+    );
+    assert_values_match(
+        "can_broadcast_shape_to([0], [2]) == false",
+        "%d",
+        false,
+        ndarray_util::can_broadcast_shape_to(1, (int32_t[]) { 0 }, 1, (int32_t[]) { 2 })
+    );
+    assert_values_match(
+        "can_broadcast_shape_to([0, 4, 0, 0], [1]) == true",
+        "%d",
+        true,
+        ndarray_util::can_broadcast_shape_to(4, (int32_t[]) { 0, 4, 0, 0 }, 1, (int32_t[]) { 1 })
+    );
+    assert_values_match(
+        "can_broadcast_shape_to([0, 4, 0, 0], [1, 1, 1, 1]) == true",
+        "%d",
+        true,
+        ndarray_util::can_broadcast_shape_to(4, (int32_t[]) { 0, 4, 0, 0 }, 4, (int32_t[]) { 1, 1, 1, 1 })
+    );
+    assert_values_match(
+        "can_broadcast_shape_to([0, 4, 0, 0], [1, 4, 1, 1]) == true",
+        "%d",
+        true,
+        ndarray_util::can_broadcast_shape_to(4, (int32_t[]) { 0, 4, 0, 0 }, 4, (int32_t[]) { 1, 4, 1, 1 })
+    );
+    assert_values_match(
+        "can_broadcast_shape_to([4, 3], [0, 3]) == false",
+        "%d",
+        false,
+        ndarray_util::can_broadcast_shape_to(2, (int32_t[]) { 4, 3 }, 2, (int32_t[]) { 0, 3 })
+    );
+    assert_values_match(
+        "can_broadcast_shape_to([4, 3], [0, 0]) == false",
+        "%d",
+        false,
+        ndarray_util::can_broadcast_shape_to(2, (int32_t[]) { 4, 3 }, 2, (int32_t[]) { 0, 0 })
+    );
+}
+
+void test_ndarray_broadcast_1() {
+    /*
+    # array = np.array([[19.9, 29.9, 39.9, 49.9]], dtype=np.float64)
+    # >>> [[19.9 29.9 39.9 49.9]]
+    #
+    # array = np.broadcast_to(array, (2, 3, 4))
+    # >>> [[[19.9 29.9 39.9 49.9]
+    # >>>   [19.9 29.9 39.9 49.9]
+    # >>>   [19.9 29.9 39.9 49.9]]
+    # >>>  [[19.9 29.9 39.9 49.9]
+    # >>>   [19.9 29.9 39.9 49.9]
+    # >>>   [19.9 29.9 39.9 49.9]]]
+    #
+    # assery array.strides == (0, 0, 8)
+
+    */
+    BEGIN_TEST();
+
+    double in_data[4] = { 19.9, 29.9, 39.9, 49.9 };
+    const int32_t in_ndims = 2;
+    int32_t in_shape[in_ndims] = {1, 4};
+    int32_t in_strides[in_ndims] = {};
+    NDArray<int32_t> ndarray = {
+        .data = (uint8_t*) in_data,
+        .itemsize = sizeof(double),
+        .ndims = in_ndims,
+        .shape = in_shape,
+        .strides = in_strides
+    };
+    ndarray.set_strides_by_shape();
+
+    const int32_t dst_ndims = 3;
+    int32_t dst_shape[dst_ndims] = {2, 3, 4};
+    int32_t dst_strides[dst_ndims] = {};
+    NDArray<int32_t> dst_ndarray = {
+        .ndims = dst_ndims,
+        .shape = dst_shape,
+        .strides = dst_strides
+    };
+
+    ndarray.broadcast_to(&dst_ndarray);
+
+    assert_arrays_match("dst_ndarray->strides", "%d", dst_ndims, (int32_t[]) { 0, 0, 8 }, dst_ndarray.strides);
+
+    assert_values_match("dst_ndarray[0, 0, 0]", "%f", 19.9, *((double*) dst_ndarray.get_pelement((int32_t[]) {0, 0, 0})));
+    assert_values_match("dst_ndarray[0, 0, 1]", "%f", 29.9, *((double*) dst_ndarray.get_pelement((int32_t[]) {0, 0, 1})));
+    assert_values_match("dst_ndarray[0, 0, 2]", "%f", 39.9, *((double*) dst_ndarray.get_pelement((int32_t[]) {0, 0, 2})));
+    assert_values_match("dst_ndarray[0, 0, 3]", "%f", 49.9, *((double*) dst_ndarray.get_pelement((int32_t[]) {0, 0, 3})));
+    assert_values_match("dst_ndarray[0, 1, 0]", "%f", 19.9, *((double*) dst_ndarray.get_pelement((int32_t[]) {0, 1, 0})));
+    assert_values_match("dst_ndarray[0, 1, 1]", "%f", 29.9, *((double*) dst_ndarray.get_pelement((int32_t[]) {0, 1, 1})));
+    assert_values_match("dst_ndarray[0, 1, 2]", "%f", 39.9, *((double*) dst_ndarray.get_pelement((int32_t[]) {0, 1, 2})));
+    assert_values_match("dst_ndarray[0, 1, 3]", "%f", 49.9, *((double*) dst_ndarray.get_pelement((int32_t[]) {0, 1, 3})));
+    assert_values_match("dst_ndarray[1, 2, 3]", "%f", 49.9, *((double*) dst_ndarray.get_pelement((int32_t[]) {1, 2, 3})));
+}
+
+void test_assign_with() {
+    /*
+        ```
+        xs = np.array([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0], [7.0, 8.0, 9.0]], dtype=np.float64)
+        ys = xs.shape
+        ```
+    */
+}
+
+int main() {
+    test_calc_size_from_shape_normal();
+    test_calc_size_from_shape_has_zero();
+    test_set_strides_by_shape();
+    test_ndarray_indices_iter_normal();
+    test_ndarray_fill_generic();
+    test_ndarray_set_to_eye();
+    test_slice_1();
+    test_slice_2();
+    test_slice_3();
+    test_slice_4();
+    test_ndslice_1();
+    test_ndslice_2();
+    test_can_broadcast_shape();
+    test_ndarray_broadcast_1();
+    test_assign_with();
+    return 0;
+}

nac3core/irrt/irrt_typedefs.hpp

@@ -0,0 +1,14 @@
+#pragma once
+
+// This is made toggleable since `irrt_test.cpp` itself would include
+// headers that define the `int_t` family.
+#ifndef IRRT_DONT_TYPEDEF_INTS
+typedef _BitInt(8) int8_t;
+typedef unsigned _BitInt(8) uint8_t;
+typedef _BitInt(32) int32_t;
+typedef unsigned _BitInt(32) uint32_t;
+typedef _BitInt(64) int64_t;
+typedef unsigned _BitInt(64) uint64_t;
+#endif
+
+typedef int32_t SliceIndex;

nac3core/irrt/irrt_utils.hpp

@@ -0,0 +1,37 @@
+#pragma once
+
+#include "irrt_typedefs.hpp"
+
+namespace {
+    template <typename T>
+    T max(T a, T b) {
+        return a > b ? a : b;
+    }
+
+    template <typename T>
+    T min(T a, T b) {
+        return a > b ? b : a;
+    }
+
+    template <typename T>
+    bool arrays_match(int len, T *as, T *bs) {
+        for (int i = 0; i < len; i++) {
+            if (as[i] != bs[i]) return false;
+        }
+        return true;
+    }
+
+    void irrt_panic() {
+        // Crash the program for now.
+        // TODO: Don't crash the program
+        //       ... or at least produce a good message when doing testing IRRT
+            
+        uint8_t* death = nullptr;
+        *death = 0; // TODO: address 0 on hardware might be writable?
+    }
+
+    // TODO: Make this a macro and allow it to be toggled on/off (e.g., debug vs release)
+    void irrt_assert(bool condition) {
+        if (!condition) irrt_panic();
+    }
+}

nac3core/nac3core_derive/Cargo.toml

@@ -1,21 +0,0 @@
-[package]
-name = "nac3core_derive"
-version = "0.1.0"
-edition = "2024"
-
-[lib]
-proc-macro = true
-
-[[test]]
-name = "structfields_tests"
-path = "tests/structfields_test.rs"
-
-[dev-dependencies]
-nac3core = { path = ".." }
-trybuild = { version = "1.0", features = ["diff"] }
-
-[dependencies]
-proc-macro2 = "1.0"
-proc-macro-error = "1.0"
-syn = "2.0"
-quote = "1.0"

nac3core/nac3core_derive/src/lib.rs

@@ -1,315 +0,0 @@
-use proc_macro::TokenStream;
-use proc_macro_error::{abort, proc_macro_error};
-use quote::quote;
-use syn::{
-    Data, DataStruct, Expr, ExprField, ExprMethodCall, ExprPath, GenericArgument, Ident, LitStr,
-    Path, PathArguments, Type, TypePath, parse_macro_input, spanned::Spanned,
-};
-
-/// Extracts all generic arguments of a [`Type`] into a [`Vec`].
-///
-/// Returns [`Some`] of a possibly-empty [`Vec`] if the path of `ty` matches with
-/// `expected_ty_name`, otherwise returns [`None`].
-fn extract_generic_args(expected_ty_name: &'static str, ty: &Type) -> Option<Vec<GenericArgument>> {
-    let Type::Path(TypePath { qself: None, path, .. }) = ty else {
-        return None;
-    };
-
-    let segments = &path.segments;
-    if segments.len() != 1 {
-        return None;
-    };
-
-    let segment = segments.iter().next().unwrap();
-    if segment.ident != expected_ty_name {
-        return None;
-    }
-
-    let PathArguments::AngleBracketed(path_args) = &segment.arguments else {
-        return Some(Vec::new());
-    };
-    let args = &path_args.args;
-
-    Some(args.iter().cloned().collect::<Vec<_>>())
-}
-
-/// Maps a `path` matching one of the `target_idents` into the `replacement` [`Ident`].
-fn map_path_to_ident(path: &Path, target_idents: &[&str], replacement: &str) -> Option<Ident> {
-    path.require_ident()
-        .ok()
-        .filter(|ident| target_idents.iter().any(|target| ident == target))
-        .map(|ident| Ident::new(replacement, ident.span()))
-}
-
-/// Extracts the left-hand side of a dot-expression.
-fn extract_dot_operand(expr: &Expr) -> Option<&Expr> {
-    match expr {
-        Expr::MethodCall(ExprMethodCall { receiver: operand, .. })
-        | Expr::Field(ExprField { base: operand, .. }) => Some(operand),
-        _ => None,
-    }
-}
-
-/// Replaces the top-level receiver of a dot-expression with an [`Ident`], returning `Some(&mut expr)` if the
-/// replacement is performed.
-///
-/// The top-level receiver is the left-most receiver expression, e.g. the top-level receiver of `a.b.c.foo()` is `a`.
-fn replace_top_level_receiver(expr: &mut Expr, ident: Ident) -> Option<&mut Expr> {
-    if let Expr::MethodCall(ExprMethodCall { receiver: operand, .. })
-    | Expr::Field(ExprField { base: operand, .. }) = expr
-    {
-        return if extract_dot_operand(operand).is_some() {
-            if replace_top_level_receiver(operand, ident).is_some() { Some(expr) } else { None }
-        } else {
-            *operand = Box::new(Expr::Path(ExprPath {
-                attrs: Vec::default(),
-                qself: None,
-                path: ident.into(),
-            }));
-
-            Some(expr)
-        };
-    }
-
-    None
-}
-
-/// Iterates all operands to the left-hand side of the `.` of an [expression][`Expr`], i.e. the container operand of all
-/// [`Expr::Field`] and the receiver operand of all [`Expr::MethodCall`].
-///
-/// The iterator will return the operand expressions in reverse order of appearance. For example, `a.b.c.func()` will
-/// return `vec![c, b, a]`.
-fn iter_dot_operands(expr: &Expr) -> impl Iterator<Item = &Expr> {
-    let mut o = extract_dot_operand(expr);
-
-    std::iter::from_fn(move || {
-        let this = o;
-        o = o.as_ref().and_then(|o| extract_dot_operand(o));
-
-        this
-    })
-}
-
-/// Normalizes a value expression for use when creating an instance of this structure, returning a
-/// [`proc_macro2::TokenStream`] of tokens representing the normalized expression.
-fn normalize_value_expr(expr: &Expr) -> proc_macro2::TokenStream {
-    match &expr {
-        Expr::Path(ExprPath { qself: None, path, .. }) => {
-            if let Some(ident) = map_path_to_ident(path, &["usize", "size_t"], "llvm_usize") {
-                quote! { #ident }
-            } else {
-                abort!(
-                    path,
-                    format!(
-                        "Expected one of `size_t`, `usize`, or an implicit call expression in #[value_type(...)], found {}",
-                        quote!(#expr).to_string(),
-                    )
-                )
-            }
-        }
-
-        Expr::Call(_) => {
-            quote! { ctx.#expr }
-        }
-
-        Expr::MethodCall(_) => {
-            let base_receiver = iter_dot_operands(expr).last();
-
-            match base_receiver {
-                // `usize.{...}`, `size_t.{...}` -> Rewrite the identifiers to `llvm_usize`
-                Some(Expr::Path(ExprPath { qself: None, path, .. }))
-                    if map_path_to_ident(path, &["usize", "size_t"], "llvm_usize").is_some() =>
-                {
-                    let ident =
-                        map_path_to_ident(path, &["usize", "size_t"], "llvm_usize").unwrap();
-
-                    let mut expr = expr.clone();
-                    let expr = replace_top_level_receiver(&mut expr, ident).unwrap();
-
-                    quote!(#expr)
-                }
-
-                // `ctx.{...}`, `context.{...}` -> Rewrite the identifiers to `ctx`
-                Some(Expr::Path(ExprPath { qself: None, path, .. }))
-                    if map_path_to_ident(path, &["ctx", "context"], "ctx").is_some() =>
-                {
-                    let ident = map_path_to_ident(path, &["ctx", "context"], "ctx").unwrap();
-
-                    let mut expr = expr.clone();
-                    let expr = replace_top_level_receiver(&mut expr, ident).unwrap();
-
-                    quote!(#expr)
-                }
-
-                // No reserved identifier prefix -> Prepend `ctx.` to the entire expression
-                _ => quote! { ctx.#expr },
-            }
-        }
-
-        _ => {
-            abort!(
-                expr,
-                format!(
-                    "Expected one of `size_t`, `usize`, or an implicit call expression in #[value_type(...)], found {}",
-                    quote!(#expr).to_string(),
-                )
-            )
-        }
-    }
-}
-
-/// Derives an implementation of `codegen::types::structure::StructFields`.
-///
-/// The benefit of using `#[derive(StructFields)]` is that all index- or order-dependent logic required by
-/// `impl StructFields` is automatically generated by this implementation, including the field index as required by
-/// `StructField::new` and the fields as returned by `StructFields::to_vec`.
-///
-/// # Prerequisites
-///
-/// In order to derive from [`StructFields`], you must implement (or derive) [`Eq`] and [`Copy`] as required by
-/// `StructFields`.
-///
-/// Moreover, `#[derive(StructFields)]` can only be used for `struct`s with named fields, and may only contain fields
-/// with either `StructField` or [`PhantomData`] types.
-///
-/// # Attributes for [`StructFields`]
-///
-/// Each `StructField` field must be declared with the `#[value_type(...)]` attribute. The argument of `value_type`
-/// accepts one of the following:
-///
-/// - An expression returning an instance of `inkwell::types::BasicType` (with or without the receiver `ctx`/`context`).
-/// For example, `context.i8_type()`, `ctx.i8_type()`, and `i8_type()` all refer to `i8`.
-/// - The reserved identifiers `usize` and `size_t` referring to an `inkwell::types::IntType` of the platform-dependent
-/// integer size. `usize` and `size_t` can also be used as the receiver to other method calls, e.g.
-/// `usize.array_type(3)`.
-///
-/// # Example
-///
-/// The following is an example of an LLVM slice implemented using `#[derive(StructFields)]`.
-///
-/// ```rust,ignore
-/// use nac3core::{
-///     codegen::types::structure::StructField,
-///     inkwell::{
-///         values::{IntValue, PointerValue},
-///         AddressSpace,
-///     },
-/// };
-/// use nac3core_derive::StructFields;
-///
-/// // All classes that implement StructFields must also implement Eq and Copy
-/// #[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-/// pub struct SliceValue<'ctx> {
-///     // Declares ptr have a value type of i8*
-///     //
-///     // Can also be written as `ctx.i8_type().ptr_type(...)` or `context.i8_type().ptr_type(...)`
-///     #[value_type(i8_type().ptr_type(AddressSpace::default()))]
-///     ptr: StructField<'ctx, PointerValue<'ctx>>,
-///
-///     // Declares len have a value type of usize, depending on the target compilation platform
-///     #[value_type(usize)]
-///     len: StructField<'ctx, IntValue<'ctx>>,
-/// }
-/// ```
-#[proc_macro_derive(StructFields, attributes(value_type))]
-#[proc_macro_error]
-pub fn derive(input: TokenStream) -> TokenStream {
-    let input = parse_macro_input!(input as syn::DeriveInput);
-    let ident = &input.ident;
-
-    let Data::Struct(DataStruct { fields, .. }) = &input.data else {
-        abort!(input, "Only structs with named fields are supported");
-    };
-    if let Err(err_span) = fields
-        .iter()
-        .try_for_each(|field| if field.ident.is_some() { Ok(()) } else { Err(field.span()) })
-    {
-        abort!(err_span, "Only structs with named fields are supported");
-    };
-
-    // Check if struct<'ctx>
-    if input.generics.params.len() != 1 {
-        abort!(input.generics, "Expected exactly 1 generic parameter")
-    }
-
-    let phantom_info = fields
-        .iter()
-        .filter(|field| extract_generic_args("PhantomData", &field.ty).is_some())
-        .map(|field| field.ident.as_ref().unwrap())
-        .cloned()
-        .collect::<Vec<_>>();
-
-    let field_info = fields
-        .iter()
-        .filter(|field| extract_generic_args("PhantomData", &field.ty).is_none())
-        .map(|field| {
-            let ident = field.ident.as_ref().unwrap();
-            let ty = &field.ty;
-
-            let Some(_) = extract_generic_args("StructField", ty) else {
-                abort!(field, "Only StructField and PhantomData are allowed")
-            };
-
-            let attrs = &field.attrs;
-            let Some(value_type_attr) =
-                attrs.iter().find(|attr| attr.path().is_ident("value_type"))
-            else {
-                abort!(field, "Expected #[value_type(...)] attribute for field");
-            };
-
-            let Ok(value_type_expr) = value_type_attr.parse_args::<Expr>() else {
-                abort!(value_type_attr, "Expected expression in #[value_type(...)]");
-            };
-
-            let value_expr_toks = normalize_value_expr(&value_type_expr);
-
-            (ident.clone(), value_expr_toks)
-        })
-        .collect::<Vec<_>>();
-
-    // `<*>::new` impl of `StructField` and `PhantomData` for `StructFields::new`
-    let phantoms_create = phantom_info
-        .iter()
-        .map(|id| quote! { #id: ::std::marker::PhantomData })
-        .collect::<Vec<_>>();
-    let fields_create = field_info
-        .iter()
-        .map(|(id, ty)| {
-            let id_lit = LitStr::new(&id.to_string(), id.span());
-            quote! {
-                #id: ::nac3core::codegen::types::structure::StructField::create(
-                    &mut counter,
-                    #id_lit,
-                    #ty,
-                )
-            }
-        })
-        .collect::<Vec<_>>();
-
-    // `.into()` impl of `StructField` for `StructFields::to_vec`
-    let fields_into =
-        field_info.iter().map(|(id, _)| quote! { self.#id.into() }).collect::<Vec<_>>();
-
-    let impl_block = quote! {
-        impl<'ctx> ::nac3core::codegen::types::structure::StructFields<'ctx> for #ident<'ctx> {
-            fn new(ctx: impl ::nac3core::inkwell::context::AsContextRef<'ctx>, llvm_usize: ::nac3core::inkwell::types::IntType<'ctx>) -> Self {
-                let ctx = unsafe { ::nac3core::inkwell::context::ContextRef::new(ctx.as_ctx_ref()) };
-
-                let mut counter = ::nac3core::codegen::types::structure::FieldIndexCounter::default();
-
-                #ident {
-                    #(#fields_create),*
-                    #(#phantoms_create),*
-                }
-            }
-
-            fn to_vec(&self) -> ::std::vec::Vec<(&'static str, ::nac3core::inkwell::types::BasicTypeEnum<'ctx>)> {
-                vec![
-                    #(#fields_into),*
-                ]
-            }
-        }
-    };
-
-    impl_block.into()
-}

nac3core/nac3core_derive/tests/structfields_empty.rs

@@ -1,9 +0,0 @@
-use nac3core_derive::StructFields;
-use std::marker::PhantomData;
-
-#[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-pub struct EmptyValue<'ctx> {
-    _phantom: PhantomData<&'ctx ()>,
-}
-
-fn main() {}

nac3core/nac3core_derive/tests/structfields_ndarray.rs

@@ -1,20 +0,0 @@
-use nac3core::{
-    codegen::types::structure::StructField,
-    inkwell::{
-        AddressSpace,
-        values::{IntValue, PointerValue},
-    },
-};
-use nac3core_derive::StructFields;
-
-#[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-pub struct NDArrayValue<'ctx> {
-    #[value_type(usize)]
-    ndims: StructField<'ctx, IntValue<'ctx>>,
-    #[value_type(usize.ptr_type(AddressSpace::default()))]
-    shape: StructField<'ctx, PointerValue<'ctx>>,
-    #[value_type(i8_type().ptr_type(AddressSpace::default()))]
-    data: StructField<'ctx, PointerValue<'ctx>>,
-}
-
-fn main() {}

nac3core/nac3core_derive/tests/structfields_slice.rs

@@ -1,18 +0,0 @@
-use nac3core::{
-    codegen::types::structure::StructField,
-    inkwell::{
-        AddressSpace,
-        values::{IntValue, PointerValue},
-    },
-};
-use nac3core_derive::StructFields;
-
-#[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-pub struct SliceValue<'ctx> {
-    #[value_type(i8_type().ptr_type(AddressSpace::default()))]
-    ptr: StructField<'ctx, PointerValue<'ctx>>,
-    #[value_type(usize)]
-    len: StructField<'ctx, IntValue<'ctx>>,
-}
-
-fn main() {}

nac3core/nac3core_derive/tests/structfields_slice_context.rs

@@ -1,18 +0,0 @@
-use nac3core::{
-    codegen::types::structure::StructField,
-    inkwell::{
-        AddressSpace,
-        values::{IntValue, PointerValue},
-    },
-};
-use nac3core_derive::StructFields;
-
-#[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-pub struct SliceValue<'ctx> {
-    #[value_type(context.i8_type().ptr_type(AddressSpace::default()))]
-    ptr: StructField<'ctx, PointerValue<'ctx>>,
-    #[value_type(usize)]
-    len: StructField<'ctx, IntValue<'ctx>>,
-}
-
-fn main() {}

nac3core/nac3core_derive/tests/structfields_slice_ctx.rs

@@ -1,18 +0,0 @@
-use nac3core::{
-    codegen::types::structure::StructField,
-    inkwell::{
-        AddressSpace,
-        values::{IntValue, PointerValue},
-    },
-};
-use nac3core_derive::StructFields;
-
-#[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-pub struct SliceValue<'ctx> {
-    #[value_type(ctx.i8_type().ptr_type(AddressSpace::default()))]
-    ptr: StructField<'ctx, PointerValue<'ctx>>,
-    #[value_type(usize)]
-    len: StructField<'ctx, IntValue<'ctx>>,
-}
-
-fn main() {}

nac3core/nac3core_derive/tests/structfields_slice_sizet.rs

@@ -1,18 +0,0 @@
-use nac3core::{
-    codegen::types::structure::StructField,
-    inkwell::{
-        AddressSpace,
-        values::{IntValue, PointerValue},
-    },
-};
-use nac3core_derive::StructFields;
-
-#[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-pub struct SliceValue<'ctx> {
-    #[value_type(i8_type().ptr_type(AddressSpace::default()))]
-    ptr: StructField<'ctx, PointerValue<'ctx>>,
-    #[value_type(size_t)]
-    len: StructField<'ctx, IntValue<'ctx>>,
-}
-
-fn main() {}

nac3core/nac3core_derive/tests/structfields_test.rs

@@ -1,10 +0,0 @@
-#[test]
-fn test_parse_empty() {
-    let t = trybuild::TestCases::new();
-    t.pass("tests/structfields_empty.rs");
-    t.pass("tests/structfields_slice.rs");
-    t.pass("tests/structfields_slice_ctx.rs");
-    t.pass("tests/structfields_slice_context.rs");
-    t.pass("tests/structfields_slice_sizet.rs");
-    t.pass("tests/structfields_ndarray.rs");
-}

nac3core/src/codegen/concrete_type.rs

@@ -1,20 +1,18 @@
-use std::collections::HashMap;
-
-use indexmap::IndexMap;
-
-use nac3parser::ast::StrRef;
-
 use crate::{
     symbol_resolver::SymbolValue,
     toplevel::DefinitionId,
     typecheck::{
         type_inferencer::PrimitiveStore,
         typedef::{
-            FunSignature, FuncArg, Type, TypeEnum, TypeVar, TypeVarId, Unifier, into_var_map,
+            into_var_map, FunSignature, FuncArg, Type, TypeEnum, TypeVar, TypeVarId, Unifier,
         },
     },
 };
 
+use indexmap::IndexMap;
+use nac3parser::ast::StrRef;
+use std::collections::HashMap;
+
 pub struct ConcreteTypeStore {
     store: Vec<ConcreteTypeEnum>,
 }
@@ -27,7 +25,6 @@ pub struct ConcreteFuncArg {
     pub name: StrRef,
     pub ty: ConcreteType,
     pub default_value: Option<SymbolValue>,
-    pub is_vararg: bool,
 }
 
