core/codegen: fix bug in call_ceil function

- distinguish between setitem and getitem
- allow starred assignment targets, but the assigned value would be a tuple
- allow both [...] and (...) to be target lists

.clang-format

@@ -1,3 +0,0 @@
-BasedOnStyle: Google
-IndentWidth: 4
-ReflowComments: false

.gitignore

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

Cargo.lock

@@ -60,7 +60,7 @@ version = "1.1.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "6d36fc52c7f6c869915e99412912f22093507da8d9e942ceaf66fe4b7c14422a"
 dependencies = [
- "windows-sys",
+ "windows-sys 0.52.0",
 ]
 
 [[package]]
@@ -70,7 +70,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "5bf74e1b6e971609db8ca7a9ce79fd5768ab6ae46441c572e46cf596f59e57f8"
 dependencies = [
  "anstyle",
- "windows-sys",
+ "windows-sys 0.52.0",
 ]
 
 [[package]]
@@ -117,9 +117,9 @@ checksum = "1fd0f2584146f6f2ef48085050886acf353beff7305ebd1ae69500e27c67f64b"
 
 [[package]]
 name = "cc"
-version = "1.1.6"
+version = "1.1.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2aba8f4e9906c7ce3c73463f62a7f0c65183ada1a2d47e397cc8810827f9694f"
+checksum = "26a5c3fd7bfa1ce3897a3a3501d362b2d87b7f2583ebcb4a949ec25911025cbc"
 
 [[package]]
 name = "cfg-if"
@@ -129,9 +129,9 @@ checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"
 
 [[package]]
 name = "clap"
-version = "4.5.11"
+version = "4.5.13"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "35723e6a11662c2afb578bcf0b88bf6ea8e21282a953428f240574fcc3a2b5b3"
+checksum = "0fbb260a053428790f3de475e304ff84cdbc4face759ea7a3e64c1edd938a7fc"
 dependencies = [
  "clap_builder",
  "clap_derive",
@@ -139,9 +139,9 @@ dependencies = [
 
 [[package]]
 name = "clap_builder"
-version = "4.5.11"
+version = "4.5.13"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "49eb96cbfa7cfa35017b7cd548c75b14c3118c98b423041d70562665e07fb0fa"
+checksum = "64b17d7ea74e9f833c7dbf2cbe4fb12ff26783eda4782a8975b72f895c9b4d99"
 dependencies = [
  "anstream",
  "anstyle",
@@ -151,9 +151,9 @@ dependencies = [
 
 [[package]]
 name = "clap_derive"
-version = "4.5.11"
+version = "4.5.13"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5d029b67f89d30bbb547c89fd5161293c0aec155fc691d7924b64550662db93e"
+checksum = "501d359d5f3dcaf6ecdeee48833ae73ec6e42723a1e52419c79abf9507eec0a0"
 dependencies = [
  "heck 0.5.0",
  "proc-macro2",
@@ -182,7 +182,7 @@ dependencies = [
  "encode_unicode",
  "lazy_static",
  "libc",
- "windows-sys",
+ "windows-sys 0.52.0",
 ]
 
 [[package]]
@@ -302,7 +302,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "534c5cf6194dfab3db3242765c03bbe257cf92f22b38f6bc0c58d59108a820ba"
 dependencies = [
  "libc",
- "windows-sys",
+ "windows-sys 0.52.0",
 ]
 
 [[package]]
@@ -385,9 +385,9 @@ dependencies = [
 
 [[package]]
 name = "indexmap"
-version = "2.2.6"
+version = "2.3.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "168fb715dda47215e360912c096649d23d58bf392ac62f73919e831745e40f26"
+checksum = "de3fc2e30ba82dd1b3911c8de1ffc143c74a914a14e99514d7637e3099df5ea0"
 dependencies = [
  "equivalent",
  "hashbrown 0.14.5",
@@ -616,7 +616,7 @@ name = "nac3core"
 version = "0.1.0"
 dependencies = [
  "crossbeam",
- "indexmap 2.2.6",
+ "indexmap 2.3.0",
  "indoc",
  "inkwell",
  "insta",
@@ -706,7 +706,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "b4c5cc86750666a3ed20bdaf5ca2a0344f9c67674cae0515bec2da16fbaa47db"
 dependencies = [
  "fixedbitset",
- "indexmap 2.2.6",
+ "indexmap 2.3.0",
 ]
 
 [[package]]
@@ -784,9 +784,12 @@ checksum = "da544ee218f0d287a911e9c99a39a8c9bc8fcad3cb8db5959940044ecfc67265"
 
 [[package]]
 name = "ppv-lite86"
-version = "0.2.17"
+version = "0.2.20"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5b40af805b3121feab8a3c29f04d8ad262fa8e0561