 #[derive(Clone, Debug)]
@@ -49,17 +46,12 @@ pub enum ConcreteTypeEnum {
     TPrimitive(Primitive),
     TTuple {
         ty: Vec<ConcreteType>,
-        is_vararg_ctx: bool,
     },
     TObj {
         obj_id: DefinitionId,
         fields: HashMap<StrRef, (ConcreteType, bool)>,
         params: IndexMap<TypeVarId, ConcreteType>,
     },
-    TModule {
-        module_id: DefinitionId,
-        methods: HashMap<StrRef, (ConcreteType, bool)>,
-    },
     TVirtual {
         ty: ConcreteType,
     },
@@ -110,16 +102,8 @@ impl ConcreteTypeStore {
                 .iter()
                 .map(|arg| ConcreteFuncArg {
                     name: arg.name,
-                    ty: if arg.is_vararg {
-                        let tuple_ty = unifier
-                            .add_ty(TypeEnum::TTuple { ty: vec![arg.ty], is_vararg_ctx: true });
-
-                        self.from_unifier_type(unifier, primitives, tuple_ty, cache)
-                    } else {
-                        self.from_unifier_type(unifier, primitives, arg.ty, cache)
-                    },
+                    ty: self.from_unifier_type(unifier, primitives, arg.ty, cache),
                     default_value: arg.default_value.clone(),
-                    is_vararg: arg.is_vararg,
                 })
                 .collect(),
             ret: self.from_unifier_type(unifier, primitives, signature.ret, cache),
@@ -174,12 +158,11 @@ impl ConcreteTypeStore {
             cache.insert(ty, None);
             let ty_enum = unifier.get_ty(ty);
             let result = match &*ty_enum {
-                TypeEnum::TTuple { ty, is_vararg_ctx } => ConcreteTypeEnum::TTuple {
+                TypeEnum::TTuple { ty } => ConcreteTypeEnum::TTuple {
                     ty: ty
                         .iter()
                         .map(|t| self.from_unifier_type(unifier, primitives, *t, cache))
                         .collect(),
-                    is_vararg_ctx: *is_vararg_ctx,
                 },
                 TypeEnum::TObj { obj_id, fields, params } => ConcreteTypeEnum::TObj {
                     obj_id: *obj_id,
@@ -209,19 +192,6 @@ impl ConcreteTypeStore {
                         })
                         .collect(),
                 },
-                TypeEnum::TModule { module_id, attributes } => ConcreteTypeEnum::TModule {
-                    module_id: *module_id,
-                    methods: attributes
-                        .iter()
-                        .filter_map(|(name, ty)| match &*unifier.get_ty(ty.0) {
-                            TypeEnum::TFunc(..) | TypeEnum::TObj { .. } => None,
-                            _ => Some((
-                                *name,
-                                (self.from_unifier_type(unifier, primitives, ty.0, cache), ty.1),
-                            )),
-                        })
-                        .collect(),
-                },
                 TypeEnum::TVirtual { ty } => ConcreteTypeEnum::TVirtual {
                     ty: self.from_unifier_type(unifier, primitives, *ty, cache),
                 },
@@ -278,12 +248,11 @@ impl ConcreteTypeStore {
                 *cache.get_mut(&cty).unwrap() = Some(ty);
                 return ty;
             }
-            ConcreteTypeEnum::TTuple { ty, is_vararg_ctx } => TypeEnum::TTuple {
+            ConcreteTypeEnum::TTuple { ty } => TypeEnum::TTuple {
                 ty: ty
                     .iter()
                     .map(|cty| self.to_unifier_type(unifier, primitives, *cty, cache))
                     .collect(),
-                is_vararg_ctx: *is_vararg_ctx,
             },
             ConcreteTypeEnum::TVirtual { ty } => {
                 TypeEnum::TVirtual { ty: self.to_unifier_type(unifier, primitives, *ty, cache) }
@@ -301,15 +270,6 @@ impl ConcreteTypeStore {
                     TypeVar { id, ty }
                 })),
             },
-            ConcreteTypeEnum::TModule { module_id, methods } => TypeEnum::TModule {
-                module_id: *module_id,
-                attributes: methods
-                    .iter()
-                    .map(|(name, cty)| {
-                        (*name, (self.to_unifier_type(unifier, primitives, cty.0, cache), cty.1))
-                    })
-                    .collect::<HashMap<_, _>>(),
-            },
             ConcreteTypeEnum::TFunc { args, ret, vars } => TypeEnum::TFunc(FunSignature {
                 args: args
                     .iter()
@@ -317,7 +277,6 @@ impl ConcreteTypeStore {
                         name: arg.name,
                         ty: self.to_unifier_type(unifier, primitives, arg.ty, cache),
                         default_value: arg.default_value.clone(),
-                        is_vararg: false,
                     })
                     .collect(),
                 ret: self.to_unifier_type(unifier, primitives, *ret, cache),

nac3core/src/codegen/extern_fns.rs

@@ -1,9 +1,8 @@
-use inkwell::{
-    attributes::{Attribute, AttributeLoc},
-    values::{BasicValueEnum, FloatValue},
-};
+use inkwell::attributes::{Attribute, AttributeLoc};
+use inkwell::values::{BasicValueEnum, CallSiteValue, FloatValue, IntValue};
+use itertools::Either;
 
-use super::{CodeGenContext, expr::infer_and_call_function};
+use crate::codegen::CodeGenContext;
 
 /// Macro to generate extern function
 /// Both function return type and function parameter type are `FloatValue`
@@ -14,11 +13,11 @@ use super::{CodeGenContext, expr::infer_and_call_function};
 /// * `$extern_fn:literal`: Name of underlying extern function
 ///
 /// Optional Arguments:
-/// * `$(,$attributes:literal)*)`: Attributes linked with the extern function.
-///   The default attributes are "mustprogress", "nofree", "nounwind", "willreturn", and "writeonly".
-///   These will be used unless other attributes are specified
+/// * `$(,$attributes:literal)*)`: Attributes linked with the extern function
+/// The default attributes are "mustprogress", "nofree", "nounwind", "willreturn", and "writeonly"
+/// These will be used unless other attributes are specified
 /// * `$(,$args:ident)*`: Operands of the extern function
-///   The data type of these operands will be set to `FloatValue`
+/// The data type of these operands will be set to `FloatValue`
 ///  
 macro_rules! generate_extern_fn {
     ("unary", $fn_name:ident, $extern_fn:literal) => {
@@ -36,8 +35,8 @@ macro_rules! generate_extern_fn {
     ($fn_name:ident, $extern_fn:literal $(,$args:ident)* $(,$attributes:literal)*) => {
         #[doc = concat!("Invokes the [`", stringify!($extern_fn), "`](https://en.cppreference.com/w/c/numeric/math/", stringify!($llvm_name), ") function." )]
         pub fn $fn_name<'ctx>(
-            ctx: &CodeGenContext<'ctx, '_>,
-            $($args: FloatValue<'ctx>,)*
+            ctx: &CodeGenContext<'ctx, '_>
+            $(,$args: FloatValue<'ctx>)*,
             name: Option<&str>,
         ) -> FloatValue<'ctx> {
             const FN_NAME: &str = $extern_fn;
@@ -45,87 +44,89 @@ macro_rules! generate_extern_fn {
             let llvm_f64 = ctx.ctx.f64_type();
             $(debug_assert_eq!($args.get_type(), llvm_f64);)*
 
-            infer_and_call_function(
-                ctx,
-                FN_NAME,
-                Some(llvm_f64.into()),
-                &[$($args.into()),*],
-                name,
-                Some(&|func| {
-                   for attr in [$($attributes),*] {
-                        func.add_attribute(
-                            AttributeLoc::Function,
-                            ctx.ctx.create_enum_attribute(Attribute::get_named_enum_kind_id(attr), 0),
-                        );
-                    }
-                })
-            )
-            .map(BasicValueEnum::into_float_value)
-            .unwrap()
-        }
-    };
-}
-
-generate_extern_fn!("unary", call_j1, "j1", "nounwind");
-
-/// Macro to generate `np_linalg` and `sp_linalg` functions
-/// The function takes as input `NDArray` and returns ()
-///
-/// Arguments:
-/// * `$fn_name:ident`: The identifier of the rust function to be generated
-/// * `$extern_fn:ident`: Name of underlying extern function
-/// * (2/3/4): Number of `NDArray` that function takes as input
-///
-/// Note:
-/// The operands and resulting `NDArray` are both passed as input to the funcion
-/// It is the responsibility of caller to ensure that output `NDArray` is properly allocated on stack
-/// The function changes the content of the output `NDArray` in-place
-macro_rules! generate_linalg_extern_fn {
-    ($fn_name:ident, $extern_fn:ident, 2) => {
-        generate_linalg_extern_fn!($fn_name, $extern_fn, mat1, mat2);
-    };
-    ($fn_name:ident, $extern_fn:ident, 3) => {
-        generate_linalg_extern_fn!($fn_name, $extern_fn, mat1, mat2, mat3);
-    };
-    ($fn_name:ident, $extern_fn:ident, 4) => {
-        generate_linalg_extern_fn!($fn_name, $extern_fn, mat1, mat2, mat3, mat4);
-    };
-    ($fn_name:ident, $extern_fn:ident $(,$input_matrix:ident)*) => {
-        #[doc = concat!("Invokes the linalg `", stringify!($extern_fn), "` function." )]
-        pub fn $fn_name<'ctx>(
-            ctx: &mut CodeGenContext<'ctx, '_>,
-            $($input_matrix: BasicValueEnum<'ctx>,)*
-            name: Option<&str>,
-        ) {
-            const FN_NAME: &str = stringify!($extern_fn);
-
-            infer_and_call_function(
-                ctx,
-                FN_NAME,
-                None,
-                &[$($input_matrix.into(),)*],
-                name,
-                Some(&|func| {
+            let extern_fn = ctx.module.get_function(FN_NAME).unwrap_or_else(|| {
+                let fn_type = llvm_f64.fn_type(&[$($args.get_type().into()),*], false);
+                let func = ctx.module.add_function(FN_NAME, fn_type, None);
+                for attr in [$($attributes),*] {
                     func.add_attribute(
                         AttributeLoc::Function,
-                        ctx.ctx.create_enum_attribute(
-                            Attribute::get_named_enum_kind_id("nounwind"),
-                            0,
-                        ),
-                    )
-                }),
-            );
+                        ctx.ctx.create_enum_attribute(Attribute::get_named_enum_kind_id(attr), 0),
+                    );
+                }
+                func
+            });
+
+            ctx.builder
+                .build_call(extern_fn, &[$($args.into()),*], name.unwrap_or_default())
+                .map(CallSiteValue::try_as_basic_value)
+                .map(|v| v.map_left(BasicValueEnum::into_float_value))
+                .map(Either::unwrap_left)
+                .unwrap()
         }
     };
 }
 
-generate_linalg_extern_fn!(call_np_linalg_cholesky, np_linalg_cholesky, 2);
-generate_linalg_extern_fn!(call_np_linalg_qr, np_linalg_qr, 3);
-generate_linalg_extern_fn!(call_np_linalg_svd, np_linalg_svd, 4);
-generate_linalg_extern_fn!(call_np_linalg_inv, np_linalg_inv, 2);
-generate_linalg_extern_fn!(call_np_linalg_pinv, np_linalg_pinv, 2);
-generate_linalg_extern_fn!(call_np_linalg_matrix_power, np_linalg_matrix_power, 3);
-generate_linalg_extern_fn!(call_np_linalg_det, np_linalg_det, 2);
-generate_linalg_extern_fn!(call_sp_linalg_lu, sp_linalg_lu, 3);
-generate_linalg_extern_fn!(call_sp_linalg_schur, sp_linalg_schur, 3);
-generate_linalg_extern_fn!(call_sp_linalg_hessenberg, sp_linalg_hessenberg, 3);
+generate_extern_fn!("unary", call_tan, "tan");
+generate_extern_fn!("unary", call_asin, "asin");
+generate_extern_fn!("unary", call_acos, "acos");
+generate_extern_fn!("unary", call_atan, "atan");
+generate_extern_fn!("unary", call_sinh, "sinh");
+generate_extern_fn!("unary", call_cosh, "cosh");
+generate_extern_fn!("unary", call_tanh, "tanh");
+generate_extern_fn!("unary", call_asinh, "asinh");
+generate_extern_fn!("unary", call_acosh, "acosh");
+generate_extern_fn!("unary", call_atanh, "atanh");
+generate_extern_fn!("unary", call_expm1, "expm1");
+generate_extern_fn!(
+    "unary",
+    call_cbrt,
+    "cbrt",
+    "mustprogress",
+    "nofree",
+    "nosync",
+    "nounwind",
+    "readonly",
+    "willreturn"
+);
+generate_extern_fn!("unary", call_erf, "erf", "nounwind");
+generate_extern_fn!("unary", call_erfc, "erfc", "nounwind");
+generate_extern_fn!("unary", call_j1, "j1", "nounwind");
+
+generate_extern_fn!("binary", call_atan2, "atan2");
+generate_extern_fn!("binary", call_hypot, "hypot", "nounwind");
+generate_extern_fn!("binary", call_nextafter, "nextafter", "nounwind");
+
+/// Invokes the [`ldexp`](https://en.cppreference.com/w/c/numeric/math/ldexp) function.
+pub fn call_ldexp<'ctx>(
+    ctx: &CodeGenContext<'ctx, '_>,
+    arg: FloatValue<'ctx>,
+    exp: IntValue<'ctx>,
+    name: Option<&str>,
+) -> FloatValue<'ctx> {
+    const FN_NAME: &str = "ldexp";
+
+    let llvm_f64 = ctx.ctx.f64_type();
+    let llvm_i32 = ctx.ctx.i32_type();
+    debug_assert_eq!(arg.get_type(), llvm_f64);
+    debug_assert_eq!(exp.get_type(), llvm_i32);
+
+    let extern_fn = ctx.module.get_function(FN_NAME).unwrap_or_else(|| {
+        let fn_type = llvm_f64.fn_type(&[llvm_f64.into(), llvm_i32.into()], false);
+        let func = ctx.module.add_function(FN_NAME, fn_type, None);
+        for attr in ["mustprogress", "nofree", "nounwind", "willreturn"] {
+            func.add_attribute(
+                AttributeLoc::Function,
+                ctx.ctx.create_enum_attribute(Attribute::get_named_enum_kind_id(attr), 0),
+            );
+        }
+
+        func
+    });
+
+    ctx.builder
+        .build_call(extern_fn, &[arg.into(), exp.into()], name.unwrap_or_default())
+        .map(CallSiteValue::try_as_basic_value)
+        .map(|v| v.map_left(BasicValueEnum::into_float_value))
+        .map(Either::unwrap_left)
+        .unwrap()
+}

nac3core/src/codegen/generator.rs

@@ -1,27 +1,20 @@
-use inkwell::{
-    context::Context,
-    targets::TargetMachine,
-    types::{BasicTypeEnum, IntType},
-    values::{BasicValueEnum, IntValue, PointerValue},
-};
-
-use nac3parser::ast::{Expr, Stmt, StrRef};
-
-use super::{CodeGenContext, bool_to_int_type, expr::*, stmt::*, values::ArraySliceValue};
 use crate::{
+    codegen::{bool_to_i1, bool_to_i8, classes::ArraySliceValue, expr::*, stmt::*, CodeGenContext},
     symbol_resolver::ValueEnum,
     toplevel::{DefinitionId, TopLevelDef},
     typecheck::typedef::{FunSignature, Type},
 };
+use inkwell::{
+    context::Context,
+    types::{BasicTypeEnum, IntType},
+    values::{BasicValueEnum, IntValue, PointerValue},
+};
+use nac3parser::ast::{Expr, Stmt, StrRef};
 
 pub trait CodeGenerator {
     /// Return the module name for the code generator.
     fn get_name(&self) -> &str;
 
-    /// Return an instance of [`IntType`] corresponding to the type of `size_t` for this instance.
-    ///
-    /// Prefer using [`CodeGenContext::get_size_type`] if [`CodeGenContext`] is available, as it is
-    /// equivalent to this function in a more concise syntax.
     fn get_size_type<'ctx>(&self, ctx: &'ctx Context) -> IntType<'ctx>;
 
     /// Generate function call and returns the function return value.
@@ -64,7 +57,6 @@ pub trait CodeGenerator {
     /// - fun: Function signature, definition ID and the substitution key.
     /// - params: Function parameters. Note that this does not include the object even if the
     ///   function is a class method.
-    ///
     /// Note that this function should check if the function is generated in another thread (due to
     /// possible race condition), see the default implementation for an example.
     fn gen_func_instance<'ctx>(
@@ -131,45 +123,11 @@ pub trait CodeGenerator {
         ctx: &mut CodeGenContext<'ctx, '_>,
         target: &Expr<Option<Type>>,
         value: ValueEnum<'ctx>,
-        value_ty: Type,
     ) -> Result<(), String>
     where
         Self: Sized,
     {
-        gen_assign(self, ctx, target, value, value_ty)
-    }
-
-    /// Generate code for an assignment expression where LHS is a `"target_list"`.
-    ///
-    /// See <https://docs.python.org/3/reference/simple_stmts.html#assignment-statements>.
-    fn gen_assign_target_list<'ctx>(
-        &mut self,
-        ctx: &mut CodeGenContext<'ctx, '_>,
-        targets: &Vec<Expr<Option<Type>>>,
-        value: ValueEnum<'ctx>,
-        value_ty: Type,
-    ) -> Result<(), String>
-    where
-        Self: Sized,
-    {
-        gen_assign_target_list(self, ctx, targets, value, value_ty)
-    }
-
-    /// Generate code for an item assignment.
-    ///
-    /// i.e., `target[key] = value`
-    fn gen_setitem<'ctx>(
-        &mut self,
-        ctx: &mut CodeGenContext<'ctx, '_>,
-        target: &Expr<Option<Type>>,
-        key: &Expr<Option<Type>>,
-        value: ValueEnum<'ctx>,
-        value_ty: Type,
-    ) -> Result<(), String>
-    where
-        Self: Sized,
-    {
-        gen_setitem(self, ctx, target, key, value, value_ty)
+        gen_assign(self, ctx, target, value)
     }
 
     /// Generate code for a while expression.
@@ -248,32 +206,22 @@ pub trait CodeGenerator {
         gen_block(self, ctx, stmts)
     }
 
-    /// Converts the value of a boolean-like value `bool_value` into an `i1`.
+    /// See [`bool_to_i1`].
     fn bool_to_i1<'ctx>(
         &self,
         ctx: &CodeGenContext<'ctx, '_>,
         bool_value: IntValue<'ctx>,
     ) -> IntValue<'ctx> {
-        self.bool_to_int_type(ctx, bool_value, ctx.ctx.bool_type())
+        bool_to_i1(&ctx.builder, bool_value)
     }
 
-    /// Converts the value of a boolean-like value `bool_value` into an `i8`.
+    /// See [`bool_to_i8`].
     fn bool_to_i8<'ctx>(
         &self,
         ctx: &CodeGenContext<'ctx, '_>,
         bool_value: IntValue<'ctx>,
     ) -> IntValue<'ctx> {
-        self.bool_to_int_type(ctx, bool_value, ctx.ctx.i8_type())
-    }
-
-    /// See [`bool_to_int_type`].
-    fn bool_to_int_type<'ctx>(
-        &self,
-        ctx: &CodeGenContext<'ctx, '_>,
-        bool_value: IntValue<'ctx>,
-        ty: IntType<'ctx>,
-    ) -> IntValue<'ctx> {
-        bool_to_int_type(&ctx.builder, bool_value, ty)
+        bool_to_i8(&ctx.builder, ctx.ctx, bool_value)
     }
 }
 
@@ -284,30 +232,26 @@ pub struct DefaultCodeGenerator {
 
 impl DefaultCodeGenerator {
     #[must_use]
-    pub fn new(name: String, size_t: IntType<'_>) -> DefaultCodeGenerator {
-        assert!(matches!(size_t.get_bit_width(), 32 | 64));
-        DefaultCodeGenerator { name, size_t: size_t.get_bit_width() }
-    }
-
-    #[must_use]
-    pub fn with_target_machine(
-        name: String,
-        ctx: &Context,
-        target_machine: &TargetMachine,
-    ) -> DefaultCodeGenerator {
-        let llvm_usize = ctx.ptr_sized_int_type(&target_machine.get_target_data(), None);
-        Self::new(name, llvm_usize)
+    pub fn new(name: String, size_t: u32) -> DefaultCodeGenerator {
+        assert!(matches!(size_t, 32 | 64));
+        DefaultCodeGenerator { name, size_t }
     }
 }
 
 impl CodeGenerator for DefaultCodeGenerator {
+    /// Returns the name for this [`CodeGenerator`].
     fn get_name(&self) -> &str {
         &self.name
     }
 
+    /// Returns an LLVM integer type representing `size_t`.
     fn get_size_type<'ctx>(&self, ctx: &'ctx Context) -> IntType<'ctx> {
         // it should be unsigned, but we don't really need unsigned and this could save us from
         // having to do a bit cast...
-        if self.size_t == 32 { ctx.i32_type() } else { ctx.i64_type() }
+        if self.size_t == 32 {
+            ctx.i32_type()
+        } else {
+            ctx.i64_type()
+        }
     }
 }

nac3core/src/codegen/irrt/cc_builtins.rs

@@ -1,131 +0,0 @@
-use inkwell::{
-    attributes::{Attribute, AttributeLoc},
-    values::{BasicValueEnum, FloatValue, IntValue},
-};
-
-use crate::codegen::{CodeGenContext, expr::infer_and_call_function};
-
-/// Generates a call to [`isinf`](https://en.cppreference.com/w/c/numeric/math/isinf) in IR. Returns
-/// an `i1` representing the result.
-pub fn call_isinf<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    v: FloatValue<'ctx>,
-    name: Option<&str>,
-) -> IntValue<'ctx> {
-    let llvm_i1 = ctx.ctx.bool_type();
-    let llvm_f64 = ctx.ctx.f64_type();
-
-    assert_eq!(v.get_type(), llvm_f64);
-
-    infer_and_call_function(ctx, "__nac3_isinf", Some(llvm_i1.into()), &[v.into()], name, None)
-        .map(BasicValueEnum::into_int_value)
-        .unwrap()
-}
-
-/// Generates a call to [`isnan`](https://en.cppreference.com/w/c/numeric/math/isnan) in IR. Returns
-/// an `i1` representing the result.
-pub fn call_isnan<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    v: FloatValue<'ctx>,
-    name: Option<&str>,
-) -> IntValue<'ctx> {
-    let llvm_i1 = ctx.ctx.bool_type();
-    let llvm_f64 = ctx.ctx.f64_type();
-
-    assert_eq!(v.get_type(), llvm_f64);
-
-    infer_and_call_function(ctx, "__nac3_isnan", Some(llvm_i1.into()), &[v.into()], name, None)
-        .map(BasicValueEnum::into_int_value)
-        .unwrap()
-}
-
-/// Macro to generate N-ary functions accepting an arbitrary number of `f64` as arguments and
-/// returning `f64`.
-///
-/// Arguments:
-///
-/// - `$fn_name:ident`: The name of the Rust function to be generated.
-/// - `$builtin_fn:ident`: The name of the builtin function to be invoked in the body of the
-///   generated function. The corresponding function in IRRT must be prefixed with `__nac3_`.
-/// - `$(,$args:ident)*`: The parameter name(s) to the IRRT function.
-macro_rules! generate_f64_nary_fn {
-    ($fn_name:ident, $builtin_fn:ident $(,$args:ident)* $(,)?) => {
-        #[doc = concat!("Generates a call to [`", stringify!($builtin_fn), "`](https://en.cppreference.com/w/c/numeric/math/", stringify!($builtin_fn), ") in IR." )]
-        pub fn $fn_name<'ctx>(
-            ctx: &CodeGenContext<'ctx, '_>,
-            $($args: FloatValue<'ctx>,)*
-            name: Option<&str>,
-        ) -> FloatValue<'ctx> {
-            const FN_NAME: &str = concat!("__nac3_", stringify!($builtin_fn));
-
-            let llvm_f64 = ctx.ctx.f64_type();
-            $(debug_assert_eq!($args.get_type(), llvm_f64);)*
-
-            infer_and_call_function(
-                ctx,
-                FN_NAME,
-                Some(llvm_f64.into()),
-                &[$($args.into()),*],
-                name,
-                Some(&|func| {
-                    func.add_attribute(
-                        AttributeLoc::Function,
-                        ctx.ctx.create_enum_attribute(Attribute::get_named_enum_kind_id("nounwind"), 0)
-                    );
-                }),
-            )
-            .map(BasicValueEnum::into_float_value)
-            .unwrap()
-        }
-    };
-}
-
-generate_f64_nary_fn!(call_tan, tan, arg);
-generate_f64_nary_fn!(call_asin, asin, arg);
-generate_f64_nary_fn!(call_acos, acos, arg);
-generate_f64_nary_fn!(call_atan, atan, arg);
-generate_f64_nary_fn!(call_sinh, sinh, arg);
-generate_f64_nary_fn!(call_cosh, cosh, arg);
-generate_f64_nary_fn!(call_tanh, tanh, arg);
-generate_f64_nary_fn!(call_asinh, asinh, arg);
-generate_f64_nary_fn!(call_acosh, acosh, arg);
-generate_f64_nary_fn!(call_atanh, atanh, arg);
-generate_f64_nary_fn!(call_expm1, expm1, arg);
-generate_f64_nary_fn!(call_cbrt, cbrt, arg);
-generate_f64_nary_fn!(call_erf, erf, arg);
-generate_f64_nary_fn!(call_erfc, erfc, arg);
-generate_f64_nary_fn!(call_gamma, gamma, z);
-generate_f64_nary_fn!(call_atan2, atan2, y, x);
-generate_f64_nary_fn!(call_hypot, hypot, x, y);
-generate_f64_nary_fn!(call_nextafter, nextafter, from, to);
-
-/// Invokes the [`ldexp`](https://en.cppreference.com/w/c/numeric/math/ldexp) function.
-pub fn call_ldexp<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    arg: FloatValue<'ctx>,
-    exp: IntValue<'ctx>,
-    name: Option<&str>,
-) -> FloatValue<'ctx> {
-    const FN_NAME: &str = "__nac3_ldexp";
-
-    let llvm_f64 = ctx.ctx.f64_type();
-    let llvm_i32 = ctx.ctx.i32_type();
-    debug_assert_eq!(arg.get_type(), llvm_f64);
-    debug_assert_eq!(exp.get_type(), llvm_i32);
-
-    infer_and_call_function(
-        ctx,
-        FN_NAME,
-        Some(llvm_f64.into()),
-        &[arg.into(), exp.into()],
-        name,
-        Some(&|func| {
-            func.add_attribute(
-                AttributeLoc::Function,
-                ctx.ctx.create_enum_attribute(Attribute::get_named_enum_kind_id("nounwind"), 0),
-            );
-        }),
-    )
-    .map(BasicValueEnum::into_float_value)
-    .unwrap()
-}

nac3core/src/codegen/irrt/list.rs

@@ -1,166 +0,0 @@
-use inkwell::{
-    AddressSpace, IntPredicate,
-    types::BasicTypeEnum,
-    values::{BasicValueEnum, IntValue},
-};
-
-use super::calculate_len_for_slice_range;
-use crate::codegen::{
-    CodeGenContext, CodeGenerator,
-    expr::infer_and_call_function,
-    macros::codegen_unreachable,
-    stmt::gen_if_callback,
-    values::{ArrayLikeValue, ListValue},
-};
-
-/// This function handles 'end' **inclusively**.
-/// Order of tuples `assign_idx` and `value_idx` is ('start', 'end', 'step').
-/// Negative index should be handled before entering this function
-pub fn list_slice_assignment<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &mut G,
-    ctx: &mut CodeGenContext<'ctx, '_>,
-    ty: BasicTypeEnum<'ctx>,
-    dest_arr: ListValue<'ctx>,
-    dest_idx: (IntValue<'ctx>, IntValue<'ctx>, IntValue<'ctx>),
-    src_arr: ListValue<'ctx>,
-    src_idx: (IntValue<'ctx>, IntValue<'ctx>, IntValue<'ctx>),
-) {
-    let llvm_usize = ctx.get_size_type();
-    let llvm_pi8 = ctx.ctx.i8_type().ptr_type(AddressSpace::default());
-    let llvm_i32 = ctx.ctx.i32_type();
-
-    assert_eq!(dest_idx.0.get_type(), llvm_i32);
-    assert_eq!(dest_idx.1.get_type(), llvm_i32);
-    assert_eq!(dest_idx.2.get_type(), llvm_i32);
-    assert_eq!(src_idx.0.get_type(), llvm_i32);
-    assert_eq!(src_idx.1.get_type(), llvm_i32);
-    assert_eq!(src_idx.2.get_type(), llvm_i32);
-
-    let (fun_symbol, elem_ptr_type) = ("__nac3_list_slice_assign_var_size", llvm_pi8);
-
-    let zero = llvm_i32.const_zero();
-    let one = llvm_i32.const_int(1, false);
-    let dest_arr_ptr = dest_arr.data().base_ptr(ctx, generator);
-    let dest_arr_ptr =
-        ctx.builder.build_pointer_cast(dest_arr_ptr, elem_ptr_type, "dest_arr_ptr_cast").unwrap();
-    let dest_len = dest_arr.load_size(ctx, Some("dest.len"));
-    let dest_len =
-        ctx.builder.build_int_truncate_or_bit_cast(dest_len, llvm_i32, "srclen32").unwrap();
-    let src_arr_ptr = src_arr.data().base_ptr(ctx, generator);
-    let src_arr_ptr =
-        ctx.builder.build_pointer_cast(src_arr_ptr, elem_ptr_type, "src_arr_ptr_cast").unwrap();
-    let src_len = src_arr.load_size(ctx, Some("src.len"));
-    let src_len =
-        ctx.builder.build_int_truncate_or_bit_cast(src_len, llvm_i32, "srclen32").unwrap();
-
-    // index in bound and positive should be done
-    // assert if dest.step == 1 then len(src) <= len(dest) else len(src) == len(dest), and
-    // throw exception if not satisfied
-    let src_end = ctx
-        .builder
-        .build_select(
-            ctx.builder.build_int_compare(IntPredicate::SLT, src_idx.2, zero, "is_neg").unwrap(),
-            ctx.builder.build_int_sub(src_idx.1, one, "e_min_one").unwrap(),
-            ctx.builder.build_int_add(src_idx.1, one, "e_add_one").unwrap(),
-            "final_e",
-        )
-        .map(BasicValueEnum::into_int_value)
-        .unwrap();
-    let dest_end = ctx
-        .builder
-        .build_select(
-            ctx.builder.build_int_compare(IntPredicate::SLT, dest_idx.2, zero, "is_neg").unwrap(),
-            ctx.builder.build_int_sub(dest_idx.1, one, "e_min_one").unwrap(),
-            ctx.builder.build_int_add(dest_idx.1, one, "e_add_one").unwrap(),
-            "final_e",
-        )
-        .map(BasicValueEnum::into_int_value)
-        .unwrap();
-    let src_slice_len =
-        calculate_len_for_slice_range(generator, ctx, src_idx.0, src_end, src_idx.2);
-    let dest_slice_len =
-        calculate_len_for_slice_range(generator, ctx, dest_idx.0, dest_end, dest_idx.2);
-    let src_eq_dest = ctx
-        .builder
-        .build_int_compare(IntPredicate::EQ, src_slice_len, dest_slice_len, "slice_src_eq_dest")
-        .unwrap();
-    let src_slt_dest = ctx
-        .builder
-        .build_int_compare(IntPredicate::SLT, src_slice_len, dest_slice_len, "slice_src_slt_dest")
-        .unwrap();
-    let dest_step_eq_one = ctx
-        .builder
-        .build_int_compare(
-            IntPredicate::EQ,
-            dest_idx.2,
-            dest_idx.2.get_type().const_int(1, false),
-            "slice_dest_step_eq_one",
-        )
-        .unwrap();
-    let cond_1 = ctx.builder.build_and(dest_step_eq_one, src_slt_dest, "slice_cond_1").unwrap();
-    let cond = ctx.builder.build_or(src_eq_dest, cond_1, "slice_cond").unwrap();
-    ctx.make_assert(
-        generator,
-        cond,
-        "0:ValueError",
-        "attempt to assign sequence of size {0} to slice of size {1} with step size {2}",
-        [Some(src_slice_len), Some(dest_slice_len), Some(dest_idx.2)],
-        ctx.current_loc,
-    );
-
-    let new_len = {
-        let args = [
-            dest_idx.0.into(),   // dest start idx
-            dest_idx.1.into(),   // dest end idx
-            dest_idx.2.into(),   // dest step
-            dest_arr_ptr.into(), // dest arr ptr
-            dest_len.into(),     // dest arr len
-            src_idx.0.into(),    // src start idx
-            src_idx.1.into(),    // src end idx
-            src_idx.2.into(),    // src step
-            src_arr_ptr.into(),  // src arr ptr
-            src_len.into(),      // src arr len
-            {
-                let s = match ty {
-                    BasicTypeEnum::FloatType(t) => t.size_of(),
-                    BasicTypeEnum::IntType(t) => t.size_of(),
-                    BasicTypeEnum::PointerType(t) => t.size_of(),
-                    BasicTypeEnum::StructType(t) => t.size_of().unwrap(),
-                    _ => codegen_unreachable!(ctx),
-                };
-                ctx.builder.build_int_truncate_or_bit_cast(s, llvm_i32, "size").unwrap()
-            }
-            .into(),
-        ];
-        infer_and_call_function(
-            ctx,
-            fun_symbol,
-            Some(llvm_i32.into()),
-            &args,
-            Some("slice_assign"),
-            None,
-        )
-        .map(BasicValueEnum::into_int_value)
-        .unwrap()
-    };
-
-    // update length
-    gen_if_callback(
-        generator,
-        ctx,
-        |_, ctx| {
-            Ok(ctx
-                .builder
-                .build_int_compare(IntPredicate::NE, new_len, dest_len, "need_update")
-                .unwrap())
-        },
-        |_, ctx| {
-            let new_len =
-                ctx.builder.build_int_z_extend_or_bit_cast(new_len, llvm_usize, "new_len").unwrap();
-            dest_arr.store_size(ctx, new_len);
-            Ok(())
-        },
-        |_, _| Ok(()),
-    )
-    .unwrap();
-}

nac3core/src/codegen/irrt/math.rs

@@ -1,89 +0,0 @@
-use inkwell::{
-    IntPredicate,
-    values::{BasicValueEnum, FloatValue, IntValue},
-};
-
-use crate::codegen::{
-    expr::infer_and_call_function,
-    macros::codegen_unreachable,
-    {CodeGenContext, CodeGenerator},
-};
-
-// repeated squaring method adapted from GNU Scientific Library:
-// https://git.savannah.gnu.org/cgit/gsl.git/tree/sys/pow_int.c
-pub fn integer_power<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &mut G,
-    ctx: &mut CodeGenContext<'ctx, '_>,
-    base: IntValue<'ctx>,
-    exp: IntValue<'ctx>,
-    signed: bool,
-) -> IntValue<'ctx> {
-    let base_type = base.get_type();
-
-    let symbol = match (base_type.get_bit_width(), exp.get_type().get_bit_width(), signed) {
-        (32, 32, true) => "__nac3_int_exp_int32_t",
-        (64, 64, true) => "__nac3_int_exp_int64_t",
-        (32, 32, false) => "__nac3_int_exp_uint32_t",
-        (64, 64, false) => "__nac3_int_exp_uint64_t",
-        _ => codegen_unreachable!(ctx),
-    };
-
-    // throw exception when exp < 0
-    let ge_zero = ctx
-        .builder
-        .build_int_compare(
-            IntPredicate::SGE,
-            exp,
-            exp.get_type().const_zero(),
-            "assert_int_pow_ge_0",
-        )
-        .unwrap();
-    ctx.make_assert(
-        generator,
-        ge_zero,
-        "0:ValueError",
-        "integer power must be positive or zero",
-        [None, None, None],
-        ctx.current_loc,
-    );
-
-    infer_and_call_function(
-        ctx,
-        symbol,
-        Some(base_type.into()),
-        &[base.into(), exp.into()],
-        Some("call_int_pow"),
-        None,
-    )
-    .map(BasicValueEnum::into_int_value)
-    .unwrap()
-}
-
-/// Generates a call to `gammaln` in IR. Returns an `f64` representing the result.
-pub fn call_gammaln<'ctx>(ctx: &CodeGenContext<'ctx, '_>, v: FloatValue<'ctx>) -> FloatValue<'ctx> {
-    let llvm_f64 = ctx.ctx.f64_type();
-
-    assert_eq!(v.get_type(), llvm_f64);
-
-    infer_and_call_function(
-        ctx,
-        "__nac3_gammaln",
-        Some(llvm_f64.into()),
-        &[v.into()],
-        Some("gammaln"),
-        None,
-    )
-    .map(BasicValueEnum::into_float_value)
-    .unwrap()
-}
-
-/// Generates a call to `j0` in IR. Returns an `f64` representing the result.
-pub fn call_j0<'ctx>(ctx: &CodeGenContext<'ctx, '_>, v: FloatValue<'ctx>) -> FloatValue<'ctx> {
-    let llvm_f64 = ctx.ctx.f64_type();
-
-    assert_eq!(v.get_type(), llvm_f64);
-
-    infer_and_call_function(ctx, "__nac3_j0", Some(llvm_f64.into()), &[v.into()], Some("j0"), None)
-        .map(BasicValueEnum::into_float_value)
-        .unwrap()
-}

nac3core/src/codegen/irrt/mod.rs

@@ -1,33 +1,30 @@
+use crate::{typecheck::typedef::Type, util::SizeVariant};
+
+mod test;
+
+use super::{
+    classes::{
+        ArrayLikeIndexer, ArrayLikeValue, ArraySliceValue, ListValue, NDArrayValue, NpArrayType,
+        NpArrayValue, TypedArrayLikeAdapter, UntypedArrayLikeAccessor,
+    },
+    llvm_intrinsics, CodeGenContext, CodeGenerator,
+};
+use crate::codegen::classes::TypedArrayLikeAccessor;
+use crate::codegen::stmt::gen_for_callback_incrementing;
 use inkwell::{
-    IntPredicate,
     attributes::{Attribute, AttributeLoc},
     context::Context,
     memory_buffer::MemoryBuffer,
     module::Module,
-    values::{BasicValue, BasicValueEnum, IntValue},
+    types::{BasicType, BasicTypeEnum, FunctionType, IntType, PointerType},
+    values::{BasicValueEnum, CallSiteValue, FloatValue, FunctionValue, IntValue},
+    AddressSpace, IntPredicate,
 };
-
+use itertools::Either;
 use nac3parser::ast::Expr;
 
-use super::{CodeGenContext, CodeGenerator};
-use crate::{symbol_resolver::SymbolResolver, typecheck::typedef::Type};
-pub use cc_builtins::*;
-pub use list::*;
-pub use math::*;
-pub use range::*;
-pub use slice::*;
-pub use string::*;
-
-mod cc_builtins;
-mod list;
-mod math;
-pub mod ndarray;
-mod range;
-mod slice;
-mod string;
-
 #[must_use]
-pub fn load_irrt<'ctx>(ctx: &'ctx Context, symbol_resolver: &dyn SymbolResolver) -> Module<'ctx> {
+pub fn load_irrt(ctx: &Context) -> Module {
     let bitcode_buf = MemoryBuffer::create_from_memory_range(
         include_bytes!(concat!(env!("OUT_DIR"), "/irrt.bc")),
         "irrt_bitcode_buffer",
@@ -43,43 +40,89 @@ pub fn load_irrt<'ctx>(ctx: &'ctx Context, symbol_resolver: &dyn SymbolResolver)
         let function = irrt_mod.get_function(symbol).unwrap();
         function.add_attribute(AttributeLoc::Function, ctx.create_enum_attribute(inline_attr, 0));
     }
-
-    // Initialize all global `EXN_*` exception IDs in IRRT with the [`SymbolResolver`].
-    let exn_id_type = ctx.i32_type();
-    let errors = &[
-        ("EXN_INDEX_ERROR", "0:IndexError"),
-        ("EXN_VALUE_ERROR", "0:ValueError"),
-        ("EXN_ASSERTION_ERROR", "0:AssertionError"),
-        ("EXN_TYPE_ERROR", "0:TypeError"),
-    ];
-    for (irrt_name, symbol_name) in errors {
-        let exn_id = symbol_resolver.get_string_id(symbol_name);
-        let exn_id = exn_id_type.const_int(exn_id as u64, false).as_basic_value_enum();
-
-        let global = irrt_mod.get_global(irrt_name).unwrap_or_else(|| {
-            panic!("Exception symbol name '{irrt_name}' should exist in the IRRT LLVM module")
-        });
-        global.set_initializer(&exn_id);
-    }
-
     irrt_mod
 }
 
-/// Returns the name of a function which contains variants for 32-bit and 64-bit `size_t`.
-///
-/// - When [`TypeContext::size_type`] is 32-bits, the function name is `fn_name}`.
-/// - When [`TypeContext::size_type`] is 64-bits, the function name is `{fn_name}64`.
-#[must_use]
-pub fn get_usize_dependent_function_name(ctx: &CodeGenContext<'_, '_>, name: &str) -> String {
-    let mut name = name.to_owned();
-    match ctx.get_size_type().get_bit_width() {
-        32 => {}
-        64 => name.push_str("64"),
-        bit_width => {
-            panic!("Unsupported int type bit width {bit_width}, must be either 32-bits or 64-bits")
-        }
-    }
-    name
+// repeated squaring method adapted from GNU Scientific Library:
+// https://git.savannah.gnu.org/cgit/gsl.git/tree/sys/pow_int.c
+pub fn integer_power<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    base: IntValue<'ctx>,
+    exp: IntValue<'ctx>,
+    signed: bool,
+) -> IntValue<'ctx> {
+    let symbol = match (base.get_type().get_bit_width(), exp.get_type().get_bit_width(), signed) {
+        (32, 32, true) => "__nac3_int_exp_int32_t",
+        (64, 64, true) => "__nac3_int_exp_int64_t",
+        (32, 32, false) => "__nac3_int_exp_uint32_t",
+        (64, 64, false) => "__nac3_int_exp_uint64_t",
+        _ => unreachable!(),
+    };
+    let base_type = base.get_type();
+    let pow_fun = ctx.module.get_function(symbol).unwrap_or_else(|| {
+        let fn_type = base_type.fn_type(&[base_type.into(), base_type.into()], false);
+        ctx.module.add_function(symbol, fn_type, None)
+    });
+    // throw exception when exp < 0
+    let ge_zero = ctx
+        .builder
+        .build_int_compare(
+            IntPredicate::SGE,
+            exp,
+            exp.get_type().const_zero(),
+            "assert_int_pow_ge_0",
+        )
+        .unwrap();
+    ctx.make_assert(
+        generator,
+        ge_zero,
+        "0:ValueError",
+        "integer power must be positive or zero",
+        [None, None, None],
+        ctx.current_loc,
+    );
+    ctx.builder
+        .build_call(pow_fun, &[base.into(), exp.into()], "call_int_pow")
+        .map(CallSiteValue::try_as_basic_value)
+        .map(|v| v.map_left(BasicValueEnum::into_int_value))
+        .map(Either::unwrap_left)
+        .unwrap()
+}
+
+pub fn calculate_len_for_slice_range<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    start: IntValue<'ctx>,
+    end: IntValue<'ctx>,
+    step: IntValue<'ctx>,
+) -> IntValue<'ctx> {
+    const SYMBOL: &str = "__nac3_range_slice_len";
+    let len_func = ctx.module.get_function(SYMBOL).unwrap_or_else(|| {
+        let i32_t = ctx.ctx.i32_type();
+        let fn_t = i32_t.fn_type(&[i32_t.into(), i32_t.into(), i32_t.into()], false);
+        ctx.module.add_function(SYMBOL, fn_t, None)
+    });
+
+    // assert step != 0, throw exception if not
+    let not_zero = ctx
+        .builder
+        .build_int_compare(IntPredicate::NE, step, step.get_type().const_zero(), "range_step_ne")
+        .unwrap();
+    ctx.make_assert(
+        generator,
+        not_zero,
+        "0:ValueError",
+        "step must not be zero",
+        [None, None, None],
+        ctx.current_loc,
+    );
+    ctx.builder
+        .build_call(len_func, &[start.into(), end.into(), step.into()], "calc_len")
+        .map(CallSiteValue::try_as_basic_value)
+        .map(|v| v.map_left(BasicValueEnum::into_int_value))
+        .map(Either::unwrap_left)
+        .unwrap()
 }
 
 /// NOTE: the output value of the end index of this function should be compared ***inclusively***,
@@ -130,11 +173,10 @@ pub fn handle_slice_indices<'ctx, G: CodeGenerator>(
     generator: &mut G,
     length: IntValue<'ctx>,
 ) -> Result<Option<(IntValue<'ctx>, IntValue<'ctx>, IntValue<'ctx>)>, String> {
-    let llvm_i32 = ctx.ctx.i32_type();
-
-    let zero = llvm_i32.const_zero();
-    let one = llvm_i32.const_int(1, false);
-    let length = ctx.builder.build_int_truncate_or_bit_cast(length, llvm_i32, "leni32").unwrap();
+    let int32 = ctx.ctx.i32_type();
+    let zero = int32.const_zero();
+    let one = int32.const_int(1, false);
+    let length = ctx.builder.build_int_truncate_or_bit_cast(length, int32, "leni32").unwrap();
     Ok(Some(match (start, end, step) {
         (s, e, None) => (
             if let Some(s) = s.as_ref() {
@@ -143,7 +185,7 @@ pub fn handle_slice_indices<'ctx, G: CodeGenerator>(
                     None => return Ok(None),
                 }
             } else {
-                llvm_i32.const_zero()
+                int32.const_zero()
             },
             {
                 let e = if let Some(s) = e.as_ref() {
@@ -248,3 +290,702 @@ pub fn handle_slice_indices<'ctx, G: CodeGenerator>(
         }
     }))
 }
+
+/// this function allows index out of range, since python
+/// allows index out of range in slice (`a = [1,2,3]; a[1:10] == [2,3]`).
+pub fn handle_slice_index_bound<'ctx, G: CodeGenerator>(
+    i: &Expr<Option<Type>>,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    generator: &mut G,
+    length: IntValue<'ctx>,
+) -> Result<Option<IntValue<'ctx>>, String> {
+    const SYMBOL: &str = "__nac3_slice_index_bound";
+    let func = ctx.module.get_function(SYMBOL).unwrap_or_else(|| {
+        let i32_t = ctx.ctx.i32_type();
+        let fn_t = i32_t.fn_type(&[i32_t.into(), i32_t.into()], false);
+        ctx.module.add_function(SYMBOL, fn_t, None)
+    });
+
+    let i = if let Some(v) = generator.gen_expr(ctx, i)? {
+        v.to_basic_value_enum(ctx, generator, i.custom.unwrap())?
+    } else {
+        return Ok(None);
+    };
+    Ok(Some(
+        ctx.builder
+            .build_call(func, &[i.into(), length.into()], "bounded_ind")
+            .map(CallSiteValue::try_as_basic_value)
+            .map(|v| v.map_left(BasicValueEnum::into_int_value))
+            .map(Either::unwrap_left)
+            .unwrap(),
+    ))
+}
+
+/// This function handles 'end' **inclusively**.
+/// Order of tuples `assign_idx` and `value_idx` is ('start', 'end', 'step').
+/// Negative index should be handled before entering this function
+pub fn list_slice_assignment<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    ty: BasicTypeEnum<'ctx>,
+    dest_arr: ListValue<'ctx>,
+    dest_idx: (IntValue<'ctx>, IntValue<'ctx>, IntValue<'ctx>),
+    src_arr: ListValue<'ctx>,
+    src_idx: (IntValue<'ctx>, IntValue<'ctx>, IntValue<'ctx>),
+) {
+    let size_ty = generator.get_size_type(ctx.ctx);
+    let int8_ptr = ctx.ctx.i8_type().ptr_type(AddressSpace::default());
+    let int32 = ctx.ctx.i32_type();
+    let (fun_symbol, elem_ptr_type) = ("__nac3_list_slice_assign_var_size", int8_ptr);
+    let slice_assign_fun = {
+        let ty_vec = vec![
+            int32.into(),         // dest start idx
+            int32.into(),         // dest end idx
+            int32.into(),         // dest step
+            elem_ptr_type.into(), // dest arr ptr
+            int32.into(),         // dest arr len
+            int32.into(),         // src start idx
+            int32.into(),         // src end idx
+            int32.into(),         // src step
+            elem_ptr_type.into(), // src arr ptr
+            int32.into(),         // src arr len
+            int32.into(),         // size
+        ];
+        ctx.module.get_function(fun_symbol).unwrap_or_else(|| {
+            let fn_t = int32.fn_type(ty_vec.as_slice(), false);
+            ctx.module.add_function(fun_symbol, fn_t, None)
+        })
+    };
+
+    let zero = int32.const_zero();
+    let one = int32.const_int(1, false);
+    let dest_arr_ptr = dest_arr.data().base_ptr(ctx, generator);
+    let dest_arr_ptr =
+        ctx.builder.build_pointer_cast(dest_arr_ptr, elem_ptr_type, "dest_arr_ptr_cast").unwrap();
+    let dest_len = dest_arr.load_size(ctx, Some("dest.len"));
+    let dest_len = ctx.builder.build_int_truncate_or_bit_cast(dest_len, int32, "srclen32").unwrap();
+    let src_arr_ptr = src_arr.data().base_ptr(ctx, generator);
+    let src_arr_ptr =
+        ctx.builder.build_pointer_cast(src_arr_ptr, elem_ptr_type, "src_arr_ptr_cast").unwrap();
+    let src_len = src_arr.load_size(ctx, Some("src.len"));
+    let src_len = ctx.builder.build_int_truncate_or_bit_cast(src_len, int32, "srclen32").unwrap();
+
+    // index in bound and positive should be done
+    // assert if dest.step == 1 then len(src) <= len(dest) else len(src) == len(dest), and
+    // throw exception if not satisfied
+    let src_end = ctx
+        .builder
+        .build_select(
+            ctx.builder.build_int_compare(IntPredicate::SLT, src_idx.2, zero, "is_neg").unwrap(),
+            ctx.builder.build_int_sub(src_idx.1, one, "e_min_one").unwrap(),
+            ctx.builder.build_int_add(src_idx.1, one, "e_add_one").unwrap(),
+            "final_e",
+        )
+        .map(BasicValueEnum::into_int_value)
+        .unwrap();
+    let dest_end = ctx
+        .builder
+        .build_select(
+            ctx.builder.build_int_compare(IntPredicate::SLT, dest_idx.2, zero, "is_neg").unwrap(),
+            ctx.builder.build_int_sub(dest_idx.1, one, "e_min_one").unwrap(),
+            ctx.builder.build_int_add(dest_idx.1, one, "e_add_one").unwrap(),
+            "final_e",
+        )
+        .map(BasicValueEnum::into_int_value)
+        .unwrap();
+    let src_slice_len =
+        calculate_len_for_slice_range(generator, ctx, src_idx.0, src_end, src_idx.2);
+    let dest_slice_len =
+        calculate_len_for_slice_range(generator, ctx, dest_idx.0, dest_end, dest_idx.2);
+    let src_eq_dest = ctx
+        .builder
+        .build_int_compare(IntPredicate::EQ, src_slice_len, dest_slice_len, "slice_src_eq_dest")
+        .unwrap();
+    let src_slt_dest = ctx
+        .builder
+        .build_int_compare(IntPredicate::SLT, src_slice_len, dest_slice_len, "slice_src_slt_dest")
+        .unwrap();
+    let dest_step_eq_one = ctx
+        .builder
+        .build_int_compare(
+            IntPredicate::EQ,
+            dest_idx.2,
+            dest_idx.2.get_type().const_int(1, false),
+            "slice_dest_step_eq_one",
+        )
+        .unwrap();
+    let cond_1 = ctx.builder.build_and(dest_step_eq_one, src_slt_dest, "slice_cond_1").unwrap();
+    let cond = ctx.builder.build_or(src_eq_dest, cond_1, "slice_cond").unwrap();
+    ctx.make_assert(
+        generator,
+        cond,
+        "0:ValueError",
+        "attempt to assign sequence of size {0} to slice of size {1} with step size {2}",
+        [Some(src_slice_len), Some(dest_slice_len), Some(dest_idx.2)],
+        ctx.current_loc,
+    );
+
+    let new_len = {
+        let args = vec![
+            dest_idx.0.into(),   // dest start idx
+            dest_idx.1.into(),   // dest end idx
+            dest_idx.2.into(),   // dest step
+            dest_arr_ptr.into(), // dest arr ptr
+            dest_len.into(),     // dest arr len
+            src_idx.0.into(),    // src start idx
+            src_idx.1.into(),    // src end idx
+            src_idx.2.into(),    // src step
+            src_arr_ptr.into(),  // src arr ptr
+            src_len.into(),      // src arr len
+            {
+                let s = match ty {
+                    BasicTypeEnum::FloatType(t) => t.size_of(),
+                    BasicTypeEnum::IntType(t) => t.size_of(),
+                    BasicTypeEnum::PointerType(t) => t.size_of(),
+                    BasicTypeEnum::StructType(t) => t.size_of().unwrap(),
+                    _ => unreachable!(),
+                };
+                ctx.builder.build_int_truncate_or_bit_cast(s, int32, "size").unwrap()
+            }
+            .into(),
+        ];
+        ctx.builder
+            .build_call(slice_assign_fun, args.as_slice(), "slice_assign")
+            .map(CallSiteValue::try_as_basic_value)
+            .map(|v| v.map_left(BasicValueEnum::into_int_value))
+            .map(Either::unwrap_left)
+            .unwrap()
+    };
+    // update length
+    let need_update =
+        ctx.builder.build_int_compare(IntPredicate::NE, new_len, dest_len, "need_update").unwrap();
+    let current = ctx.builder.get_insert_block().unwrap().get_parent().unwrap();
+    let update_bb = ctx.ctx.append_basic_block(current, "update");
+    let cont_bb = ctx.ctx.append_basic_block(current, "cont");
+    ctx.builder.build_conditional_branch(need_update, update_bb, cont_bb).unwrap();
+    ctx.builder.position_at_end(update_bb);
+    let new_len = ctx.builder.build_int_z_extend_or_bit_cast(new_len, size_ty, "new_len").unwrap();
+    dest_arr.store_size(ctx, generator, new_len);
+    ctx.builder.build_unconditional_branch(cont_bb).unwrap();
+    ctx.builder.position_at_end(cont_bb);
+}
+
+/// Generates a call to `isinf` in IR. Returns an `i1` representing the result.
+pub fn call_isinf<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &CodeGenContext<'ctx, '_>,
+    v: FloatValue<'ctx>,
+) -> IntValue<'ctx> {
+    let intrinsic_fn = ctx.module.get_function("__nac3_isinf").unwrap_or_else(|| {
+        let fn_type = ctx.ctx.i32_type().fn_type(&[ctx.ctx.f64_type().into()], false);
+        ctx.module.add_function("__nac3_isinf", fn_type, None)
+    });
+
+    let ret = ctx
+        .builder
+        .build_call(intrinsic_fn, &[v.into()], "isinf")
+        .map(CallSiteValue::try_as_basic_value)
+        .map(|v| v.map_left(BasicValueEnum::into_int_value))
+        .map(Either::unwrap_left)
+        .unwrap();
+
+    generator.bool_to_i1(ctx, ret)
+}
+
+/// Generates a call to `isnan` in IR. Returns an `i1` representing the result.
+pub fn call_isnan<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &CodeGenContext<'ctx, '_>,
+    v: FloatValue<'ctx>,
+) -> IntValue<'ctx> {
+    let intrinsic_fn = ctx.module.get_function("__nac3_isnan").unwrap_or_else(|| {
+        let fn_type = ctx.ctx.i32_type().fn_type(&[ctx.ctx.f64_type().into()], false);
+        ctx.module.add_function("__nac3_isnan", fn_type, None)
+    });
+
+    let ret = ctx
+        .builder
+        .build_call(intrinsic_fn, &[v.into()], "isnan")
+        .map(CallSiteValue::try_as_basic_value)
+        .map(|v| v.map_left(BasicValueEnum::into_int_value))
+        .map(Either::unwrap_left)
+        .unwrap();
+
+    generator.bool_to_i1(ctx, ret)
+}
+
+/// Generates a call to `gamma` in IR. Returns an `f64` representing the result.
+pub fn call_gamma<'ctx>(ctx: &CodeGenContext<'ctx, '_>, v: FloatValue<'ctx>) -> FloatValue<'ctx> {
+    let llvm_f64 = ctx.ctx.f64_type();
+
+    let intrinsic_fn = ctx.module.get_function("__nac3_gamma").unwrap_or_else(|| {
+        let fn_type = llvm_f64.fn_type(&[llvm_f64.into()], false);
+        ctx.module.add_function("__nac3_gamma", fn_type, None)
+    });
+
+    ctx.builder
+        .build_call(intrinsic_fn, &[v.into()], "gamma")
+        .map(CallSiteValue::try_as_basic_value)
+        .map(|v| v.map_left(BasicValueEnum::into_float_value))
+        .map(Either::unwrap_left)
+        .unwrap()
+}
+
+/// Generates a call to `gammaln` in IR. Returns an `f64` representing the result.
+pub fn call_gammaln<'ctx>(ctx: &CodeGenContext<'ctx, '_>, v: FloatValue<'ctx>) -> FloatValue<'ctx> {
+    let llvm_f64 = ctx.ctx.f64_type();
+
+    let intrinsic_fn = ctx.module.get_function("__nac3_gammaln").unwrap_or_else(|| {
+        let fn_type = llvm_f64.fn_type(&[llvm_f64.into()], false);
+        ctx.module.add_function("__nac3_gammaln", fn_type, None)
+    });
+
+    ctx.builder
+        .build_call(intrinsic_fn, &[v.into()], "gammaln")
+        .map(CallSiteValue::try_as_basic_value)
+        .map(|v| v.map_left(BasicValueEnum::into_float_value))
+        .map(Either::unwrap_left)
+        .unwrap()
+}
+
+/// Generates a call to `j0` in IR. Returns an `f64` representing the result.
+pub fn call_j0<'ctx>(ctx: &CodeGenContext<'ctx, '_>, v: FloatValue<'ctx>) -> FloatValue<'ctx> {
+    let llvm_f64 = ctx.ctx.f64_type();
+
+    let intrinsic_fn = ctx.module.get_function("__nac3_j0").unwrap_or_else(|| {
+        let fn_type = llvm_f64.fn_type(&[llvm_f64.into()], false);
+        ctx.module.add_function("__nac3_j0", fn_type, None)
+    });
+
+    ctx.builder
+        .build_call(intrinsic_fn, &[v.into()], "j0")
+        .map(CallSiteValue::try_as_basic_value)
+        .map(|v| v.map_left(BasicValueEnum::into_float_value))
+        .map(Either::unwrap_left)
+        .unwrap()
+}
+
+/// Generates a call to `__nac3_ndarray_calc_size`. Returns an [`IntValue`] representing the
+/// calculated total size.
+///
+/// * `dims` - An [`ArrayLikeIndexer`] containing the size of each dimension.
+/// * `range` - The dimension index to begin and end (exclusively) calculating the dimensions for,
+/// or [`None`] if starting from the first dimension and ending at the last dimension respectively.
+pub fn call_ndarray_calc_size<'ctx, G, Dims>(
+    generator: &G,
+    ctx: &CodeGenContext<'ctx, '_>,
+    dims: &Dims,
+    (begin, end): (Option<IntValue<'ctx>>, Option<IntValue<'ctx>>),
+) -> IntValue<'ctx>
+where
+    G: CodeGenerator + ?Sized,
+    Dims: ArrayLikeIndexer<'ctx>,
+{
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+    let llvm_pusize = llvm_usize.ptr_type(AddressSpace::default());
+
+    let ndarray_calc_size_fn_name = match llvm_usize.get_bit_width() {
+        32 => "__nac3_ndarray_calc_size",
+        64 => "__nac3_ndarray_calc_size64",
+        bw => unreachable!("Unsupported size type bit width: {}", bw),
+    };
+    let ndarray_calc_size_fn_t = llvm_usize.fn_type(
+        &[llvm_pusize.into(), llvm_usize.into(), llvm_usize.into(), llvm_usize.into()],
+        false,
+    );
+    let ndarray_calc_size_fn =
+        ctx.module.get_function(ndarray_calc_size_fn_name).unwrap_or_else(|| {
+            ctx.module.add_function(ndarray_calc_size_fn_name, ndarray_calc_size_fn_t, None)
+        });
+
+    let begin = begin.unwrap_or_else(|| llvm_usize.const_zero());
+    let end = end.unwrap_or_else(|| dims.size(ctx, generator));
+    ctx.builder
+        .build_call(
+            ndarray_calc_size_fn,
+            &[
+                dims.base_ptr(ctx, generator).into(),
+                dims.size(ctx, generator).into(),
+                begin.into(),
+                end.into(),
+            ],
+            "",
+        )
+        .map(CallSiteValue::try_as_basic_value)
+        .map(|v| v.map_left(BasicValueEnum::into_int_value))
+        .map(Either::unwrap_left)
+        .unwrap()
+}
+
+/// Generates a call to `__nac3_ndarray_calc_nd_indices`. Returns a [`TypeArrayLikeAdpater`]
+/// containing `i32` indices of the flattened index.
+///
+/// * `index` - The index to compute the multidimensional index for.
+/// * `ndarray` - LLVM pointer to the `NDArray`. This value must be the LLVM representation of an
+/// `NDArray`.
+pub fn call_ndarray_calc_nd_indices<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    index: IntValue<'ctx>,
+    ndarray: NDArrayValue<'ctx>,
+) -> TypedArrayLikeAdapter<'ctx, IntValue<'ctx>> {
+    let llvm_void = ctx.ctx.void_type();
+    let llvm_i32 = ctx.ctx.i32_type();
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+    let llvm_pi32 = llvm_i32.ptr_type(AddressSpace::default());
+    let llvm_pusize = llvm_usize.ptr_type(AddressSpace::default());
+
+    let ndarray_calc_nd_indices_fn_name = match llvm_usize.get_bit_width() {
+        32 => "__nac3_ndarray_calc_nd_indices",
+        64 => "__nac3_ndarray_calc_nd_indices64",
+        bw => unreachable!("Unsupported size type bit width: {}", bw),
+    };
+    let ndarray_calc_nd_indices_fn =
+        ctx.module.get_function(ndarray_calc_nd_indices_fn_name).unwrap_or_else(|| {
+            let fn_type = llvm_void.fn_type(
+                &[llvm_usize.into(), llvm_pusize.into(), llvm_usize.into(), llvm_pi32.into()],
+                false,
+            );
+
+            ctx.module.add_function(ndarray_calc_nd_indices_fn_name, fn_type, None)
+        });
+
+    let ndarray_num_dims = ndarray.load_ndims(ctx);
+    let ndarray_dims = ndarray.dim_sizes();
+
+    let indices = ctx.builder.build_array_alloca(llvm_i32, ndarray_num_dims, "").unwrap();
+
+    ctx.builder
+        .build_call(
+            ndarray_calc_nd_indices_fn,
+            &[
+                index.into(),
+                ndarray_dims.base_ptr(ctx, generator).into(),
+                ndarray_num_dims.into(),
+                indices.into(),
+            ],
+            "",
+        )
+        .unwrap();
+
+    TypedArrayLikeAdapter::from(
+        ArraySliceValue::from_ptr_val(indices, ndarray_num_dims, None),
+        Box::new(|_, v| v.into_int_value()),
+        Box::new(|_, v| v.into()),
+    )
+}
+
+fn call_ndarray_flatten_index_impl<'ctx, G, Indices>(
+    generator: &G,
+    ctx: &CodeGenContext<'ctx, '_>,
+    ndarray: NDArrayValue<'ctx>,
+    indices: &Indices,
+) -> IntValue<'ctx>
+where
+    G: CodeGenerator + ?Sized,
+    Indices: ArrayLikeIndexer<'ctx>,
+{
+    let llvm_i32 = ctx.ctx.i32_type();
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+
+    let llvm_pi32 = llvm_i32.ptr_type(AddressSpace::default());
+    let llvm_pusize = llvm_usize.ptr_type(AddressSpace::default());
+
+    debug_assert_eq!(
+        IntType::try_from(indices.element_type(ctx, generator))
+            .map(IntType::get_bit_width)
+            .unwrap_or_default(),
+        llvm_i32.get_bit_width(),
+        "Expected i32 value for argument `indices` to `call_ndarray_flatten_index_impl`"
+    );
+    debug_assert_eq!(
+        indices.size(ctx, generator).get_type().get_bit_width(),
+        llvm_usize.get_bit_width(),
+        "Expected usize integer value for argument `indices_size` to `call_ndarray_flatten_index_impl`"
+    );
+
+    let ndarray_flatten_index_fn_name = match llvm_usize.get_bit_width() {
+        32 => "__nac3_ndarray_flatten_index",
+        64 => "__nac3_ndarray_flatten_index64",
+        bw => unreachable!("Unsupported size type bit width: {}", bw),
+    };
+    let ndarray_flatten_index_fn =
+        ctx.module.get_function(ndarray_flatten_index_fn_name).unwrap_or_else(|| {
+            let fn_type = llvm_usize.fn_type(
+                &[llvm_pusize.into(), llvm_usize.into(), llvm_pi32.into(), llvm_usize.into()],
+                false,
+            );
+
+            ctx.module.add_function(ndarray_flatten_index_fn_name, fn_type, None)
+        });
+
+    let ndarray_num_dims = ndarray.load_ndims(ctx);
+    let ndarray_dims = ndarray.dim_sizes();
+
+    let index = ctx
+        .builder
+        .build_call(
+            ndarray_flatten_index_fn,
+            &[
+                ndarray_dims.base_ptr(ctx, generator).into(),
+                ndarray_num_dims.into(),
+                indices.base_ptr(ctx, generator).into(),
+                indices.size(ctx, generator).into(),
+            ],
+            "",
+        )
+        .map(CallSiteValue::try_as_basic_value)
+        .map(|v| v.map_left(BasicValueEnum::into_int_value))
+        .map(Either::unwrap_left)
+        .unwrap();
+
+    index
+}
+
+/// Generates a call to `__nac3_ndarray_flatten_index`. Returns the flattened index for the
+/// multidimensional index.
+///
+/// * `ndarray` - LLVM pointer to the `NDArray`. This value must be the LLVM representation of an
+/// `NDArray`.
+/// * `indices` - The multidimensional index to compute the flattened index for.
+pub fn call_ndarray_flatten_index<'ctx, G, Index>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    ndarray: NDArrayValue<'ctx>,
+    indices: &Index,
+) -> IntValue<'ctx>
+where
+    G: CodeGenerator + ?Sized,
+    Index: ArrayLikeIndexer<'ctx>,
+{
+    call_ndarray_flatten_index_impl(generator, ctx, ndarray, indices)
+}
+
+/// Generates a call to `__nac3_ndarray_calc_broadcast`. Returns a tuple containing the number of
+/// dimension and size of each dimension of the resultant `ndarray`.
+pub fn call_ndarray_calc_broadcast<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    lhs: NDArrayValue<'ctx>,
+    rhs: NDArrayValue<'ctx>,
+) -> TypedArrayLikeAdapter<'ctx, IntValue<'ctx>> {
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+    let llvm_pusize = llvm_usize.ptr_type(AddressSpace::default());
+
+    let ndarray_calc_broadcast_fn_name = match llvm_usize.get_bit_width() {
+        32 => "__nac3_ndarray_calc_broadcast",
+        64 => "__nac3_ndarray_calc_broadcast64",
+        bw => unreachable!("Unsupported size type bit width: {}", bw),
+    };
+    let ndarray_calc_broadcast_fn =
+        ctx.module.get_function(ndarray_calc_broadcast_fn_name).unwrap_or_else(|| {
+            let fn_type = llvm_usize.fn_type(
+                &[
+                    llvm_pusize.into(),
+                    llvm_usize.into(),
+                    llvm_pusize.into(),
+                    llvm_usize.into(),
+                    llvm_pusize.into(),
+                ],
+                false,
+            );
+
+            ctx.module.add_function(ndarray_calc_broadcast_fn_name, fn_type, None)
+        });
+
+    let lhs_ndims = lhs.load_ndims(ctx);
+    let rhs_ndims = rhs.load_ndims(ctx);
+    let min_ndims = llvm_intrinsics::call_int_umin(ctx, lhs_ndims, rhs_ndims, None);
+
+    gen_for_callback_incrementing(
+        generator,
+        ctx,
+        llvm_usize.const_zero(),
+        (min_ndims, false),
+        |generator, ctx, _, idx| {
+            let idx = ctx.builder.build_int_sub(min_ndims, idx, "").unwrap();
+            let (lhs_dim_sz, rhs_dim_sz) = unsafe {
+                (
+                    lhs.dim_sizes().get_typed_unchecked(ctx, generator, &idx, None),
+                    rhs.dim_sizes().get_typed_unchecked(ctx, generator, &idx, None),
+                )
+            };
+
+            let llvm_usize_const_one = llvm_usize.const_int(1, false);
+            let lhs_eqz = ctx
+                .builder
+                .build_int_compare(IntPredicate::EQ, lhs_dim_sz, llvm_usize_const_one, "")
+                .unwrap();
+            let rhs_eqz = ctx
+                .builder
+                .build_int_compare(IntPredicate::EQ, rhs_dim_sz, llvm_usize_const_one, "")
+                .unwrap();
+            let lhs_or_rhs_eqz = ctx.builder.build_or(lhs_eqz, rhs_eqz, "").unwrap();
+
+            let lhs_eq_rhs = ctx
+                .builder
+                .build_int_compare(IntPredicate::EQ, lhs_dim_sz, rhs_dim_sz, "")
+                .unwrap();
+
+            let is_compatible = ctx.builder.build_or(lhs_or_rhs_eqz, lhs_eq_rhs, "").unwrap();
+
+            ctx.make_assert(
+                generator,
+                is_compatible,
+                "0:ValueError",
+                "operands could not be broadcast together",
+                [None, None, None],
+                ctx.current_loc,
+            );
+
+            Ok(())
+        },
+        llvm_usize.const_int(1, false),
+    )
+    .unwrap();
+
+    let max_ndims = llvm_intrinsics::call_int_umax(ctx, lhs_ndims, rhs_ndims, None);
+    let lhs_dims = lhs.dim_sizes().base_ptr(ctx, generator);
+    let lhs_ndims = lhs.load_ndims(ctx);
+    let rhs_dims = rhs.dim_sizes().base_ptr(ctx, generator);
+    let rhs_ndims = rhs.load_ndims(ctx);
+    let out_dims = ctx.builder.build_array_alloca(llvm_usize, max_ndims, "").unwrap();
+    let out_dims = ArraySliceValue::from_ptr_val(out_dims, max_ndims, None);
+
+    ctx.builder
+        .build_call(
+            ndarray_calc_broadcast_fn,
+            &[
+                lhs_dims.into(),
+                lhs_ndims.into(),
+                rhs_dims.into(),
+                rhs_ndims.into(),
+                out_dims.base_ptr(ctx, generator).into(),
+            ],
+            "",
+        )
+        .unwrap();
+
+    TypedArrayLikeAdapter::from(
+        out_dims,
+        Box::new(|_, v| v.into_int_value()),
+        Box::new(|_, v| v.into()),
+    )
+}
+
+/// Generates a call to `__nac3_ndarray_calc_broadcast_idx`. Returns an [`ArrayAllocaValue`]
+/// containing the indices used for accessing `array` corresponding to the index of the broadcasted
+/// array `broadcast_idx`.
+pub fn call_ndarray_calc_broadcast_index<
+    'ctx,
+    G: CodeGenerator + ?Sized,
+    BroadcastIdx: UntypedArrayLikeAccessor<'ctx>,
+>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    array: NDArrayValue<'ctx>,
+    broadcast_idx: &BroadcastIdx,
+) -> TypedArrayLikeAdapter<'ctx, IntValue<'ctx>> {
+    let llvm_i32 = ctx.ctx.i32_type();
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+    let llvm_pi32 = llvm_i32.ptr_type(AddressSpace::default());
+    let llvm_pusize = llvm_usize.ptr_type(AddressSpace::default());
+
+    let ndarray_calc_broadcast_fn_name = match llvm_usize.get_bit_width() {
+        32 => "__nac3_ndarray_calc_broadcast_idx",
+        64 => "__nac3_ndarray_calc_broadcast_idx64",
+        bw => unreachable!("Unsupported size type bit width: {}", bw),
+    };
+    let ndarray_calc_broadcast_fn =
+        ctx.module.get_function(ndarray_calc_broadcast_fn_name).unwrap_or_else(|| {
+            let fn_type = llvm_usize.fn_type(
+                &[llvm_pusize.into(), llvm_usize.into(), llvm_pi32.into(), llvm_pi32.into()],
+                false,
+            );
+
+            ctx.module.add_function(ndarray_calc_broadcast_fn_name, fn_type, None)
+        });
+
+    let broadcast_size = broadcast_idx.size(ctx, generator);
+    let out_idx = ctx.builder.build_array_alloca(llvm_i32, broadcast_size, "").unwrap();
+
+    let array_dims = array.dim_sizes().base_ptr(ctx, generator);
+    let array_ndims = array.load_ndims(ctx);
+    let broadcast_idx_ptr = unsafe {
+        broadcast_idx.ptr_offset_unchecked(ctx, generator, &llvm_usize.const_zero(), None)
+    };
+
+    ctx.builder
+        .build_call(
+            ndarray_calc_broadcast_fn,
+            &[array_dims.into(), array_ndims.into(), broadcast_idx_ptr.into(), out_idx.into()],
+            "",
+        )
+        .unwrap();
+
+    TypedArrayLikeAdapter::from(
+        ArraySliceValue::from_ptr_val(out_idx, broadcast_size, None),
+        Box::new(|_, v| v.into_int_value()),
+        Box::new(|_, v| v.into()),
+    )
+}
+
+fn get_size_variant<'ctx>(ty: IntType<'ctx>) -> SizeVariant {
+    match ty.get_bit_width() {
+        32 => SizeVariant::Bits32,
+        64 => SizeVariant::Bits64,
+        _ => unreachable!("Unsupported int type bit width {}", ty.get_bit_width()),
+    }
+}
+
+fn get_size_type_dependent_function<'ctx, BuildFuncTypeFn>(
+    ctx: &CodeGenContext<'ctx, '_>,
+    size_type: IntType<'ctx>,
+    base_name: &str,
+    build_func_type: BuildFuncTypeFn,
+) -> FunctionValue<'ctx>
+where
+    BuildFuncTypeFn: Fn() -> FunctionType<'ctx>,
+{
+    let mut fn_name = base_name.to_owned();
+    match get_size_variant(size_type) {
+        SizeVariant::Bits32 => {
+            // The original fn_name is the correct function name
+        }
+        SizeVariant::Bits64 => {
+            // Append "64" at the end, this is the naming convention for 64-bit
+            fn_name.push_str("64");
+        }
+    }
+
+    // Get (or declare then get if does not exist) the corresponding function
+    ctx.module.get_function(&fn_name).unwrap_or_else(|| {
+        let fn_type = build_func_type();
+        ctx.module.add_function(&fn_name, fn_type, None)
+    })
+}
+
+fn get_ndarray_struct_ptr<'ctx>(ctx: &'ctx Context, size_type: IntType<'ctx>) -> PointerType<'ctx> {
+    let i8_type = ctx.i8_type();
+
+    let ndarray_ty = NpArrayType { size_type, elem_type: i8_type.as_basic_type_enum() };
+    let struct_ty = ndarray_ty.fields().whole_struct.as_struct_type(ctx);
+    struct_ty.ptr_type(AddressSpace::default())
+}
+
+pub fn call_nac3_ndarray_size<'ctx>(
+    ctx: &CodeGenContext<'ctx, '_>,
+    ndarray: NpArrayValue<'ctx>,
+) -> IntValue<'ctx> {
+    let size_type = ndarray.ty.size_type;
+    let function = get_size_type_dependent_function(ctx, size_type, "__nac3_ndarray_size", || {
+        size_type.fn_type(&[get_ndarray_struct_ptr(ctx.ctx, size_type).into()], false)
+    });
+
+    ctx.builder
+        .build_call(function, &[ndarray.ptr.into()], "size")
+        .unwrap()
+        .try_as_basic_value()
+        .unwrap_left()
+        .into_int_value()
+}

nac3core/src/codegen/irrt/ndarray/array.rs

@@ -1,72 +0,0 @@
-use inkwell::{types::BasicTypeEnum, values::IntValue};
-
-use crate::codegen::{
-    CodeGenContext, CodeGenerator,
-    expr::infer_and_call_function,
-    irrt::get_usize_dependent_function_name,
-    values::{ListValue, ProxyValue, TypedArrayLikeAccessor, ndarray::NDArrayValue},
-};
-
-/// Generates a call to `__nac3_ndarray_array_set_and_validate_list_shape`.
-///
-/// Deduces the target shape of the `ndarray` from the provided `list`, raising an exception if
-/// there is any issue with the resultant `shape`.
-///
-/// `shape` must be pre-allocated by the caller of this function to `[usize; ndims]`, and must be
-/// initialized to all `-1`s.
-pub fn call_nac3_ndarray_array_set_and_validate_list_shape<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    list: ListValue<'ctx>,
-    ndims: IntValue<'ctx>,
-    shape: &impl TypedArrayLikeAccessor<'ctx, G, IntValue<'ctx>>,
-) {
-    let llvm_usize = ctx.get_size_type();
-    assert_eq!(list.get_type().element_type().unwrap(), ctx.ctx.i8_type().into());
-    assert_eq!(ndims.get_type(), llvm_usize);
-    assert_eq!(
-        BasicTypeEnum::try_from(shape.element_type(ctx, generator)).unwrap(),
-        llvm_usize.into()
-    );
-
-    let name =
-        get_usize_dependent_function_name(ctx, "__nac3_ndarray_array_set_and_validate_list_shape");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        None,
-        &[list.as_abi_value(ctx).into(), ndims.into(), shape.base_ptr(ctx, generator).into()],
-        None,
-        None,
-    );
-}
-
-/// Generates a call to `__nac3_ndarray_array_write_list_to_array`.
-///
-/// Copies the contents stored in `list` into `ndarray`.
-///
-/// The `ndarray` must fulfill the following preconditions:
-///
-/// - `ndarray.itemsize`: Must be initialized.
-/// - `ndarray.ndims`: Must be initialized.
-/// - `ndarray.shape`: Must be initialized.
-/// - `ndarray.data`: Must be allocated and contiguous.
-pub fn call_nac3_ndarray_array_write_list_to_array<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    list: ListValue<'ctx>,
-    ndarray: NDArrayValue<'ctx>,
-) {
-    assert_eq!(list.get_type().element_type().unwrap(), ctx.ctx.i8_type().into());
-
-    let name = get_usize_dependent_function_name(ctx, "__nac3_ndarray_array_write_list_to_array");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        None,
-        &[list.as_abi_value(ctx).into(), ndarray.as_abi_value(ctx).into()],
-        None,
-        None,
-    );
-}

nac3core/src/codegen/irrt/ndarray/basic.rs

@@ -1,257 +0,0 @@
-use inkwell::{
-    AddressSpace,
-    values::{BasicValueEnum, IntValue, PointerValue},
-};
-
-use crate::codegen::{
-    CodeGenContext, CodeGenerator,
-    expr::infer_and_call_function,
-    irrt::get_usize_dependent_function_name,
-    values::{ProxyValue, TypedArrayLikeAccessor, ndarray::NDArrayValue},
-};
-
-/// Generates a call to `__nac3_ndarray_util_assert_shape_no_negative`.
-///
-/// Assets that `shape` does not contain negative dimensions.
-pub fn call_nac3_ndarray_util_assert_shape_no_negative<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    shape: &impl TypedArrayLikeAccessor<'ctx, G, IntValue<'ctx>>,
-) {
-    let llvm_usize = ctx.get_size_type();
-
-    assert_eq!(shape.element_type(ctx, generator), llvm_usize.into());
-
-    let name =
-        get_usize_dependent_function_name(ctx, "__nac3_ndarray_util_assert_shape_no_negative");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_usize.into()),
-        &[shape.size(ctx, generator).into(), shape.base_ptr(ctx, generator).into()],
-        None,
-        None,
-    );
-}
-
-/// Generates a call to `__nac3_ndarray_util_assert_shape_output_shape_same`.
-///
-/// Asserts that `ndarray_shape` and `output_shape` are the same in the context of writing output to
-/// an `ndarray`.
-pub fn call_nac3_ndarray_util_assert_output_shape_same<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray_shape: &impl TypedArrayLikeAccessor<'ctx, G, IntValue<'ctx>>,
-    output_shape: &impl TypedArrayLikeAccessor<'ctx, G, IntValue<'ctx>>,
-) {
-    let llvm_usize = ctx.get_size_type();
-
-    assert_eq!(ndarray_shape.element_type(ctx, generator), llvm_usize.into());
-    assert_eq!(output_shape.element_type(ctx, generator), llvm_usize.into());
-
-    let name =
-        get_usize_dependent_function_name(ctx, "__nac3_ndarray_util_assert_output_shape_same");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_usize.into()),
-        &[
-            ndarray_shape.size(ctx, generator).into(),
-            ndarray_shape.base_ptr(ctx, generator).into(),
-            output_shape.size(ctx, generator).into(),
-            output_shape.base_ptr(ctx, generator).into(),
-        ],
-        None,
-        None,
-    );
-}
-
-/// Generates a call to `__nac3_ndarray_size`.
-///
-/// Returns a [`usize`][CodeGenerator::get_size_type] value of the number of elements of an
-/// `ndarray`, corresponding to the value of `ndarray.size`.
-pub fn call_nac3_ndarray_size<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-) -> IntValue<'ctx> {
-    let llvm_usize = ctx.get_size_type();
-
-    let name = get_usize_dependent_function_name(ctx, "__nac3_ndarray_size");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_usize.into()),
-        &[ndarray.as_abi_value(ctx).into()],
-        Some("size"),
-        None,
-    )
-    .map(BasicValueEnum::into_int_value)
-    .unwrap()
-}
-
-/// Generates a call to `__nac3_ndarray_nbytes`.
-///
-/// Returns a [`usize`][CodeGenerator::get_size_type] value of the number of bytes consumed by the
-/// data of the `ndarray`, corresponding to the value of `ndarray.nbytes`.
-pub fn call_nac3_ndarray_nbytes<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-) -> IntValue<'ctx> {
-    let llvm_usize = ctx.get_size_type();
-
-    let name = get_usize_dependent_function_name(ctx, "__nac3_ndarray_nbytes");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_usize.into()),
-        &[ndarray.as_abi_value(ctx).into()],
-        Some("nbytes"),
-        None,
-    )
-    .map(BasicValueEnum::into_int_value)
-    .unwrap()
-}
-
-/// Generates a call to `__nac3_ndarray_len`.
-///
-/// Returns a [`usize`][CodeGenerator::get_size_type] value of the size of the topmost dimension of
-/// the `ndarray`, corresponding to the value of `ndarray.__len__`.
-pub fn call_nac3_ndarray_len<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-) -> IntValue<'ctx> {
-    let llvm_usize = ctx.get_size_type();
-
-    let name = get_usize_dependent_function_name(ctx, "__nac3_ndarray_len");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_usize.into()),
-        &[ndarray.as_abi_value(ctx).into()],
-        Some("len"),
-        None,
-    )
-    .map(BasicValueEnum::into_int_value)
-    .unwrap()
-}
-
-/// Generates a call to `__nac3_ndarray_is_c_contiguous`.
-///
-/// Returns an `i1` value indicating whether the `ndarray` is C-contiguous.
-pub fn call_nac3_ndarray_is_c_contiguous<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-) -> IntValue<'ctx> {
-    let llvm_i1 = ctx.ctx.bool_type();
-
-    let name = get_usize_dependent_function_name(ctx, "__nac3_ndarray_is_c_contiguous");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_i1.into()),
-        &[ndarray.as_abi_value(ctx).into()],
-        Some("is_c_contiguous"),
-        None,
-    )
-    .map(BasicValueEnum::into_int_value)
-    .unwrap()
-}
-
-/// Generates a call to `__nac3_ndarray_get_nth_pelement`.
-///
-/// Returns a [`PointerValue`] to the `index`-th flattened element of the `ndarray`.
-pub fn call_nac3_ndarray_get_nth_pelement<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-    index: IntValue<'ctx>,
-) -> PointerValue<'ctx> {
-    let llvm_i8 = ctx.ctx.i8_type();
-    let llvm_pi8 = llvm_i8.ptr_type(AddressSpace::default());
-    let llvm_usize = ctx.get_size_type();
-
-    assert_eq!(index.get_type(), llvm_usize);
-
-    let name = get_usize_dependent_function_name(ctx, "__nac3_ndarray_get_nth_pelement");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_pi8.into()),
-        &[ndarray.as_abi_value(ctx).into(), index.into()],
-        Some("pelement"),
-        None,
-    )
-    .map(BasicValueEnum::into_pointer_value)
-    .unwrap()
-}
-
-/// Generates a call to `__nac3_ndarray_get_pelement_by_indices`.
-///
-/// `indices` must have the same number of elements as the number of dimensions in `ndarray`.
-///
-/// Returns a [`PointerValue`] to the element indexed by `indices`.
-pub fn call_nac3_ndarray_get_pelement_by_indices<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-    indices: &impl TypedArrayLikeAccessor<'ctx, G, IntValue<'ctx>>,
-) -> PointerValue<'ctx> {
-    let llvm_i8 = ctx.ctx.i8_type();
-    let llvm_pi8 = llvm_i8.ptr_type(AddressSpace::default());
-    let llvm_usize = ctx.get_size_type();
-
-    assert_eq!(indices.element_type(ctx, generator), llvm_usize.into());
-
-    let name = get_usize_dependent_function_name(ctx, "__nac3_ndarray_get_pelement_by_indices");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_pi8.into()),
-        &[ndarray.as_abi_value(ctx).into(), indices.base_ptr(ctx, generator).into()],
-        Some("pelement"),
-        None,
-    )
-    .map(BasicValueEnum::into_pointer_value)
-    .unwrap()
-}
-
-/// Generates a call to `__nac3_ndarray_set_strides_by_shape`.
-///
-/// Sets `ndarray.strides` assuming that `ndarray.shape` is C-contiguous.
-pub fn call_nac3_ndarray_set_strides_by_shape<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-) {
-    let name = get_usize_dependent_function_name(ctx, "__nac3_ndarray_set_strides_by_shape");
-
-    infer_and_call_function(ctx, &name, None, &[ndarray.as_abi_value(ctx).into()], None, None);
-}
-
-/// Generates a call to `__nac3_ndarray_copy_data`.
-///
-/// Copies all elements from `src_ndarray` to `dst_ndarray` using their flattened views. The number
-/// of elements in `src_ndarray` must be greater than or equal to the number of elements in
-/// `dst_ndarray`.
-pub fn call_nac3_ndarray_copy_data<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    src_ndarray: NDArrayValue<'ctx>,
-    dst_ndarray: NDArrayValue<'ctx>,
-) {
-    let name = get_usize_dependent_function_name(ctx, "__nac3_ndarray_copy_data");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        None,
-        &[src_ndarray.as_abi_value(ctx).into(), dst_ndarray.as_abi_value(ctx).into()],
-        None,
-        None,
-    );
-}

nac3core/src/codegen/irrt/ndarray/broadcast.rs

@@ -1,82 +0,0 @@
-use inkwell::values::IntValue;
-
-use crate::codegen::{
-    CodeGenContext, CodeGenerator,
-    expr::infer_and_call_function,
-    irrt::get_usize_dependent_function_name,
-    types::{ProxyType, ndarray::ShapeEntryType},
-    values::{
-        ArrayLikeValue, ArraySliceValue, ProxyValue, TypedArrayLikeAccessor, TypedArrayLikeMutator,
-        ndarray::NDArrayValue,
-    },
-};
-
-/// Generates a call to `__nac3_ndarray_broadcast_to`.
-///
-/// Attempts to broadcast `src_ndarray` to the new shape defined by `dst_ndarray`.
-///
-/// `dst_ndarray` must meet the following preconditions:
-///
-/// - `dst_ndarray.ndims` must be initialized and matching the length of `dst_ndarray.shape`.
-/// - `dst_ndarray.shape` must be initialized and contains the target broadcast shape.
-/// - `dst_ndarray.strides` must be allocated and may contain uninitialized values.
-pub fn call_nac3_ndarray_broadcast_to<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    src_ndarray: NDArrayValue<'ctx>,
-    dst_ndarray: NDArrayValue<'ctx>,
-) {
-    let name = get_usize_dependent_function_name(ctx, "__nac3_ndarray_broadcast_to");
-    infer_and_call_function(
-        ctx,
-        &name,
-        None,
-        &[src_ndarray.as_abi_value(ctx).into(), dst_ndarray.as_abi_value(ctx).into()],
-        None,
-        None,
-    );
-}
-
-/// Generates a call to `__nac3_ndarray_broadcast_shapes`.
-///
-/// Attempts to calculate the resultant shape from broadcasting all shapes in `shape_entries`,
-/// writing the result to `dst_shape`.
-pub fn call_nac3_ndarray_broadcast_shapes<'ctx, G, Shape>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    num_shape_entries: IntValue<'ctx>,
-    shape_entries: ArraySliceValue<'ctx>,
-    dst_ndims: IntValue<'ctx>,
-    dst_shape: &Shape,
-) where
-    G: CodeGenerator + ?Sized,
-    Shape: TypedArrayLikeAccessor<'ctx, G, IntValue<'ctx>>
-        + TypedArrayLikeMutator<'ctx, G, IntValue<'ctx>>,
-{
-    let llvm_usize = ctx.get_size_type();
-
-    assert_eq!(num_shape_entries.get_type(), llvm_usize);
-    assert!(
-        ShapeEntryType::is_representable(
-            shape_entries.base_ptr(ctx, generator).get_type(),
-            llvm_usize,
-        )
-        .is_ok()
-    );
-    assert_eq!(dst_ndims.get_type(), llvm_usize);
-    assert_eq!(dst_shape.element_type(ctx, generator), llvm_usize.into());
-
-    let name = get_usize_dependent_function_name(ctx, "__nac3_ndarray_broadcast_shapes");
-    infer_and_call_function(
-        ctx,
-        &name,
-        None,
-        &[
-            num_shape_entries.into(),
-            shape_entries.base_ptr(ctx, generator).into(),
-            dst_ndims.into(),
-            dst_shape.base_ptr(ctx, generator).into(),
-        ],
-        None,
-        None,
-    );
-}

nac3core/src/codegen/irrt/ndarray/indexing.rs

@@ -1,34 +0,0 @@
-use crate::codegen::{
-    CodeGenContext, CodeGenerator,
-    expr::infer_and_call_function,
-    irrt::get_usize_dependent_function_name,
-    values::{ArrayLikeValue, ArraySliceValue, ProxyValue, ndarray::NDArrayValue},
-};
-
-/// Generates a call to `__nac3_ndarray_index`.
-///
-/// Performs [basic indexing](https://numpy.org/doc/stable/user/basics.indexing.html#basic-indexing)
-/// on `src_ndarray` using `indices`, writing the result to `dst_ndarray`, corresponding to the
-/// operation `dst_ndarray = src_ndarray[indices]`.
-pub fn call_nac3_ndarray_index<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    indices: ArraySliceValue<'ctx>,
-    src_ndarray: NDArrayValue<'ctx>,
-    dst_ndarray: NDArrayValue<'ctx>,
-) {
-    let name = get_usize_dependent_function_name(ctx, "__nac3_ndarray_index");
-    infer_and_call_function(
-        ctx,
-        &name,
-        None,
-        &[
-            indices.size(ctx, generator).into(),
-            indices.base_ptr(ctx, generator).into(),
-            src_ndarray.as_abi_value(ctx).into(),
-            dst_ndarray.as_abi_value(ctx).into(),
-        ],
-        None,
-        None,
-    );
-}

nac3core/src/codegen/irrt/ndarray/iter.rs

@@ -1,72 +0,0 @@
-use inkwell::values::{BasicValueEnum, IntValue};
-
-use crate::codegen::{
-    CodeGenContext, CodeGenerator,
-    expr::infer_and_call_function,
-    irrt::get_usize_dependent_function_name,
-    values::{
-        ProxyValue, TypedArrayLikeAccessor,
-        ndarray::{NDArrayValue, NDIterValue},
-    },
-};
-
-/// Generates a call to `__nac3_nditer_initialize`.
-///
-/// Initializes the `iter` object.
-pub fn call_nac3_nditer_initialize<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    iter: NDIterValue<'ctx>,
-    ndarray: NDArrayValue<'ctx>,
-    indices: &impl TypedArrayLikeAccessor<'ctx, G, IntValue<'ctx>>,
-) {
-    let llvm_usize = ctx.get_size_type();
-
-    assert_eq!(indices.element_type(ctx, generator), llvm_usize.into());
-
-    let name = get_usize_dependent_function_name(ctx, "__nac3_nditer_initialize");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        None,
-        &[
-            iter.as_abi_value(ctx).into(),
-            ndarray.as_abi_value(ctx).into(),
-            indices.base_ptr(ctx, generator).into(),
-        ],
-        None,
-        None,
-    );
-}
-
-/// Generates a call to `__nac3_nditer_initialize_has_element`.
-///
-/// Returns an `i1` value indicating whether there are elements left to traverse for the `iter`
-/// object.
-pub fn call_nac3_nditer_has_element<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    iter: NDIterValue<'ctx>,
-) -> IntValue<'ctx> {
-    let name = get_usize_dependent_function_name(ctx, "__nac3_nditer_has_element");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        Some(ctx.ctx.bool_type().into()),
-        &[iter.as_abi_value(ctx).into()],
-        None,
-        None,
-    )
-    .map(BasicValueEnum::into_int_value)
-    .unwrap()
-}
-
-/// Generates a call to `__nac3_nditer_next`.
-///
-/// Moves `iter` to point to the next element.
-pub fn call_nac3_nditer_next<'ctx>(ctx: &CodeGenContext<'ctx, '_>, iter: NDIterValue<'ctx>) {
-    let name = get_usize_dependent_function_name(ctx, "__nac3_nditer_next");
-
-    infer_and_call_function(ctx, &name, None, &[iter.as_abi_value(ctx).into()], None, None);
-}

nac3core/src/codegen/irrt/ndarray/matmul.rs

@@ -1,51 +0,0 @@
-use inkwell::values::IntValue;
-
-use crate::codegen::{
-    CodeGenContext, CodeGenerator, expr::infer_and_call_function,
-    irrt::get_usize_dependent_function_name, values::TypedArrayLikeAccessor,
-};
-
-/// Generates a call to `__nac3_ndarray_matmul_calculate_shapes`.
-///
-/// Calculates the broadcasted shapes for `a`, `b`, and the `ndarray` holding the final values of
-/// `a @ b`.
-#[allow(clippy::too_many_arguments)]
-pub fn call_nac3_ndarray_matmul_calculate_shapes<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    a_shape: &impl TypedArrayLikeAccessor<'ctx, G, IntValue<'ctx>>,
-    b_shape: &impl TypedArrayLikeAccessor<'ctx, G, IntValue<'ctx>>,
-    final_ndims: IntValue<'ctx>,
-    new_a_shape: &impl TypedArrayLikeAccessor<'ctx, G, IntValue<'ctx>>,
-    new_b_shape: &impl TypedArrayLikeAccessor<'ctx, G, IntValue<'ctx>>,
-    dst_shape: &impl TypedArrayLikeAccessor<'ctx, G, IntValue<'ctx>>,
-) {
-    let llvm_usize = ctx.get_size_type();
-
-    assert_eq!(a_shape.element_type(ctx, generator), llvm_usize.into());
-    assert_eq!(b_shape.element_type(ctx, generator), llvm_usize.into());
-    assert_eq!(final_ndims.get_type(), llvm_usize);
-    assert_eq!(new_a_shape.element_type(ctx, generator), llvm_usize.into());
-    assert_eq!(new_b_shape.element_type(ctx, generator), llvm_usize.into());
-    assert_eq!(dst_shape.element_type(ctx, generator), llvm_usize.into());
-
-    let name = get_usize_dependent_function_name(ctx, "__nac3_ndarray_matmul_calculate_shapes");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        None,
-        &[
-            a_shape.size(ctx, generator).into(),
-            a_shape.base_ptr(ctx, generator).into(),
-            b_shape.size(ctx, generator).into(),
-            b_shape.base_ptr(ctx, generator).into(),
-            final_ndims.into(),
-            new_a_shape.base_ptr(ctx, generator).into(),
-            new_b_shape.base_ptr(ctx, generator).into(),
-            dst_shape.base_ptr(ctx, generator).into(),
-        ],
-        None,
-        None,
-    );
-}

nac3core/src/codegen/irrt/ndarray/mod.rs

@@ -1,17 +0,0 @@
-pub use array::*;
-pub use basic::*;
-pub use broadcast::*;
-pub use indexing::*;
-pub use iter::*;
-pub use matmul::*;
-pub use reshape::*;
-pub use transpose::*;
-
-mod array;
-mod basic;
-mod broadcast;
-mod indexing;
-mod iter;
-mod matmul;
-mod reshape;
-mod transpose;

nac3core/src/codegen/irrt/ndarray/reshape.rs

@@ -1,39 +0,0 @@
-use inkwell::values::IntValue;
-
-use crate::codegen::{
-    CodeGenContext, CodeGenerator,
-    expr::infer_and_call_function,
-    irrt::get_usize_dependent_function_name,
-    values::{ArrayLikeValue, ArraySliceValue},
-};
-
-/// Generates a call to `__nac3_ndarray_reshape_resolve_and_check_new_shape`.
-///
-/// Resolves unknown dimensions in `new_shape` for `numpy.reshape(<ndarray>, new_shape)`, raising an
-/// assertion if multiple dimensions are unknown (`-1`).
-pub fn call_nac3_ndarray_reshape_resolve_and_check_new_shape<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    size: IntValue<'ctx>,
-    new_ndims: IntValue<'ctx>,
-    new_shape: ArraySliceValue<'ctx>,
-) {
-    let llvm_usize = ctx.get_size_type();
-
-    assert_eq!(size.get_type(), llvm_usize);
-    assert_eq!(new_ndims.get_type(), llvm_usize);
-    assert_eq!(new_shape.element_type(ctx, generator), llvm_usize.into());
-
-    let name = get_usize_dependent_function_name(
-        ctx,
-        "__nac3_ndarray_reshape_resolve_and_check_new_shape",
-    );
-    infer_and_call_function(
-        ctx,
-        &name,
-        None,
-        &[size.into(), new_ndims.into(), new_shape.base_ptr(ctx, generator).into()],
-        None,
-        None,
-    );
-}

nac3core/src/codegen/irrt/ndarray/transpose.rs

@@ -1,48 +0,0 @@
-use inkwell::{AddressSpace, values::IntValue};
-
-use crate::codegen::{
-    CodeGenContext, CodeGenerator,
-    expr::infer_and_call_function,
-    irrt::get_usize_dependent_function_name,
-    values::{ProxyValue, TypedArrayLikeAccessor, ndarray::NDArrayValue},
-};
-
-/// Generates a call to `__nac3_ndarray_transpose`.
-///
-/// Creates a transpose view of `src_ndarray` and writes the result to `dst_ndarray`.
-///
-/// `dst_ndarray` must fulfill the following preconditions:
-///
-/// - `dst_ndarray.ndims` must be initialized and must be equal to `src_ndarray.ndims`.
-/// - `dst_ndarray.shape` must be allocated and may contain uninitialized values.
-/// - `dst_ndarray.strides` must be allocated and may contain uninitialized values.
-pub fn call_nac3_ndarray_transpose<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    src_ndarray: NDArrayValue<'ctx>,
-    dst_ndarray: NDArrayValue<'ctx>,
-    axes: Option<&impl TypedArrayLikeAccessor<'ctx, G, IntValue<'ctx>>>,
-) {
-    let llvm_usize = ctx.get_size_type();
-
-    assert!(axes.is_none_or(|axes| axes.size(ctx, generator).get_type() == llvm_usize));
-    assert!(axes.is_none_or(|axes| axes.element_type(ctx, generator) == llvm_usize.into()));
-
-    let name = get_usize_dependent_function_name(ctx, "__nac3_ndarray_transpose");
-    infer_and_call_function(
-        ctx,
-        &name,
-        None,
-        &[
-            src_ndarray.as_abi_value(ctx).into(),
-            dst_ndarray.as_abi_value(ctx).into(),
-            axes.map_or(llvm_usize.const_zero(), |axes| axes.size(ctx, generator)).into(),
-            axes.map_or(llvm_usize.ptr_type(AddressSpace::default()).const_null(), |axes| {
-                axes.base_ptr(ctx, generator)
-            })
-            .into(),
-        ],
-        None,
-        None,
-    );
-}

nac3core/src/codegen/irrt/range.rs

@@ -1,53 +0,0 @@
-use inkwell::{
-    IntPredicate,
-    values::{BasicValueEnum, IntValue},
-};
-
-use crate::codegen::{CodeGenContext, CodeGenerator, expr::infer_and_call_function};
-
-/// Invokes the `__nac3_range_slice_len` in IRRT.
-///
-/// - `start`: The `i32` start value for the slice.
-/// - `end`: The `i32` end value for the slice.
-/// - `step`: The `i32` step value for the slice.
-///
-/// Returns an `i32` value of the length of the slice.
-pub fn calculate_len_for_slice_range<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &mut G,
-    ctx: &mut CodeGenContext<'ctx, '_>,
-    start: IntValue<'ctx>,
-    end: IntValue<'ctx>,
-    step: IntValue<'ctx>,
-) -> IntValue<'ctx> {
-    const SYMBOL: &str = "__nac3_range_slice_len";
-
-    let llvm_i32 = ctx.ctx.i32_type();
-    assert_eq!(start.get_type(), llvm_i32);
-    assert_eq!(end.get_type(), llvm_i32);
-    assert_eq!(step.get_type(), llvm_i32);
-
-    // assert step != 0, throw exception if not
-    let not_zero = ctx
-        .builder
-        .build_int_compare(IntPredicate::NE, step, step.get_type().const_zero(), "range_step_ne")
-        .unwrap();
-    ctx.make_assert(
-        generator,
-        not_zero,
-        "0:ValueError",
-        "step must not be zero",
-        [None, None, None],
-        ctx.current_loc,
-    );
-
-    infer_and_call_function(
-        ctx,
-        SYMBOL,
-        Some(llvm_i32.into()),
-        &[start.into(), end.into(), step.into()],
-        Some("calc_len"),
-        None,
-    )
-    .map(BasicValueEnum::into_int_value)
-    .unwrap()
-}

nac3core/src/codegen/irrt/slice.rs

@@ -1,41 +0,0 @@
-use inkwell::values::{BasicValueEnum, IntValue};
-
-use nac3parser::ast::Expr;
-
-use crate::{
-    codegen::{CodeGenContext, CodeGenerator, expr::infer_and_call_function},
-    typecheck::typedef::Type,
-};
-
-/// this function allows index out of range, since python
-/// allows index out of range in slice (`a = [1,2,3]; a[1:10] == [2,3]`).
-pub fn handle_slice_index_bound<'ctx, G: CodeGenerator>(
-    i: &Expr<Option<Type>>,
-    ctx: &mut CodeGenContext<'ctx, '_>,
-    generator: &mut G,
-    length: IntValue<'ctx>,
-) -> Result<Option<IntValue<'ctx>>, String> {
-    const SYMBOL: &str = "__nac3_slice_index_bound";
-
-    let llvm_i32 = ctx.ctx.i32_type();
-    assert_eq!(length.get_type(), llvm_i32);
-
-    let i = if let Some(v) = generator.gen_expr(ctx, i)? {
-        v.to_basic_value_enum(ctx, generator, i.custom.unwrap())?
-    } else {
-        return Ok(None);
-    };
-
-    Ok(Some(
-        infer_and_call_function(
-            ctx,
-            SYMBOL,
-            Some(llvm_i32.into()),
-            &[i, length.into()],
-            Some("bounded_ind"),
-            None,
-        )
-        .map(BasicValueEnum::into_int_value)
-        .unwrap(),
-    ))
-}

nac3core/src/codegen/irrt/string.rs

@@ -1,31 +0,0 @@
-use inkwell::values::{BasicValueEnum, IntValue};
-
-use super::get_usize_dependent_function_name;
-use crate::codegen::{CodeGenContext, expr::infer_and_call_function, values::StringValue};
-
-/// Generates a call to string equality comparison. Returns an `i1` representing whether the strings are equal.
-pub fn call_string_eq<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    str1: StringValue<'ctx>,
-    str2: StringValue<'ctx>,
-) -> IntValue<'ctx> {
-    let llvm_i1 = ctx.ctx.bool_type();
-
-    let func_name = get_usize_dependent_function_name(ctx, "nac3_str_eq");
-
-    infer_and_call_function(
-        ctx,
-        &func_name,
-        Some(llvm_i1.into()),
-        &[
-            str1.extract_ptr(ctx).into(),
-            str1.extract_len(ctx).into(),
-            str2.extract_ptr(ctx).into(),
-            str2.extract_len(ctx).into(),
-        ],
-        Some("str_eq_call"),
-        None,
-    )
-    .map(BasicValueEnum::into_int_value)
-    .unwrap()
-}

nac3core/src/codegen/irrt/test.rs

@@ -0,0 +1,26 @@
+#[cfg(test)]
+mod tests {
+    use std::{path::Path, process::Command};
+
+    #[test]
+    fn run_irrt_test() {
+        assert!(
+            cfg!(feature = "test"),
+            "Please do `cargo test -F test` to compile `irrt_test.out` and run test"
+        );
+
+        let irrt_test_out_path = Path::new(concat!(env!("OUT_DIR"), "/irrt_test.out"));
+        let output = Command::new(irrt_test_out_path.to_str().unwrap()).output().unwrap();
+
+        if !output.status.success() {
+            eprintln!("irrt_test failed with status {}:", output.status);
+            eprintln!("====== stdout ======");
+            eprintln!("{}", String::from_utf8(output.stdout).unwrap());
+            eprintln!("====== stderr ======");
+            eprintln!("{}", String::from_utf8(output.stderr).unwrap());
+            eprintln!("====================");
+
+            panic!("irrt_test failed");
+        }
+    }
+}

nac3core/src/codegen/llvm_intrinsics.rs

@@ -1,45 +1,38 @@
-use inkwell::{
-    AddressSpace,
-    intrinsics::Intrinsic,
-    types::AnyTypeEnum::IntType,
-    values::{BasicValueEnum, CallSiteValue, FloatValue, IntValue, PointerValue},
-};
+use crate::codegen::CodeGenContext;
+use inkwell::context::Context;
+use inkwell::intrinsics::Intrinsic;
+use inkwell::types::AnyTypeEnum::IntType;
+use inkwell::types::FloatType;
+use inkwell::values::{BasicValueEnum, CallSiteValue, FloatValue, IntValue, PointerValue};
+use inkwell::AddressSpace;
 use itertools::Either;
 
-use super::CodeGenContext;
+/// Returns the string representation for the floating-point type `ft` when used in intrinsic
+/// functions.
+fn get_float_intrinsic_repr(ctx: &Context, ft: FloatType) -> &'static str {
+    // Standard LLVM floating-point types
+    if ft == ctx.f16_type() {
+        return "f16";
+    }
+    if ft == ctx.f32_type() {
+        return "f32";
+    }
+    if ft == ctx.f64_type() {
+        return "f64";
+    }
+    if ft == ctx.f128_type() {
+        return "f128";
+    }
 
-/// Invokes the [`llvm.va_start`](https://llvm.org/docs/LangRef.html#llvm-va-start-intrinsic)
-/// intrinsic.
-pub fn call_va_start<'ctx>(ctx: &CodeGenContext<'ctx, '_>, arglist: PointerValue<'ctx>) {
-    const FN_NAME: &str = "llvm.va_start";
+    // Non-standard floating-point types
+    if ft == ctx.x86_f80_type() {
+        return "f80";
+    }
+    if ft == ctx.ppc_f128_type() {
+        return "ppcf128";
+    }
 
-    let intrinsic_fn = ctx.module.get_function(FN_NAME).unwrap_or_else(|| {
-        let llvm_void = ctx.ctx.void_type();
-        let llvm_i8 = ctx.ctx.i8_type();
-        let llvm_p0i8 = llvm_i8.ptr_type(AddressSpace::default());
-        let fn_type = llvm_void.fn_type(&[llvm_p0i8.into()], false);
-
-        ctx.module.add_function(FN_NAME, fn_type, None)
-    });
-
-    ctx.builder.build_call(intrinsic_fn, &[arglist.into()], "").unwrap();
-}
-
-/// Invokes the [`llvm.va_end`](https://llvm.org/docs/LangRef.html#llvm-va-end-intrinsic)
-/// intrinsic.
-pub fn call_va_end<'ctx>(ctx: &CodeGenContext<'ctx, '_>, arglist: PointerValue<'ctx>) {
-    const FN_NAME: &str = "llvm.va_end";
-
-    let intrinsic_fn = ctx.module.get_function(FN_NAME).unwrap_or_else(|| {
-        let llvm_void = ctx.ctx.void_type();
-        let llvm_i8 = ctx.ctx.i8_type();
-        let llvm_p0i8 = llvm_i8.ptr_type(AddressSpace::default());
-        let fn_type = llvm_void.fn_type(&[llvm_p0i8.into()], false);
-
-        ctx.module.add_function(FN_NAME, fn_type, None)
-    });
-
-    ctx.builder.build_call(intrinsic_fn, &[arglist.into()], "").unwrap();
+    unreachable!()
 }
 
 /// Invokes the [`llvm.stacksave`](https://llvm.org/docs/LangRef.html#llvm-stacksave-intrinsic)
@@ -98,29 +91,41 @@ pub fn call_stackrestore<'ctx>(ctx: &CodeGenContext<'ctx, '_>, ptr: PointerValue
 /// * `dest` - The pointer to the destination. Must be a pointer to an integer type.
 /// * `src` - The pointer to the source. Must be a pointer to an integer type.
 /// * `len` - The number of bytes to copy.
+/// * `is_volatile` - Whether the `memcpy` operation should be `volatile`.
 pub fn call_memcpy<'ctx>(
     ctx: &CodeGenContext<'ctx, '_>,
     dest: PointerValue<'ctx>,
     src: PointerValue<'ctx>,
     len: IntValue<'ctx>,
+    is_volatile: IntValue<'ctx>,
 ) {
+    const FN_NAME: &str = "llvm.memcpy";
+
     debug_assert!(dest.get_type().get_element_type().is_int_type());
     debug_assert!(src.get_type().get_element_type().is_int_type());
     debug_assert_eq!(
         dest.get_type().get_element_type().into_int_type().get_bit_width(),
         src.get_type().get_element_type().into_int_type().get_bit_width(),
     );
-    debug_assert_eq!(len.get_type(), ctx.get_size_type());
+    debug_assert!(matches!(len.get_type().get_bit_width(), 32 | 64));
+    debug_assert_eq!(is_volatile.get_type().get_bit_width(), 1);
 
     let llvm_dest_t = dest.get_type();
     let llvm_src_t = src.get_type();
+    let llvm_len_t = len.get_type();
 
-    let target_data =
-        ctx.registry.llvm_options.create_target_machine().map(|tm| tm.get_target_data()).unwrap();
-    let dest_alignment = target_data.get_abi_alignment(&llvm_dest_t);
-    let src_alignment = target_data.get_abi_alignment(&llvm_src_t);
+    let intrinsic_fn = Intrinsic::find(FN_NAME)
+        .and_then(|intrinsic| {
+            intrinsic.get_declaration(
+                &ctx.module,
+                &[llvm_dest_t.into(), llvm_src_t.into(), llvm_len_t.into()],
+            )
+        })
+        .unwrap();
 
-    ctx.builder.build_memcpy(dest, dest_alignment, src, src_alignment, len).unwrap();
+    ctx.builder
+        .build_call(intrinsic_fn, &[dest.into(), src.into(), len.into(), is_volatile.into()], "")
+        .unwrap();
 }
 
 /// Invokes the `llvm.memcpy` intrinsic.
@@ -132,6 +137,7 @@ pub fn call_memcpy_generic<'ctx>(
     dest: PointerValue<'ctx>,
     src: PointerValue<'ctx>,
     len: IntValue<'ctx>,
+    is_volatile: IntValue<'ctx>,
 ) {
     let llvm_i8 = ctx.ctx.i8_type();
     let llvm_p0i8 = llvm_i8.ptr_type(AddressSpace::default());
@@ -143,7 +149,7 @@ pub fn call_memcpy_generic<'ctx>(
         dest
     } else {
         ctx.builder
-            .build_bit_cast(dest, llvm_p0i8, "")
+            .build_bitcast(dest, llvm_p0i8, "")
             .map(BasicValueEnum::into_pointer_value)
             .unwrap()
     };
@@ -151,57 +157,12 @@ pub fn call_memcpy_generic<'ctx>(
         src
     } else {
         ctx.builder
-            .build_bit_cast(src, llvm_p0i8, "")
+            .build_bitcast(src, llvm_p0i8, "")
             .map(BasicValueEnum::into_pointer_value)
             .unwrap()
     };
 
-    call_memcpy(ctx, dest, src, len);
-}
-
-/// Invokes the `llvm.memcpy` intrinsic.
-///
-/// Unlike [`call_memcpy`], this function accepts any type of pointer value. If `dest` or `src` is
-/// not a pointer to an integer, the pointer(s) will be cast to `i8*` before invoking `memcpy`.
-/// Moreover, `len` now refers to the number of elements to copy (rather than number of bytes to
-/// copy).
-pub fn call_memcpy_generic_array<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    dest: PointerValue<'ctx>,
-    src: PointerValue<'ctx>,
-    len: IntValue<'ctx>,
-) {
-    let llvm_i8 = ctx.ctx.i8_type();
-    let llvm_p0i8 = llvm_i8.ptr_type(AddressSpace::default());
-    let llvm_usize = ctx.get_size_type();
-
-    let dest_elem_t = dest.get_type().get_element_type();
-    let src_elem_t = src.get_type().get_element_type();
-
-    let dest = if matches!(dest_elem_t, IntType(t) if t.get_bit_width() == 8) {
-        dest
-    } else {
-        ctx.builder
-            .build_bit_cast(dest, llvm_p0i8, "")
-            .map(BasicValueEnum::into_pointer_value)
-            .unwrap()
-    };
-    let src = if matches!(src_elem_t, IntType(t) if t.get_bit_width() == 8) {
-        src
-    } else {
-        ctx.builder
-            .build_bit_cast(src, llvm_p0i8, "")
-            .map(BasicValueEnum::into_pointer_value)
-            .unwrap()
-    };
-
-    let sizeof_elem = ctx
-        .builder
-        .build_int_truncate_or_bit_cast(src_elem_t.size_of().unwrap(), llvm_usize, "")
-        .unwrap();
-    let len = ctx.builder.build_int_mul(len, sizeof_elem, "").unwrap();
-
-    call_memcpy(ctx, dest, src, len);
+    call_memcpy(ctx, dest, src, len, is_volatile);
 }
 
 /// Macro to find and generate build call for llvm intrinsic (body of llvm intrinsic function)
@@ -210,9 +171,8 @@ pub fn call_memcpy_generic_array<'ctx>(
 /// * `$ctx:ident`: Reference to the current Code Generation Context
 /// * `$name:ident`: Optional name to be assigned to the llvm build call (Option<&str>)
 /// * `$llvm_name:literal`: Name of underlying llvm intrinsic function
-/// * `$map_fn:ident`: Mapping function to be applied on `BasicValue` (`BasicValue` -> Function Return Type).
-///   Use `BasicValueEnum::into_int_value` for Integer return type and
-///   `BasicValueEnum::into_float_value` for Float return type
+/// * `$map_fn:ident`: Mapping function to be applied on `BasicValue` (`BasicValue` -> Function Return Type)
+/// Use `BasicValueEnum::into_int_value` for Integer return type and `BasicValueEnum::into_float_value` for Float return type
 /// * `$llvm_ty:ident`: Type of first operand
 /// * `,($val:ident)*`: Comma separated list of operands
 macro_rules! generate_llvm_intrinsic_fn_body {
@@ -228,8 +188,8 @@ macro_rules! generate_llvm_intrinsic_fn_body {
 /// Arguments:
 /// * `float/int`: Indicates the return and argument type of the function
 /// * `$fn_name:ident`: The identifier of the rust function to be generated
-/// * `$llvm_name:literal`: Name of underlying llvm intrinsic function.
-///   Omit "llvm." prefix from the function name i.e. use "ceil" instead of "llvm.ceil"
+/// * `$llvm_name:literal`: Name of underlying llvm intrinsic function
+/// Omit "llvm." prefix from the function name i.e. use "ceil" instead of "llvm.ceil"
 /// * `$val:ident`: The operand for unary operations
 /// * `$val1:ident`, `$val2:ident`: The operands for binary operations
 macro_rules! generate_llvm_intrinsic_fn {
@@ -346,25 +306,3 @@ pub fn call_float_powi<'ctx>(
         .map(Either::unwrap_left)
         .unwrap()
 }
-
-/// Invokes the [`llvm.ctpop`](https://llvm.org/docs/LangRef.html#llvm-ctpop-intrinsic) intrinsic.
-pub fn call_int_ctpop<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    src: IntValue<'ctx>,
-    name: Option<&str>,
-) -> IntValue<'ctx> {
-    const FN_NAME: &str = "llvm.ctpop";
-
-    let llvm_src_t = src.get_type();
-
-    let intrinsic_fn = Intrinsic::find(FN_NAME)
-        .and_then(|intrinsic| intrinsic.get_declaration(&ctx.module, &[llvm_src_t.into()]))
-        .unwrap();
-
-    ctx.builder
-        .build_call(intrinsic_fn, &[src.into()], name.unwrap_or_default())
-        .map(CallSiteValue::try_as_basic_value)
-        .map(|v| v.map_left(BasicValueEnum::into_int_value))
-        .map(Either::unwrap_left)
-        .unwrap()
-}

nac3core/src/codegen/mod.rs

@@ -1,16 +1,14 @@
-use std::{
-    cell::OnceCell,
-    collections::{HashMap, HashSet},
-    sync::{
-        Arc,
-        atomic::{AtomicBool, Ordering},
+use crate::{
+    codegen::classes::{ListType, NDArrayType, ProxyType, RangeType},
+    symbol_resolver::{StaticValue, SymbolResolver},
+    toplevel::{helper::PrimDef, numpy::unpack_ndarray_var_tys, TopLevelContext, TopLevelDef},
+    typecheck::{
+        type_inferencer::{CodeLocation, PrimitiveStore},
+        typedef::{CallId, FuncArg, Type, TypeEnum, Unifier},
     },
-    thread,
 };
-
-use crossbeam::channel::{Receiver, Sender, unbounded};
+use crossbeam::channel::{unbounded, Receiver, Sender};
 use inkwell::{
-    AddressSpace, IntPredicate, OptimizationLevel,
     attributes::{Attribute, AttributeLoc},
     basic_block::BasicBlock,
     builder::Builder,
@@ -21,34 +19,22 @@ use inkwell::{
     module::Module,
     passes::PassBuilderOptions,
     targets::{CodeModel, RelocMode, Target, TargetMachine, TargetTriple},
-    types::{AnyType, BasicType, BasicTypeEnum, IntType},
+    types::{AnyType, BasicType, BasicTypeEnum},
     values::{BasicValueEnum, FunctionValue, IntValue, PhiValue, PointerValue},
+    AddressSpace, IntPredicate, OptimizationLevel,
 };
 use itertools::Itertools;
-use parking_lot::{Condvar, Mutex};
-
 use nac3parser::ast::{Location, Stmt, StrRef};
-
-use crate::{
-    symbol_resolver::{StaticValue, SymbolResolver},
-    toplevel::{
-        TopLevelContext, TopLevelDef,
-        helper::{PrimDef, extract_ndims},
-        numpy::unpack_ndarray_var_tys,
-    },
-    typecheck::{
-        type_inferencer::{CodeLocation, PrimitiveStore},
-        typedef::{CallId, FuncArg, Type, TypeEnum, Unifier},
-    },
-};
-use concrete_type::{ConcreteType, ConcreteTypeEnum, ConcreteTypeStore};
-pub use generator::{CodeGenerator, DefaultCodeGenerator};
-use types::{
-    ExceptionType, ListType, OptionType, ProxyType, RangeType, StringType, TupleType,
-    ndarray::NDArrayType,
+use parking_lot::{Condvar, Mutex};
+use std::collections::{HashMap, HashSet};
+use std::sync::{
+    atomic::{AtomicBool, Ordering},
+    Arc,
 };
+use std::thread;
 
 pub mod builtin_fns;
+pub mod classes;
 pub mod concrete_type;
 pub mod expr;
 pub mod extern_fns;
@@ -57,27 +43,12 @@ pub mod irrt;
 pub mod llvm_intrinsics;
 pub mod numpy;
 pub mod stmt;
-pub mod types;
-pub mod values;
 
 #[cfg(test)]
 mod test;
 
-mod macros {
-    /// Codegen-variant of [`std::unreachable`] which accepts an instance of [`CodeGenContext`] as
-    /// its first argument to provide Python source information to indicate the codegen location
-    /// causing the assertion.
-    macro_rules! codegen_unreachable {
-        ($ctx:expr $(,)?) => {
-            std::unreachable!("unreachable code while processing {}", &$ctx.current_loc)
-        };
-        ($ctx:expr, $($arg:tt)*) => {
-            std::unreachable!("unreachable code while processing {}: {}", &$ctx.current_loc, std::format!("{}", std::format_args!($($arg)+)))
-        };
-    }
-
-    pub(crate) use codegen_unreachable;
-}
+use concrete_type::{ConcreteType, ConcreteTypeEnum, ConcreteTypeStore};
+pub use generator::{CodeGenerator, DefaultCodeGenerator};
 
 #[derive(Default)]
 pub struct StaticValueStore {
@@ -97,16 +68,6 @@ pub struct CodeGenLLVMOptions {
     pub target: CodeGenTargetMachineOptions,
 }
 
-impl CodeGenLLVMOptions {
-    /// Creates a [`TargetMachine`] using the target options specified by this struct.
-    ///
-    /// See [`Target::create_target_machine`].
-    #[must_use]
-    pub fn create_target_machine(&self) -> Option<TargetMachine> {
-        self.target.create_target_machine(self.opt_level)
-    }
-}
-
 /// Additional options for code generation for the target machine.
 #[derive(Clone, Debug, Eq, PartialEq)]
 pub struct CodeGenTargetMachineOptions {
@@ -230,33 +191,14 @@ pub struct CodeGenContext<'ctx, 'a> {
 
     /// The current source location.
     pub current_loc: Location,
-
-    /// The cached type of `size_t`.
-    llvm_usize: OnceCell<IntType<'ctx>>,
 }
 
-impl<'ctx> CodeGenContext<'ctx, '_> {
+impl<'ctx, 'a> CodeGenContext<'ctx, 'a> {
     /// Whether the [current basic block][Builder::get_insert_block] referenced by `builder`
     /// contains a [terminator statement][BasicBlock::get_terminator].
     pub fn is_terminated(&self) -> bool {
         self.builder.get_insert_block().and_then(BasicBlock::get_terminator).is_some()
     }
-
-    /// Returns a [`IntType`] representing `size_t` for the compilation target as specified by
-    /// [`self.registry`][WorkerRegistry].
-    pub fn get_size_type(&self) -> IntType<'ctx> {
-        *self.llvm_usize.get_or_init(|| {
-            self.ctx.ptr_sized_int_type(
-                &self
-                    .registry
-                    .llvm_options
-                    .create_target_machine()
-                    .map(|tm| tm.get_target_data())
-                    .unwrap(),
-                None,
-            )
-        })
-    }
 }
 
 type Fp = Box<dyn Fn(&Module) + Send + Sync>;
@@ -396,10 +338,6 @@ impl WorkerRegistry {
         let mut builder = context.create_builder();
         let mut module = context.create_module(generator.get_name());
 
-        let target_machine = self.llvm_options.create_target_machine().unwrap();
-        module.set_data_layout(&target_machine.get_target_data().get_data_layout());
-        module.set_triple(&target_machine.get_triple());
-
         module.add_basic_value_flag(
             "Debug Info Version",
             inkwell::module::FlagBehavior::Warning,
@@ -423,10 +361,6 @@ impl WorkerRegistry {
                     errors.insert(e);
                     // create a new empty module just to continue codegen and collect errors
                     module = context.create_module(&format!("{}_recover", generator.get_name()));
-
-                    let target_machine = self.llvm_options.create_target_machine().unwrap();
-                    module.set_data_layout(&target_machine.get_target_data().get_data_layout());
-                    module.set_triple(&target_machine.get_triple());
                 }
             }
             *self.task_count.lock() -= 1;
@@ -492,7 +426,7 @@ pub struct CodeGenTask {
 fn get_llvm_type<'ctx, G: CodeGenerator + ?Sized>(
     ctx: &'ctx Context,
     module: &Module<'ctx>,
-    generator: &G,
+    generator: &mut G,
     unifier: &mut Unifier,
     top_level: &TopLevelContext,
     type_cache: &mut HashMap<Type, BasicTypeEnum<'ctx>>,
@@ -504,44 +438,12 @@ fn get_llvm_type<'ctx, G: CodeGenerator + ?Sized>(
     type_cache.get(&unifier.get_representative(ty)).copied().unwrap_or_else(|| {
         let ty_enum = unifier.get_ty(ty);
         let result = match &*ty_enum {
-            TModule {module_id, attributes} => {
-                let top_level_defs = top_level.definitions.read();
-                let definition = top_level_defs.get(module_id.0).unwrap();
-                let TopLevelDef::Module { name, attributes: attribute_fields, .. } = &*definition.read() else {
-                    unreachable!()
-                };
-                let ty: BasicTypeEnum<'_> = if let Some(t) = module.get_struct_type(&name.to_string()) {
-                    t.ptr_type(AddressSpace::default()).into()
-                } else {
-                    let struct_type = ctx.opaque_struct_type(&name.to_string());
-                    type_cache.insert(
-                        unifier.get_representative(ty),
-                        struct_type.ptr_type(AddressSpace::default()).into(),
-                    );
-                    let module_fields: Vec<BasicTypeEnum<'_>> = attribute_fields.iter()
-                    .map(|f| {
-                        get_llvm_type(
-                            ctx,
-                            module,
-                            generator,
-                            unifier,
-                            top_level,
-                            type_cache,
-                            attributes[&f.0].0,
-                        )
-                    })
-                    .collect_vec();
-                    struct_type.set_body(&module_fields, false);
-                    struct_type.ptr_type(AddressSpace::default()).into()
-                };
-                return ty;
-            },
             TObj { obj_id, fields, .. } => {
                 // check to avoid treating non-class primitives as classes
                 if PrimDef::contains_id(*obj_id) {
                     return match &*unifier.get_ty_immutable(ty) {
                         TObj { obj_id, params, .. } if *obj_id == PrimDef::Option.id() => {
-                            let element_type = get_llvm_type(
+                            get_llvm_type(
                                 ctx,
                                 module,
                                 generator,
@@ -549,9 +451,9 @@ fn get_llvm_type<'ctx, G: CodeGenerator + ?Sized>(
                                 top_level,
                                 type_cache,
                                 *params.iter().next().unwrap().1,
-                            );
-
-                            OptionType::new_with_generator(generator, ctx, &element_type).as_abi_type().into()
+                            )
+                            .ptr_type(AddressSpace::default())
+                            .into()
                         }
 
                         TObj { obj_id, params, .. } if *obj_id == PrimDef::List.id() => {
@@ -565,17 +467,16 @@ fn get_llvm_type<'ctx, G: CodeGenerator + ?Sized>(
                                 *params.iter().next().unwrap().1,
                             );
 
-                            ListType::new_with_generator(generator, ctx, element_type).as_abi_type().into()
+                            ListType::new(generator, ctx, element_type).as_base_type().into()
                         }
 
                         TObj { obj_id, .. } if *obj_id == PrimDef::NDArray.id() => {
-                            let (dtype, ndims) = unpack_ndarray_var_tys(unifier, ty);
-                            let ndims = extract_ndims(unifier, ndims);
+                            let (dtype, _) = unpack_ndarray_var_tys(unifier, ty);
                             let element_type = get_llvm_type(
                                 ctx, module, generator, unifier, top_level, type_cache, dtype,
                             );
 
-                            NDArrayType::new_with_generator(generator, ctx, element_type, ndims).as_abi_type().into()
+                            NDArrayType::new(generator, ctx, element_type).as_base_type().into()
                         }
 
                         _ => unreachable!(
@@ -619,17 +520,15 @@ fn get_llvm_type<'ctx, G: CodeGenerator + ?Sized>(
                 };
                 return ty;
             }
-            TTuple { ty, is_vararg_ctx } => {
+            TTuple { ty } => {
                 // a struct with fields in the order present in the tuple
-                assert!(!is_vararg_ctx, "Tuples in vararg context must be instantiated with the correct number of arguments before calling get_llvm_type");
-
                 let fields = ty
                     .iter()
                     .map(|ty| {
                         get_llvm_type(ctx, module, generator, unifier, top_level, type_cache, *ty)
                     })
                     .collect_vec();
-                TupleType::new_with_generator(generator, ctx, &fields).as_abi_type().into()
+                ctx.struct_type(&fields, false).into()
             }
             TVirtual { .. } => unimplemented!(),
             _ => unreachable!("{}", ty_enum.get_type_name()),
@@ -652,7 +551,7 @@ fn get_llvm_type<'ctx, G: CodeGenerator + ?Sized>(
 fn get_llvm_abi_type<'ctx, G: CodeGenerator + ?Sized>(
     ctx: &'ctx Context,
     module: &Module<'ctx>,
-    generator: &G,
+    generator: &mut G,
     unifier: &mut Unifier,
     top_level: &TopLevelContext,
     type_cache: &mut HashMap<Type, BasicTypeEnum<'ctx>>,
@@ -661,11 +560,11 @@ fn get_llvm_abi_type<'ctx, G: CodeGenerator + ?Sized>(
 ) -> BasicTypeEnum<'ctx> {
     // If the type is used in the definition of a function, return `i1` instead of `i8` for ABI
     // consistency.
-    if unifier.unioned(ty, primitives.bool) {
+    return if unifier.unioned(ty, primitives.bool) {
         ctx.bool_type().into()
     } else {
         get_llvm_type(ctx, module, generator, unifier, top_level, type_cache, ty)
-    }
+    };
 }
 
 /// Whether `sret` is needed for a return value with type `ty`.
@@ -690,40 +589,6 @@ fn need_sret(ty: BasicTypeEnum) -> bool {
     need_sret_impl(ty, true)
 }
 
-/// Returns the [`BasicTypeEnum`] representing a `va_list` struct for variadic arguments.
-fn get_llvm_valist_type<'ctx>(ctx: &'ctx Context, triple: &TargetTriple) -> BasicTypeEnum<'ctx> {
-    let triple = TargetMachine::normalize_triple(triple);
-    let triple = triple.as_str().to_str().unwrap();
-    let arch = triple.split('-').next().unwrap();
-
-    let llvm_pi8 = ctx.i8_type().ptr_type(AddressSpace::default());
-
-    // Referenced from parseArch() in llvm/lib/Support/Triple.cpp
-    match arch {
-        "i386" | "i486" | "i586" | "i686" | "riscv32" => {
-            ctx.i8_type().ptr_type(AddressSpace::default()).into()
-        }
-        "amd64" | "x86_64" | "x86_64h" => {
-            let llvm_i32 = ctx.i32_type();
-
-            let va_list_tag = ctx.opaque_struct_type("struct.__va_list_tag");
-            va_list_tag.set_body(
-                &[llvm_i32.into(), llvm_i32.into(), llvm_pi8.into(), llvm_pi8.into()],
-                false,
-            );
-            va_list_tag.into()
-        }
-        "armv7" => {
-            let va_list = ctx.opaque_struct_type("struct.__va_list");
-            va_list.set_body(&[llvm_pi8.into()], false);
-            va_list.into()
-        }
-        triple => {
-            todo!("Unsupported platform for varargs: {triple}")
-        }
-    }
-}
-
 /// Implementation for generating LLVM IR for a function.
 pub fn gen_func_impl<
     'ctx,
@@ -789,11 +654,34 @@ pub fn gen_func_impl<
         (primitives.float, context.f64_type().into()),
         (primitives.bool, context.i8_type().into()),
         (primitives.str, {
-            StringType::new_with_generator(generator, context).as_abi_type().into()
+            let name = "str";
+            match module.get_struct_type(name) {
+                None => {
+                    let str_type = context.opaque_struct_type("str");
+                    let fields = [
+                        context.i8_type().ptr_type(AddressSpace::default()).into(),
+                        generator.get_size_type(context).into(),
+                    ];
+                    str_type.set_body(&fields, false);
+                    str_type.into()
+                }
+                Some(t) => t.as_basic_type_enum(),
+            }
         }),
-        (primitives.range, RangeType::new_with_generator(generator, context).as_abi_type().into()),
+        (primitives.range, RangeType::new(context).as_base_type().into()),
         (primitives.exception, {
-            ExceptionType::new_with_generator(generator, context).as_abi_type().into()
+            let name = "Exception";
+            if let Some(t) = module.get_struct_type(name) {
+                t.ptr_type(AddressSpace::default()).as_basic_type_enum()
+            } else {
+                let exception = context.opaque_struct_type("Exception");
+                let int32 = context.i32_type().into();
+                let int64 = context.i64_type().into();
+                let str_ty = module.get_struct_type("str").unwrap().as_basic_type_enum();
+                let fields = [int32, str_ty, int32, int32, str_ty, str_ty, int64, int64, int64];
+                exception.set_body(&fields, false);
+                exception.ptr_type(AddressSpace::default()).as_basic_type_enum()
+            }
         }),
     ]
     .iter()
@@ -812,7 +700,6 @@ pub fn gen_func_impl<
                 name: arg.name,
                 ty: task.store.to_unifier_type(&mut unifier, &primitives, arg.ty, &mut cache),
                 default_value: arg.default_value.clone(),
-                is_vararg: arg.is_vararg,
             })
             .collect_vec(),
         task.store.to_unifier_type(&mut unifier, &primitives, *ret, &mut cache),
@@ -832,13 +719,10 @@ pub fn gen_func_impl<
         ))
     };
 
-    let has_sret = ret_type.is_some_and(|ty| need_sret(ty));
+    let has_sret = ret_type.map_or(false, |ty| need_sret(ty));
     let mut params = args
         .iter()
-        .filter(|arg| !arg.is_vararg)
         .map(|arg| {
-            debug_assert!(!arg.is_vararg);
-
             get_llvm_abi_type(
                 context,
                 &module,
@@ -857,12 +741,9 @@ pub fn gen_func_impl<
         params.insert(0, ret_type.unwrap().ptr_type(AddressSpace::default()).into());
     }
 
-    debug_assert!(matches!(args.iter().filter(|arg| arg.is_vararg).count(), 0..=1));
-    let vararg_arg = args.iter().find(|arg| arg.is_vararg);
-
     let fn_type = match ret_type {
-        Some(ret_type) if !has_sret => ret_type.fn_type(&params, vararg_arg.is_some()),
-        _ => context.void_type().fn_type(&params, vararg_arg.is_some()),
+        Some(ret_type) if !has_sret => ret_type.fn_type(&params, false),
+        _ => context.void_type().fn_type(&params, false),
     };
 
     let symbol = &task.symbol_name;
@@ -890,10 +771,9 @@ pub fn gen_func_impl<
     builder.position_at_end(init_bb);
     let body_bb = context.append_basic_block(fn_val, "body");
 
-    // Store non-vararg argument values into local variables
     let mut var_assignment = HashMap::new();
     let offset = u32::from(has_sret);
-    for (n, arg) in args.iter().enumerate().filter(|(_, arg)| !arg.is_vararg) {
+    for (n, arg) in args.iter().enumerate() {
         let param = fn_val.get_nth_param((n as u32) + offset).unwrap();
         let local_type = get_llvm_type(
             context,
@@ -913,7 +793,7 @@ pub fn gen_func_impl<
             let param_val = param.into_int_value();
 
             if expected_ty.get_bit_width() == 8 && param_val.get_type().get_bit_width() == 1 {
-                bool_to_int_type(&builder, param_val, context.i8_type())
+                bool_to_i8(&builder, context, param_val)
             } else {
                 param_val
             }
@@ -926,8 +806,6 @@ pub fn gen_func_impl<
         var_assignment.insert(arg.name, (alloca, None, 0));
     }
 
-    // TODO: Save vararg parameters as list
-
     let return_buffer = if has_sret {
         Some(fn_val.get_nth_param(0).unwrap().into_pointer_value())
     } else {
@@ -1019,19 +897,8 @@ pub fn gen_func_impl<
         need_sret: has_sret,
         current_loc: Location::default(),
         debug_info: (dibuilder, compile_unit, func_scope.as_debug_info_scope()),
-        llvm_usize: OnceCell::default(),
     };
 
-    let target_llvm_usize = context.ptr_sized_int_type(
-        &registry.llvm_options.create_target_machine().map(|tm| tm.get_target_data()).unwrap(),
-        None,
-    );
-    let generator_llvm_usize = generator.get_size_type(context);
-    assert_eq!(
-        generator_llvm_usize, target_llvm_usize,
-        "CodeGenerator (size_t = {generator_llvm_usize}) is not compatible with CodeGen Target (size_t = {target_llvm_usize})",
-    );
-
     let loc = code_gen_context.debug_info.0.create_debug_location(
         context,
         row as u32,
@@ -1082,29 +949,43 @@ pub fn gen_func<'ctx, G: CodeGenerator>(
     })
 }
 
-/// Converts the value of a boolean-like value `value` into an arbitrary [`IntType`].
-///
-/// This has the same semantics as `(ty)(value != 0)` in C.
-///
-/// The returned value is guaranteed to either be `0` or `1`, except for `ty == i1` where only the
-/// least-significant bit would be guaranteed to be `0` or `1`.
-fn bool_to_int_type<'ctx>(
+/// Converts the value of a boolean-like value `bool_value` into an `i1`.
+fn bool_to_i1<'ctx>(builder: &Builder<'ctx>, bool_value: IntValue<'ctx>) -> IntValue<'ctx> {
+    if bool_value.get_type().get_bit_width() == 1 {
+        bool_value
+    } else {
+        builder
+            .build_int_compare(
+                IntPredicate::NE,
+                bool_value,
+                bool_value.get_type().const_zero(),
+                "tobool",
+            )
+            .unwrap()
+    }
+}
+
+/// Converts the value of a boolean-like value `bool_value` into an `i8`.
+fn bool_to_i8<'ctx>(
     builder: &Builder<'ctx>,
-    value: IntValue<'ctx>,
-    ty: IntType<'ctx>,
+    ctx: &'ctx Context,
+    bool_value: IntValue<'ctx>,
 ) -> IntValue<'ctx> {
-    // i1 -> i1    : %value                                     ; no-op
-    // i1 -> i<N>  : zext i1 %value to i<N>                     ; guaranteed to be 0 or 1 - see docs
-    // i<M> -> i<N>: zext i1 (icmp eq i<M> %value, 0) to i<N>   ; same as i<M> -> i1 -> i<N>
-    match (value.get_type().get_bit_width(), ty.get_bit_width()) {
-        (1, 1) => value,
-        (1, _) => builder.build_int_z_extend(value, ty, "frombool").unwrap(),
-        _ => bool_to_int_type(
+    let value_bits = bool_value.get_type().get_bit_width();
+    match value_bits {
+        8 => bool_value,
+        1 => builder.build_int_z_extend(bool_value, ctx.i8_type(), "frombool").unwrap(),
+        _ => bool_to_i8(
             builder,
+            ctx,
             builder
-                .build_int_compare(IntPredicate::NE, value, value.get_type().const_zero(), "tobool")
+                .build_int_compare(
+                    IntPredicate::NE,
+                    bool_value,
+                    bool_value.get_type().const_zero(),
+                    "",
+                )
                 .unwrap(),
-            ty,
         ),
     }
 }
@@ -1147,112 +1028,3 @@ fn gen_in_range_check<'ctx>(
 
     ctx.builder.build_int_compare(IntPredicate::SLT, lo, hi, "cmp").unwrap()
 }
-
-/// Returns the internal name for the `va_count` argument, used to indicate the number of arguments
-/// passed to the variadic function.
-fn get_va_count_arg_name(arg_name: StrRef) -> StrRef {
-    format!("__{}_va_count", &arg_name).into()
-}
-
-/// Returns the alignment of the type.
-///
-/// This is necessary as `get_alignment` is not implemented as part of [`BasicType`].
-pub fn get_type_alignment<'ctx>(ty: impl Into<BasicTypeEnum<'ctx>>) -> IntValue<'ctx> {
-    match ty.into() {
-        BasicTypeEnum::ArrayType(ty) => ty.get_alignment(),
-        BasicTypeEnum::FloatType(ty) => ty.get_alignment(),
-        BasicTypeEnum::IntType(ty) => ty.get_alignment(),
-        BasicTypeEnum::PointerType(ty) => ty.get_alignment(),
-        BasicTypeEnum::StructType(ty) => ty.get_alignment(),
-        BasicTypeEnum::VectorType(ty) => ty.get_alignment(),
-    }
-}
-
-/// Inserts an `alloca` instruction with allocation `size` given in bytes and the alignment of the
-/// given type.
-///
-/// The returned [`PointerValue`] will have a type of `i8*`, a size of at least `size`, and will be
-/// aligned with the alignment of `align_ty`.
-pub fn type_aligned_alloca<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &mut G,
-    ctx: &mut CodeGenContext<'ctx, '_>,
-    align_ty: impl Into<BasicTypeEnum<'ctx>>,
-    size: IntValue<'ctx>,
-    name: Option<&str>,
-) -> PointerValue<'ctx> {
-    /// Round `val` up to its modulo `power_of_two`.
-    fn round_up<'ctx>(
-        ctx: &CodeGenContext<'ctx, '_>,
-        val: IntValue<'ctx>,
-        power_of_two: IntValue<'ctx>,
-    ) -> IntValue<'ctx> {
-        debug_assert_eq!(
-            val.get_type().get_bit_width(),
-            power_of_two.get_type().get_bit_width(),
-            "`val` ({}) and `power_of_two` ({}) must be the same type",
-            val.get_type(),
-            power_of_two.get_type(),
-        );
-
-        let llvm_val_t = val.get_type();
-
-        let max_rem =
-            ctx.builder.build_int_sub(power_of_two, llvm_val_t.const_int(1, false), "").unwrap();
-        ctx.builder
-            .build_and(
-                ctx.builder.build_int_add(val, max_rem, "").unwrap(),
-                ctx.builder.build_not(max_rem, "").unwrap(),
-                "",
-            )
-            .unwrap()
-    }
-
-    let llvm_i8 = ctx.ctx.i8_type();
-    let llvm_pi8 = llvm_i8.ptr_type(AddressSpace::default());
-    let llvm_usize = ctx.get_size_type();
-    let align_ty = align_ty.into();
-
-    let size = ctx.builder.build_int_truncate_or_bit_cast(size, llvm_usize, "").unwrap();
-
-    debug_assert_eq!(
-        size.get_type().get_bit_width(),
-        llvm_usize.get_bit_width(),
-        "Expected size_t ({}) for parameter `size` of `aligned_alloca`, got {}",
-        llvm_usize,
-        size.get_type(),
-    );
-
-    let alignment = get_type_alignment(align_ty);
-    let alignment = ctx.builder.build_int_truncate_or_bit_cast(alignment, llvm_usize, "").unwrap();
-
-    if ctx.registry.llvm_options.opt_level == OptimizationLevel::None {
-        let alignment_bitcount = llvm_intrinsics::call_int_ctpop(ctx, alignment, None);
-
-        ctx.make_assert(
-            generator,
-            ctx.builder
-                .build_int_compare(
-                    IntPredicate::EQ,
-                    alignment_bitcount,
-                    alignment_bitcount.get_type().const_int(1, false),
-                    "",
-                )
-                .unwrap(),
-            "0:AssertionError",
-            "Expected power-of-two alignment for aligned_alloca, got {0}",
-            [Some(alignment), None, None],
-            ctx.current_loc,
-        );
-    }
-
-    let buffer_size = round_up(ctx, size, alignment);
-    let aligned_slices = ctx.builder.build_int_unsigned_div(buffer_size, alignment, "").unwrap();
-
-    // Just to be absolutely sure, alloca in [i8 x alignment] slices
-    let buffer = ctx.builder.build_array_alloca(align_ty, aligned_slices, "").unwrap();
-
-    ctx.builder
-        .build_bit_cast(buffer, llvm_pi8, name.unwrap_or_default())
-        .map(BasicValueEnum::into_pointer_value)
-        .unwrap()
-}

nac3core/src/codegen/snapshots/nac3core__codegen__test__test_primitives.snap

@@ -1,36 +0,0 @@
----
-source: nac3core/src/codegen/test.rs
-expression: "module.print_to_string().to_str().map(str::trim).unwrap()"
----
-; ModuleID = 'test'
-source_filename = "test"
-target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
-target triple = "x86_64-unknown-linux-gnu"
-
-; Function Attrs: mustprogress nofree norecurse nosync nounwind willreturn memory(none)
-define i32 @testing(i32 %0, i32 %1) local_unnamed_addr #0 !dbg !4 {
-init:
-  %add = add i32 %1, %0, !dbg !9
-  %cmp = icmp eq i32 %add, 1, !dbg !10
-  %. = select i1 %cmp, i32 %0, i32 0, !dbg !11
-  ret i32 %., !dbg !12
-}
-
-attributes #0 = { mustprogress nofree norecurse nosync nounwind willreturn memory(none) }
-
-!llvm.module.flags = !{!0, !1}
-!llvm.dbg.cu = !{!2}
-
-!0 = !{i32 2, !"Debug Info Version", i32 3}
-!1 = !{i32 2, !"Dwarf Version", i32 4}
-!2 = distinct !DICompileUnit(language: DW_LANG_Python, file: !3, producer: "NAC3", isOptimized: true, runtimeVersion: 0, emissionKind: FullDebug)
-!3 = !DIFile(filename: "unknown", directory: "")
-!4 = distinct !DISubprogram(name: "testing", linkageName: "testing", scope: null, file: !3, line: 1, type: !5, scopeLine: 1, flags: DIFlagPublic, spFlags: DISPFlagDefinition | DISPFlagOptimized, unit: !2, retainedNodes: !8)
-!5 = !DISubroutineType(flags: DIFlagPublic, types: !6)
-!6 = !{!7}
-!7 = !DIBasicType(name: "_", flags: DIFlagPublic)
-!8 = !{}
-!9 = !DILocation(line: 1, column: 9, scope: !4)
-!10 = !DILocation(line: 2, column: 15, scope: !4)
-!11 = !DILocation(line: 2, scope: !4)
-!12 = !DILocation(line: 3, column: 8, scope: !4)

nac3core/src/codegen/snapshots/nac3core__codegen__test__test_simple_call.snap

@@ -1,42 +0,0 @@
----
-source: nac3core/src/codegen/test.rs
-expression: "module.print_to_string().to_str().map(str::trim).unwrap()"
----
-; ModuleID = 'test'
-source_filename = "test"
-target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
-target triple = "x86_64-unknown-linux-gnu"
-
-; Function Attrs: mustprogress nofree norecurse nosync nounwind willreturn memory(none)
-define i32 @testing(i32 %0) local_unnamed_addr #0 !dbg !5 {
-init:
-  %add.i = shl i32 %0, 1, !dbg !10
-  %mul = add i32 %add.i, 2, !dbg !10
-  ret i32 %mul, !dbg !10
-}
-
-; Function Attrs: mustprogress nofree norecurse nosync nounwind willreturn memory(none)
-define i32 @foo.0(i32 %0) local_unnamed_addr #0 !dbg !11 {
-init:
-  %add = add i32 %0, 1, !dbg !12
-  ret i32 %add, !dbg !12
-}
-
-attributes #0 = { mustprogress nofree norecurse nosync nounwind willreturn memory(none) }
-
-!llvm.module.flags = !{!0, !1}
-!llvm.dbg.cu = !{!2, !4}
-
-!0 = !{i32 2, !"Debug Info Version", i32 3}
-!1 = !{i32 2, !"Dwarf Version", i32 4}
-!2 = distinct !DICompileUnit(language: DW_LANG_Python, file: !3, producer: "NAC3", isOptimized: true, runtimeVersion: 0, emissionKind: FullDebug)
-!3 = !DIFile(filename: "unknown", directory: "")
-!4 = distinct !DICompileUnit(language: DW_LANG_Python, file: !3, producer: "NAC3", isOptimized: true, runtimeVersion: 0, emissionKind: FullDebug)
-!5 = distinct !DISubprogram(name: "testing", linkageName: "testing", scope: null, file: !3, line: 1, type: !6, scopeLine: 1, flags: DIFlagPublic, spFlags: DISPFlagDefinition | DISPFlagOptimized, unit: !2, retainedNodes: !9)
-!6 = !DISubroutineType(flags: DIFlagPublic, types: !7)
-!7 = !{!8}
-!8 = !DIBasicType(name: "_", flags: DIFlagPublic)
-!9 = !{}
-!10 = !DILocation(line: 2, column: 12, scope: !5)
-!11 = distinct !DISubprogram(name: "foo.0", linkageName: "foo.0", scope: null, file: !3, line: 1, type: !6, scopeLine: 1, flags: DIFlagPublic, spFlags: DISPFlagDefinition | DISPFlagOptimized, unit: !4, retainedNodes: !9)
-!12 = !DILocation(line: 1, column: 12, scope: !11)

nac3core/src/codegen/test.rs

@@ -1,42 +1,39 @@
-use std::{
-    collections::{HashMap, HashSet},
-    sync::Arc,
-};
-
-use function_name::named;
-use indexmap::IndexMap;
-use indoc::indoc;
-use inkwell::{
-    OptimizationLevel,
-    targets::{InitializationConfig, Target},
-};
-use nac3parser::{
-    ast::{FileName, StrRef, fold::Fold},
-    parser::parse_program,
-};
-use parking_lot::RwLock;
-
-use super::{
-    CodeGenContext, CodeGenLLVMOptions, CodeGenTargetMachineOptions, CodeGenTask, CodeGenerator,
-    DefaultCodeGenerator, WithCall, WorkerRegistry,
-    concrete_type::ConcreteTypeStore,
-    types::{ListType, ProxyType, RangeType, ndarray::NDArrayType},
-};
 use crate::{
+    codegen::{
+        classes::{ListType, NDArrayType, ProxyType, RangeType},
+        concrete_type::ConcreteTypeStore,
+        CodeGenContext, CodeGenLLVMOptions, CodeGenTargetMachineOptions, CodeGenTask,
+        CodeGenerator, DefaultCodeGenerator, WithCall, WorkerRegistry,
+    },
     symbol_resolver::{SymbolResolver, ValueEnum},
     toplevel::{
-        DefinitionId, FunInstance, TopLevelContext, TopLevelDef,
         composer::{ComposerConfig, TopLevelComposer},
+        DefinitionId, FunInstance, TopLevelContext, TopLevelDef,
     },
     typecheck::{
-        type_inferencer::{FunctionData, IdentifierInfo, Inferencer, PrimitiveStore},
+        type_inferencer::{FunctionData, Inferencer, PrimitiveStore},
         typedef::{FunSignature, FuncArg, Type, TypeEnum, Unifier, VarMap},
     },
 };
+use indexmap::IndexMap;
+use indoc::indoc;
+use inkwell::{
+    targets::{InitializationConfig, Target},
+    OptimizationLevel,
+};
+use nac3parser::ast::FileName;
+use nac3parser::{
+    ast::{fold::Fold, StrRef},
+    parser::parse_program,
+};
+use parking_lot::RwLock;
+use std::collections::{HashMap, HashSet};
+use std::sync::Arc;
 
 struct Resolver {
     id_to_type: HashMap<StrRef, Type>,
     id_to_def: RwLock<HashMap<StrRef, DefinitionId>>,
+    class_names: HashMap<StrRef, Type>,
 }
 
 impl Resolver {
@@ -67,7 +64,6 @@ impl SymbolResolver for Resolver {
         &self,
         _: StrRef,
         _: &mut CodeGenContext<'ctx, '_>,
-        _: &mut dyn CodeGenerator,
     ) -> Option<ValueEnum<'ctx>> {
         unimplemented!()
     }
@@ -90,7 +86,6 @@ impl SymbolResolver for Resolver {
 }
 
 #[test]
-#[named]
 fn test_primitives() {
     let source = indoc! { "
         c = a + b
@@ -99,32 +94,23 @@ fn test_primitives() {
         "};
     let statements = parse_program(source, FileName::default()).unwrap();
 
-    let context = inkwell::context::Context::create();
-    let composer = TopLevelComposer::new(Vec::new(), Vec::new(), ComposerConfig::default(), 64).0;
+    let composer = TopLevelComposer::new(Vec::new(), ComposerConfig::default(), 32).0;
     let mut unifier = composer.unifier.clone();
     let primitives = composer.primitives_ty;
     let top_level = Arc::new(composer.make_top_level_context());
     unifier.top_level = Some(top_level.clone());
 
-    let resolver =
-        Arc::new(Resolver { id_to_type: HashMap::new(), id_to_def: RwLock::new(HashMap::new()) })
-            as Arc<dyn SymbolResolver + Send + Sync>;
+    let resolver = Arc::new(Resolver {
+        id_to_type: HashMap::new(),
+        id_to_def: RwLock::new(HashMap::new()),
+        class_names: HashMap::default(),
+    }) as Arc<dyn SymbolResolver + Send + Sync>;
 
-    let threads = vec![DefaultCodeGenerator::new("test".into(), context.i64_type()).into()];
+    let threads = vec![DefaultCodeGenerator::new("test".into(), 32).into()];
     let signature = FunSignature {
         args: vec![
-            FuncArg {
-                name: "a".into(),
-                ty: primitives.int32,
-                default_value: None,
-                is_vararg: false,
-            },
-            FuncArg {
-                name: "b".into(),
-                ty: primitives.int32,
-                default_value: None,
-                is_vararg: false,
-            },
+            FuncArg { name: "a".into(), ty: primitives.int32, default_value: None },
+            FuncArg { name: "b".into(), ty: primitives.int32, default_value: None },
         ],
         ret: primitives.int32,
         vars: VarMap::new(),
@@ -142,8 +128,7 @@ fn test_primitives() {
     };
     let mut virtual_checks = Vec::new();
     let mut calls = HashMap::new();
-    let mut identifiers: HashMap<_, _> =
-        ["a".into(), "b".into()].map(|id| (id, IdentifierInfo::default())).into();
+    let mut identifiers: HashSet<_> = ["a".into(), "b".into()].into();
     let mut inferencer = Inferencer {
         top_level: &top_level,
         function_data: &mut function_data,
@@ -183,10 +168,58 @@ fn test_primitives() {
         id: 0,
     };
     let f = Arc::new(WithCall::new(Box::new(|module| {
-        insta::assert_snapshot!(
-            function_name!(),
-            module.print_to_string().to_str().map(str::trim).unwrap()
-        );
+        // the following IR is equivalent to
+        // ```
+        // ; ModuleID = 'test.ll'
+        // source_filename = "test"
+        //
+        // ; Function Attrs: norecurse nounwind readnone
+        // define i32 @testing(i32 %0, i32 %1) local_unnamed_addr #0 {
+        // init:
+        //   %add = add i32 %1, %0
+        //   %cmp = icmp eq i32 %add, 1
+        //   %ifexpr = select i1 %cmp, i32 %0, i32 0
+        //   ret i32 %ifexpr
+        // }
+        //
+        // attributes #0 = { norecurse nounwind readnone }
+        // ```
+        // after O2 optimization
+
+        let expected = indoc! {"
+            ; ModuleID = 'test'
+            source_filename = \"test\"
+
+            ; Function Attrs: mustprogress nofree norecurse nosync nounwind readnone willreturn
+            define i32 @testing(i32 %0, i32 %1) local_unnamed_addr #0 !dbg !4 {
+            init:
+              %add = add i32 %1, %0, !dbg !9
+              %cmp = icmp eq i32 %add, 1, !dbg !10
+              %. = select i1 %cmp, i32 %0, i32 0, !dbg !11
+              ret i32 %., !dbg !12
+            }
+
+            attributes #0 = { mustprogress nofree norecurse nosync nounwind readnone willreturn }
+
+            !llvm.module.flags = !{!0, !1}
+            !llvm.dbg.cu = !{!2}
+
+            !0 = !{i32 2, !\"Debug Info Version\", i32 3}
+            !1 = !{i32 2, !\"Dwarf Version\", i32 4}
+            !2 = distinct !DICompileUnit(language: DW_LANG_Python, file: !3, producer: \"NAC3\", isOptimized: true, runtimeVersion: 0, emissionKind: FullDebug)
+            !3 = !DIFile(filename: \"unknown\", directory: \"\")
+            !4 = distinct !DISubprogram(name: \"testing\", linkageName: \"testing\", scope: null, file: !3, line: 1, type: !5, scopeLine: 1, flags: DIFlagPublic, spFlags: DISPFlagDefinition | DISPFlagOptimized, unit: !2, retainedNodes: !8)
+            !5 = !DISubroutineType(flags: DIFlagPublic, types: !6)
+            !6 = !{!7}
+            !7 = !DIBasicType(name: \"_\", flags: DIFlagPublic)
+            !8 = !{}
+            !9 = !DILocation(line: 1, column: 9, scope: !4)
+            !10 = !DILocation(line: 2, column: 15, scope: !4)
+            !11 = !DILocation(line: 0, scope: !4)
+            !12 = !DILocation(line: 3, column: 8, scope: !4)
+        "}
+        .trim();
+        assert_eq!(expected, module.print_to_string().to_str().unwrap().trim());
     })));
 
     Target::initialize_all(&InitializationConfig::default());
@@ -201,7 +234,6 @@ fn test_primitives() {
 }
 
 #[test]
-#[named]
 fn test_simple_call() {
     let source_1 = indoc! { "
         a = foo(a)
@@ -214,20 +246,14 @@ fn test_simple_call() {
         "};
     let statements_2 = parse_program(source_2, FileName::default()).unwrap();
 
-    let context = inkwell::context::Context::create();
-    let composer = TopLevelComposer::new(Vec::new(), Vec::new(), ComposerConfig::default(), 64).0;
+    let composer = TopLevelComposer::new(Vec::new(), ComposerConfig::default(), 32).0;
     let mut unifier = composer.unifier.clone();
     let primitives = composer.primitives_ty;
     let top_level = Arc::new(composer.make_top_level_context());
     unifier.top_level = Some(top_level.clone());
 
     let signature = FunSignature {
-        args: vec![FuncArg {
-            name: "a".into(),
-            ty: primitives.int32,
-            default_value: None,
-            is_vararg: false,
-        }],
+        args: vec![FuncArg { name: "a".into(), ty: primitives.int32, default_value: None }],
         ret: primitives.int32,
         vars: VarMap::new(),
     };
@@ -250,7 +276,11 @@ fn test_simple_call() {
         loc: None,
     })));
 
-    let resolver = Resolver { id_to_type: HashMap::new(), id_to_def: RwLock::new(HashMap::new()) };
+    let resolver = Resolver {
+        id_to_type: HashMap::new(),
+        id_to_def: RwLock::new(HashMap::new()),
+        class_names: HashMap::default(),
+    };
     resolver.add_id_def("foo".into(), DefinitionId(foo_id));
     let resolver = Arc::new(resolver) as Arc<dyn SymbolResolver + Send + Sync>;
 
@@ -262,7 +292,7 @@ fn test_simple_call() {
         unreachable!()
     }
 
-    let threads = vec![DefaultCodeGenerator::new("test".into(), context.i64_type()).into()];
+    let threads = vec![DefaultCodeGenerator::new("test".into(), 32).into()];
     let mut function_data = FunctionData {
         resolver: resolver.clone(),
         bound_variables: Vec::new(),
@@ -270,8 +300,7 @@ fn test_simple_call() {
     };
     let mut virtual_checks = Vec::new();
     let mut calls = HashMap::new();
-    let mut identifiers: HashMap<_, _> =
-        ["a".into(), "foo".into()].map(|id| (id, IdentifierInfo::default())).into();
+    let mut identifiers: HashSet<_> = ["a".into(), "foo".into()].into();
     let mut inferencer = Inferencer {
         top_level: &top_level,
         function_data: &mut function_data,
@@ -336,10 +365,46 @@ fn test_simple_call() {
         id: 0,
     };
     let f = Arc::new(WithCall::new(Box::new(|module| {
-        insta::assert_snapshot!(
-            function_name!(),
-            module.print_to_string().to_str().map(str::trim).unwrap()
-        );
+        let expected = indoc! {"
+            ; ModuleID = 'test'
+            source_filename = \"test\"
+
+            ; Function Attrs: mustprogress nofree norecurse nosync nounwind readnone willreturn
+            define i32 @testing(i32 %0) local_unnamed_addr #0 !dbg !5 {
+            init:
+              %add.i = shl i32 %0, 1, !dbg !10
+              %mul = add i32 %add.i, 2, !dbg !10
+              ret i32 %mul, !dbg !10
+            }
+
+            ; Function Attrs: mustprogress nofree norecurse nosync nounwind readnone willreturn
+            define i32 @foo.0(i32 %0) local_unnamed_addr #0 !dbg !11 {
+            init:
+              %add = add i32 %0, 1, !dbg !12
+              ret i32 %add, !dbg !12
+            }
+
+            attributes #0 = { mustprogress nofree norecurse nosync nounwind readnone willreturn }
+
+            !llvm.module.flags = !{!0, !1}
+            !llvm.dbg.cu = !{!2, !4}
+
+            !0 = !{i32 2, !\"Debug Info Version\", i32 3}
+            !1 = !{i32 2, !\"Dwarf Version\", i32 4}
+            !2 = distinct !DICompileUnit(language: DW_LANG_Python, file: !3, producer: \"NAC3\", isOptimized: true, runtimeVersion: 0, emissionKind: FullDebug)
+            !3 = !DIFile(filename: \"unknown\", directory: \"\")
+            !4 = distinct !DICompileUnit(language: DW_LANG_Python, file: !3, producer: \"NAC3\", isOptimized: true, runtimeVersion: 0, emissionKind: FullDebug)
+            !5 = distinct !DISubprogram(name: \"testing\", linkageName: \"testing\", scope: null, file: !3, line: 1, type: !6, scopeLine: 1, flags: DIFlagPublic, spFlags: DISPFlagDefinition | DISPFlagOptimized, unit: !2, retainedNodes: !9)
+            !6 = !DISubroutineType(flags: DIFlagPublic, types: !7)
+            !7 = !{!8}
+            !8 = !DIBasicType(name: \"_\", flags: DIFlagPublic)
+            !9 = !{}
+            !10 = !DILocation(line: 2, column: 12, scope: !5)
+            !11 = distinct !DISubprogram(name: \"foo.0\", linkageName: \"foo.0\", scope: null, file: !3, line: 1, type: !6, scopeLine: 1, flags: DIFlagPublic, spFlags: DISPFlagDefinition | DISPFlagOptimized, unit: !4, retainedNodes: !9)
+            !12 = !DILocation(line: 1, column: 12, scope: !11)
+        "}
+        .trim();
+        assert_eq!(expected, module.print_to_string().to_str().unwrap().trim());
     })));
 
     Target::initialize_all(&InitializationConfig::default());
@@ -356,34 +421,31 @@ fn test_simple_call() {
 #[test]
 fn test_classes_list_type_new() {
     let ctx = inkwell::context::Context::create();
-    let generator = DefaultCodeGenerator::new(String::new(), ctx.i64_type());
+    let generator = DefaultCodeGenerator::new(String::new(), 64);
 
     let llvm_i32 = ctx.i32_type();
     let llvm_usize = generator.get_size_type(&ctx);
 
-    let llvm_list = ListType::new_with_generator(&generator, &ctx, llvm_i32.into());
-    assert!(ListType::is_representable(llvm_list.as_abi_type(), llvm_usize).is_ok());
+    let llvm_list = ListType::new(&generator, &ctx, llvm_i32.into());
+    assert!(ListType::is_type(llvm_list.as_base_type(), llvm_usize).is_ok());
 }
 
 #[test]
 fn test_classes_range_type_new() {
     let ctx = inkwell::context::Context::create();
-    let generator = DefaultCodeGenerator::new(String::new(), ctx.i64_type());
 
-    let llvm_usize = generator.get_size_type(&ctx);
-
-    let llvm_range = RangeType::new_with_generator(&generator, &ctx);
-    assert!(RangeType::is_representable(llvm_range.as_abi_type(), llvm_usize).is_ok());
+    let llvm_range = RangeType::new(&ctx);
+    assert!(RangeType::is_type(llvm_range.as_base_type()).is_ok());
 }
 
 #[test]
 fn test_classes_ndarray_type_new() {
     let ctx = inkwell::context::Context::create();
-    let generator = DefaultCodeGenerator::new(String::new(), ctx.i64_type());
+    let generator = DefaultCodeGenerator::new(String::new(), 64);
 
     let llvm_i32 = ctx.i32_type();
     let llvm_usize = generator.get_size_type(&ctx);
 
-    let llvm_ndarray = NDArrayType::new_with_generator(&generator, &ctx, llvm_i32.into(), 2);
-    assert!(NDArrayType::is_representable(llvm_ndarray.as_abi_type(), llvm_usize).is_ok());
+    let llvm_ndarray = NDArrayType::new(&generator, &ctx, llvm_i32.into());
+    assert!(NDArrayType::is_type(llvm_ndarray.as_base_type(), llvm_usize).is_ok());
 }

nac3core/src/codegen/types/exception.rs

@@ -1,264 +0,0 @@
-use inkwell::{
-    AddressSpace,
-    context::{AsContextRef, Context},
-    types::{AnyTypeEnum, BasicType, BasicTypeEnum, IntType, PointerType, StructType},
-    values::{IntValue, PointerValue, StructValue},
-};
-use itertools::Itertools;
-
-use nac3core_derive::StructFields;
-
-use super::{
-    ProxyType,
-    structure::{StructField, StructFields, StructProxyType, check_struct_type_matches_fields},
-};
-use crate::{
-    codegen::{CodeGenContext, CodeGenerator, values::ExceptionValue},
-    typecheck::typedef::{Type, TypeEnum},
-};
-
-/// Proxy type for an `Exception` in LLVM.
-#[derive(Debug, PartialEq, Eq, Clone, Copy)]
-pub struct ExceptionType<'ctx> {
-    ty: PointerType<'ctx>,
-    llvm_usize: IntType<'ctx>,
-}
-
-#[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-pub struct ExceptionStructFields<'ctx> {
-    /// The ID of the exception name.
-    #[value_type(i32_type())]
-    pub name: StructField<'ctx, IntValue<'ctx>>,
-
-    /// The file where the exception originated from.
-    #[value_type(get_struct_type("str").unwrap())]
-    pub file: StructField<'ctx, StructValue<'ctx>>,
-
-    /// The line number where the exception originated from.
-    #[value_type(i32_type())]
-    pub line: StructField<'ctx, IntValue<'ctx>>,
-
-    /// The column number where the exception originated from.
-    #[value_type(i32_type())]
-    pub col: StructField<'ctx, IntValue<'ctx>>,
-
-    /// The function name where the exception originated from.
-    #[value_type(get_struct_type("str").unwrap())]
-    pub func: StructField<'ctx, StructValue<'ctx>>,
-
-    /// The exception message.
-    #[value_type(get_struct_type("str").unwrap())]
-    pub message: StructField<'ctx, StructValue<'ctx>>,
-
-    #[value_type(i64_type())]
-    pub param0: StructField<'ctx, IntValue<'ctx>>,
-
-    #[value_type(i64_type())]
-    pub param1: StructField<'ctx, IntValue<'ctx>>,
-
-    #[value_type(i64_type())]
-    pub param2: StructField<'ctx, IntValue<'ctx>>,
-}
-
-impl<'ctx> ExceptionType<'ctx> {
-    /// Returns an instance of [`StructFields`] containing all field accessors for this type.
-    #[must_use]
-    fn fields(
-        ctx: impl AsContextRef<'ctx>,
-        llvm_usize: IntType<'ctx>,
-    ) -> ExceptionStructFields<'ctx> {
-        ExceptionStructFields::new(ctx, llvm_usize)
-    }
-
-    /// Creates an LLVM type corresponding to the expected structure of an `Exception`.
-    #[must_use]
-    fn llvm_type(ctx: &'ctx Context, llvm_usize: IntType<'ctx>) -> PointerType<'ctx> {
-        const NAME: &str = "Exception";
-
-        assert!(ctx.get_struct_type("str").is_some());
-
-        if let Some(t) = ctx.get_struct_type(NAME) {
-            t.ptr_type(AddressSpace::default())
-        } else {
-            let exn_ty = ctx.opaque_struct_type(NAME);
-            let field_tys =
-                Self::fields(ctx, llvm_usize).into_iter().map(|field| field.1).collect_vec();
-            exn_ty.set_body(&field_tys, false);
-            exn_ty.ptr_type(AddressSpace::default())
-        }
-    }
-
-    fn new_impl(ctx: &'ctx Context, llvm_usize: IntType<'ctx>) -> Self {
-        let llvm_str = Self::llvm_type(ctx, llvm_usize);
-
-        Self { ty: llvm_str, llvm_usize }
-    }
-
-    /// Creates an instance of [`ExceptionType`].
-    #[must_use]
-    pub fn new(ctx: &CodeGenContext<'ctx, '_>) -> Self {
-        Self::new_impl(ctx.ctx, ctx.get_size_type())
-    }
-
-    /// Creates an instance of [`ExceptionType`].
-    #[must_use]
-    pub fn new_with_generator<G: CodeGenerator + ?Sized>(
-        generator: &G,
-        ctx: &'ctx Context,
-    ) -> Self {
-        Self::new_impl(ctx, generator.get_size_type(ctx))
-    }
-
-    /// Creates an [`ExceptionType`] from a [unifier type][Type].
-    #[must_use]
-    pub fn from_unifier_type(ctx: &mut CodeGenContext<'ctx, '_>, ty: Type) -> Self {
-        // Check unifier type
-        assert!(
-            matches!(&*ctx.unifier.get_ty_immutable(ty), TypeEnum::TObj { obj_id, .. } if *obj_id == ctx.primitives.exception.obj_id(&ctx.unifier).unwrap())
-        );
-
-        Self::new_impl(ctx.ctx, ctx.get_size_type())
-    }
-
-    /// Creates an [`ExceptionType`] from a [`StructType`] representing an `Exception`.
-    #[must_use]
-    pub fn from_struct_type(ty: StructType<'ctx>, llvm_usize: IntType<'ctx>) -> Self {
-        Self::from_pointer_type(ty.ptr_type(AddressSpace::default()), llvm_usize)
-    }
-
-    /// Creates an [`ExceptionType`] from a [`PointerType`] representing an `Exception`.
-    #[must_use]
-    pub fn from_pointer_type(ptr_ty: PointerType<'ctx>, llvm_usize: IntType<'ctx>) -> Self {
-        debug_assert!(Self::has_same_repr(ptr_ty, llvm_usize).is_ok());
-
-        Self { ty: ptr_ty, llvm_usize }
-    }
-
-    /// Returns an instance of [`ExceptionType`] by obtaining the LLVM representation of the builtin
-    /// `Exception` type.
-    #[must_use]
-    pub fn get_instance<G: CodeGenerator + ?Sized>(
-        generator: &mut G,
-        ctx: &mut CodeGenContext<'ctx, '_>,
-    ) -> Self {
-        Self::from_pointer_type(
-            ctx.get_llvm_type(generator, ctx.primitives.exception).into_pointer_type(),
-            ctx.get_size_type(),
-        )
-    }
-
-    /// Allocates an instance of [`ExceptionValue`] as if by calling `alloca` on the base type.
-    ///
-    /// See [`ProxyType::raw_alloca`].
-    #[must_use]
-    pub fn alloca(
-        &self,
-        ctx: &mut CodeGenContext<'ctx, '_>,
-        name: Option<&'ctx str>,
-    ) -> <Self as ProxyType<'ctx>>::Value {
-        <Self as ProxyType<'ctx>>::Value::from_pointer_value(
-            self.raw_alloca(ctx, name),
-            self.llvm_usize,
-            name,
-        )
-    }
-
-    /// Allocates an instance of [`ExceptionValue`] as if by calling `alloca` on the base type.
-    ///
-    /// See [`ProxyType::raw_alloca_var`].
-    #[must_use]
-    pub fn alloca_var<G: CodeGenerator + ?Sized>(
-        &self,
-        generator: &mut G,
-        ctx: &mut CodeGenContext<'ctx, '_>,
-        name: Option<&'ctx str>,
-    ) -> <Self as ProxyType<'ctx>>::Value {
-        <Self as ProxyType<'ctx>>::Value::from_pointer_value(
-            self.raw_alloca_var(generator, ctx, name),
-            self.llvm_usize,
-            name,
-        )
-    }
-
-    /// Converts an existing value into a [`ExceptionValue`].
-    #[must_use]
-    pub fn map_struct_value<G: CodeGenerator + ?Sized>(
-        &self,
-        generator: &mut G,
-        ctx: &mut CodeGenContext<'ctx, '_>,
-        value: StructValue<'ctx>,
-        name: Option<&'ctx str>,
-    ) -> <Self as ProxyType<'ctx>>::Value {
-        <Self as ProxyType<'ctx>>::Value::from_struct_value(
-            generator,
-            ctx,
-            value,
-            self.llvm_usize,
-            name,
-        )
-    }
-
-    /// Converts an existing value into a [`ExceptionValue`].
-    #[must_use]
-    pub fn map_pointer_value(
-        &self,
-        value: PointerValue<'ctx>,
-        name: Option<&'ctx str>,
-    ) -> <Self as ProxyType<'ctx>>::Value {
-        <Self as ProxyType<'ctx>>::Value::from_pointer_value(value, self.llvm_usize, name)
-    }
-}
-
-impl<'ctx> ProxyType<'ctx> for ExceptionType<'ctx> {
-    type ABI = PointerType<'ctx>;
-    type Base = PointerType<'ctx>;
-    type Value = ExceptionValue<'ctx>;
-
-    fn is_representable(
-        llvm_ty: impl BasicType<'ctx>,
-        llvm_usize: IntType<'ctx>,
-    ) -> Result<(), String> {
-        if let BasicTypeEnum::PointerType(ty) = llvm_ty.as_basic_type_enum() {
-            Self::has_same_repr(ty, llvm_usize)
-        } else {
-            Err(format!("Expected pointer type, got {llvm_ty:?}"))
-        }
-    }
-
-    fn has_same_repr(ty: Self::Base, llvm_usize: IntType<'ctx>) -> Result<(), String> {
-        let ctx = ty.get_context();
-
-        let llvm_ty = ty.get_element_type();
-        let AnyTypeEnum::StructType(llvm_ty) = llvm_ty else {
-            return Err(format!("Expected struct type for `list` type, got {llvm_ty}"));
-        };
-
-        check_struct_type_matches_fields(Self::fields(ctx, llvm_usize), llvm_ty, "exception", &[])
-    }
-
-    fn alloca_type(&self) -> impl BasicType<'ctx> {
-        self.as_abi_type().get_element_type().into_struct_type()
-    }
-
-    fn as_base_type(&self) -> Self::Base {
-        self.ty
-    }
-
-    fn as_abi_type(&self) -> Self::ABI {
-        self.as_base_type()
-    }
-}
-
-impl<'ctx> StructProxyType<'ctx> for ExceptionType<'ctx> {
-    type StructFields = ExceptionStructFields<'ctx>;
-
-    fn get_fields(&self) -> Self::StructFields {
-        Self::fields(self.ty.get_context(), self.llvm_usize)
-    }
-}
-
-impl<'ctx> From<ExceptionType<'ctx>> for PointerType<'ctx> {
-    fn from(value: ExceptionType<'ctx>) -> Self {
-        value.as_base_type()
-    }
-}

nac3core/src/codegen/types/list.rs

@@ -1,390 +0,0 @@
-use inkwell::{
-    AddressSpace, IntPredicate, OptimizationLevel,
-    context::Context,
-    types::{AnyTypeEnum, BasicType, BasicTypeEnum, IntType, PointerType, StructType},
-    values::{IntValue, PointerValue, StructValue},
-};
-use itertools::Itertools;
-
-use nac3core_derive::StructFields;
-
-use super::ProxyType;
-use crate::{
-    codegen::{
-        CodeGenContext, CodeGenerator,
-        types::structure::{
-            FieldIndexCounter, StructField, StructFields, StructProxyType,
-            check_struct_type_matches_fields,
-        },
-        values::ListValue,
-    },
-    typecheck::typedef::{Type, TypeEnum, iter_type_vars},
-};
-
-/// Proxy type for a `list` type in LLVM.
-#[derive(Debug, PartialEq, Eq, Clone, Copy)]
-pub struct ListType<'ctx> {
-    ty: PointerType<'ctx>,
-    item: Option<BasicTypeEnum<'ctx>>,
-    llvm_usize: IntType<'ctx>,
-}
-
-#[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-pub struct ListStructFields<'ctx> {
-    /// Array pointer to content.
-    #[value_type(i8_type().ptr_type(AddressSpace::default()))]
-    pub items: StructField<'ctx, PointerValue<'ctx>>,
-
-    /// Number of items in the array.
-    #[value_type(usize)]
-    pub len: StructField<'ctx, IntValue<'ctx>>,
-}
-
-impl<'ctx> ListStructFields<'ctx> {
-    #[must_use]
-    pub fn new_typed(item: BasicTypeEnum<'ctx>, llvm_usize: IntType<'ctx>) -> Self {
-        let mut counter = FieldIndexCounter::default();
-
-        ListStructFields {
-            items: StructField::create(
-                &mut counter,
-                "items",
-                item.ptr_type(AddressSpace::default()),
-            ),
-            len: StructField::create(&mut counter, "len", llvm_usize),
-        }
-    }
-}
-
-impl<'ctx> ListType<'ctx> {
-    /// Returns an instance of [`StructFields`] containing all field accessors for this type.
-    #[must_use]
-    fn fields(item: BasicTypeEnum<'ctx>, llvm_usize: IntType<'ctx>) -> ListStructFields<'ctx> {
-        ListStructFields::new_typed(item, llvm_usize)
-    }
-
-    /// Creates an LLVM type corresponding to the expected structure of a `List`.
-    #[must_use]
-    fn llvm_type(
-        ctx: &'ctx Context,
-        element_type: Option<BasicTypeEnum<'ctx>>,
-        llvm_usize: IntType<'ctx>,
-    ) -> PointerType<'ctx> {
-        let element_type = element_type.map_or(llvm_usize.into(), |ty| ty.as_basic_type_enum());
-
-        let field_tys =
-            Self::fields(element_type, llvm_usize).into_iter().map(|field| field.1).collect_vec();
-
-        ctx.struct_type(&field_tys, false).ptr_type(AddressSpace::default())
-    }
-
-    fn new_impl(
-        ctx: &'ctx Context,
-        element_type: Option<BasicTypeEnum<'ctx>>,
-        llvm_usize: IntType<'ctx>,
-    ) -> Self {
-        let llvm_list = Self::llvm_type(ctx, element_type, llvm_usize);
-
-        Self { ty: llvm_list, item: element_type, llvm_usize }
-    }
-
-    /// Creates an instance of [`ListType`].
-    #[must_use]
-    pub fn new(ctx: &CodeGenContext<'ctx, '_>, element_type: &impl BasicType<'ctx>) -> Self {
-        Self::new_impl(ctx.ctx, Some(element_type.as_basic_type_enum()), ctx.get_size_type())
-    }
-
-    /// Creates an instance of [`ListType`].
-    #[must_use]
-    pub fn new_with_generator<G: CodeGenerator + ?Sized>(
-        generator: &G,
-        ctx: &'ctx Context,
-        element_type: BasicTypeEnum<'ctx>,
-    ) -> Self {
-        Self::new_impl(ctx, Some(element_type.as_basic_type_enum()), generator.get_size_type(ctx))
-    }
-
-    /// Creates an instance of [`ListType`] with an unknown element type.
-    #[must_use]
-    pub fn new_untyped(ctx: &CodeGenContext<'ctx, '_>) -> Self {
-        Self::new_impl(ctx.ctx, None, ctx.get_size_type())
-    }
-
-    /// Creates an instance of [`ListType`] with an unknown element type.
-    #[must_use]
-    pub fn new_untyped_with_generator<G: CodeGenerator + ?Sized>(
-        generator: &G,
-        ctx: &'ctx Context,
-    ) -> Self {
-        Self::new_impl(ctx, None, generator.get_size_type(ctx))
-    }
-
-    /// Creates an [`ListType`] from a [unifier type][Type].
-    #[must_use]
-    pub fn from_unifier_type<G: CodeGenerator + ?Sized>(
-        generator: &G,
-        ctx: &mut CodeGenContext<'ctx, '_>,
-        ty: Type,
-    ) -> Self {
-        // Check unifier type and extract `item_type`
-        let elem_type = match &*ctx.unifier.get_ty_immutable(ty) {
-            TypeEnum::TObj { obj_id, params, .. }
-                if *obj_id == ctx.primitives.list.obj_id(&ctx.unifier).unwrap() =>
-            {
-                iter_type_vars(params).next().unwrap().ty
-            }
-
-            _ => panic!("Expected `list` type, but got {}", ctx.unifier.stringify(ty)),
-        };
-
-        let llvm_usize = ctx.get_size_type();
-        let llvm_elem_type = if let TypeEnum::TVar { .. } = &*ctx.unifier.get_ty_immutable(ty) {
-            None
-        } else {
-            Some(ctx.get_llvm_type(generator, elem_type))
-        };
-
-        Self::new_impl(ctx.ctx, llvm_elem_type, llvm_usize)
-    }
-
-    /// Creates an [`ListType`] from a [`StructType`].
-    #[must_use]
-    pub fn from_struct_type(ty: StructType<'ctx>, llvm_usize: IntType<'ctx>) -> Self {
-        Self::from_pointer_type(ty.ptr_type(AddressSpace::default()), llvm_usize)
-    }
-
-    /// Creates an [`ListType`] from a [`PointerType`].
-    #[must_use]
-    pub fn from_pointer_type(ptr_ty: PointerType<'ctx>, llvm_usize: IntType<'ctx>) -> Self {
-        debug_assert!(Self::has_same_repr(ptr_ty, llvm_usize).is_ok());
-
-        let ctx = ptr_ty.get_context();
-
-        // We are just searching for the index off a field - Slot an arbitrary element type in.
-        let item_field_idx =
-            Self::fields(ctx.i8_type().into(), llvm_usize).index_of_field(|f| f.items);
-        let item = unsafe {
-            ptr_ty
-                .get_element_type()
-                .into_struct_type()
-                .get_field_type_at_index_unchecked(item_field_idx)
-                .into_pointer_type()
-                .get_element_type()
-        };
-        let item = BasicTypeEnum::try_from(item).unwrap_or_else(|()| {
-            panic!(
-                "Expected BasicTypeEnum for list element type, got {}",
-                ptr_ty.get_element_type().print_to_string()
-            )
-        });
-
-        ListType { ty: ptr_ty, item: Some(item), llvm_usize }
-    }
-
-    /// Returns the type of the `size` field of this `list` type.
-    #[must_use]
-    pub fn size_type(&self) -> IntType<'ctx> {
-        self.llvm_usize
-    }
-
-    /// Returns the element type of this `list` type.
-    #[must_use]
-    pub fn element_type(&self) -> Option<BasicTypeEnum<'ctx>> {
-        self.item
-    }
-
-    /// Allocates an instance of [`ListValue`] as if by calling `alloca` on the base type.
-    ///
-    /// See [`ProxyType::raw_alloca`].
-    #[must_use]
-    pub fn alloca(
-        &self,
-        ctx: &mut CodeGenContext<'ctx, '_>,
-        name: Option<&'ctx str>,
-    ) -> <Self as ProxyType<'ctx>>::Value {
-        <Self as ProxyType<'ctx>>::Value::from_pointer_value(
-            self.raw_alloca(ctx, name),
-            self.llvm_usize,
-            name,
-        )
-    }
-
-    /// Allocates an instance of [`ListValue`] as if by calling `alloca` on the base type.
-    ///
-    /// See [`ProxyType::raw_alloca_var`].
-    #[must_use]
-    pub fn alloca_var<G: CodeGenerator + ?Sized>(
-        &self,
-        generator: &mut G,
-        ctx: &mut CodeGenContext<'ctx, '_>,
-        name: Option<&'ctx str>,
-    ) -> <Self as ProxyType<'ctx>>::Value {
-        <Self as ProxyType<'ctx>>::Value::from_pointer_value(
-            self.raw_alloca_var(generator, ctx, name),
-            self.llvm_usize,
-            name,
-        )
-    }
-
-    /// Allocates a [`ListValue`] on the stack using `item` of this [`ListType`] instance.
-    ///
-    /// The returned list will contain:
-    ///
-    /// - `data`: Allocated with `len` number of elements.
-    /// - `len`: Initialized to the value of `len` passed to this function.
-    #[must_use]
-    pub fn construct<G: CodeGenerator + ?Sized>(
-        &self,
-        generator: &mut G,
-        ctx: &mut CodeGenContext<'ctx, '_>,
-        len: IntValue<'ctx>,
-        name: Option<&'ctx str>,
-    ) -> <Self as ProxyType<'ctx>>::Value {
-        let len = ctx.builder.build_int_z_extend(len, self.llvm_usize, "").unwrap();
-
-        // Generate a runtime assertion if allocating a non-empty list with unknown element type
-        if ctx.registry.llvm_options.opt_level == OptimizationLevel::None && self.item.is_none() {
-            let len_eqz = ctx
-                .builder
-                .build_int_compare(IntPredicate::EQ, len, self.llvm_usize.const_zero(), "")
-                .unwrap();
-
-            ctx.make_assert(
-                generator,
-                len_eqz,
-                "0:AssertionError",
-                "Cannot allocate a non-empty list with unknown element type",
-                [None, None, None],
-                ctx.current_loc,
-            );
-        }
-
-        let plist = self.alloca_var(generator, ctx, name);
-        plist.store_size(ctx, len);
-
-        let item = self.item.unwrap_or(self.llvm_usize.into());
-        plist.create_data(ctx, item, None);
-
-        plist
-    }
-
-    /// Convenience function for creating a list with zero elements.
-    ///
-    /// This function is preferred over [`ListType::construct`] if the length is known to always be
-    /// 0, as this function avoids injecting an IR assertion for checking if a non-empty untyped
-    /// list is being allocated.
-    ///
-    /// The returned list will contain:
-    ///
-    /// - `data`: Initialized to `(T*) 0`.
-    /// - `len`: Initialized to `0`.
-    #[must_use]
-    pub fn construct_empty<G: CodeGenerator + ?Sized>(
-        &self,
-        generator: &mut G,
-        ctx: &mut CodeGenContext<'ctx, '_>,
-        name: Option<&'ctx str>,
-    ) -> <Self as ProxyType<'ctx>>::Value {
-        let plist = self.alloca_var(generator, ctx, name);
-
-        plist.store_size(ctx, self.llvm_usize.const_zero());
-        plist.create_data(ctx, self.item.unwrap_or(self.llvm_usize.into()), None);
-
-        plist
-    }
-
-    /// Converts an existing value into a [`ListValue`].
-    #[must_use]
-    pub fn map_struct_value<G: CodeGenerator + ?Sized>(
-        &self,
-        generator: &mut G,
-        ctx: &mut CodeGenContext<'ctx, '_>,
-        value: StructValue<'ctx>,
-        name: Option<&'ctx str>,
-    ) -> <Self as ProxyType<'ctx>>::Value {
-        <Self as ProxyType<'ctx>>::Value::from_struct_value(
-            generator,
-            ctx,
-            value,
-            self.llvm_usize,
-            name,
-        )
-    }
-
-    /// Converts an existing value into a [`ListValue`].
-    #[must_use]
-    pub fn map_pointer_value(
-        &self,
-        value: PointerValue<'ctx>,
-        name: Option<&'ctx str>,
-    ) -> <Self as ProxyType<'ctx>>::Value {
-        <Self as ProxyType<'ctx>>::Value::from_pointer_value(value, self.llvm_usize, name)
-    }
-}
-
-impl<'ctx> ProxyType<'ctx> for ListType<'ctx> {
-    type ABI = PointerType<'ctx>;
-    type Base = PointerType<'ctx>;
-    type Value = ListValue<'ctx>;
-
-    fn is_representable(
-        llvm_ty: impl BasicType<'ctx>,
-        llvm_usize: IntType<'ctx>,
-    ) -> Result<(), String> {
-        if let BasicTypeEnum::PointerType(ty) = llvm_ty.as_basic_type_enum() {
-            Self::has_same_repr(ty, llvm_usize)
-        } else {
-            Err(format!("Expected pointer type, got {llvm_ty:?}"))
-        }
-    }
-
-    fn has_same_repr(ty: Self::Base, llvm_usize: IntType<'ctx>) -> Result<(), String> {
-        let ctx = ty.get_context();
-
-        let llvm_ty = ty.get_element_type();
-        let AnyTypeEnum::StructType(llvm_ty) = llvm_ty else {
-            return Err(format!("Expected struct type for `list` type, got {llvm_ty}"));
-        };
-
-        let fields = ListStructFields::new(ctx, llvm_usize);
-
-        check_struct_type_matches_fields(
-            fields,
-            llvm_ty,
-            "list",
-            &[(fields.items.name(), &|ty| {
-                if ty.is_pointer_type() {
-                    Ok(())
-                } else {
-                    Err(format!("Expected T* for `list.items`, got {ty}"))
-                }
-            })],
-        )
-    }
-
-    fn alloca_type(&self) -> impl BasicType<'ctx> {
-        self.as_abi_type().get_element_type().into_struct_type()
-    }
-
-    fn as_base_type(&self) -> Self::Base {
-        self.ty
-    }
-
-    fn as_abi_type(&self) -> Self::ABI {
-        self.as_base_type()
-    }
-}
-
-impl<'ctx> StructProxyType<'ctx> for ListType<'ctx> {
-    type StructFields = ListStructFields<'ctx>;
-
-    fn get_fields(&self) -> Self::StructFields {
-        Self::fields(self.item.unwrap_or(self.llvm_usize.into()), self.llvm_usize)
-    }
-}
-
-impl<'ctx> From<ListType<'ctx>> for PointerType<'ctx> {
-    fn from(value: ListType<'ctx>) -> Self {
-        value.as_base_type()
-    }
-}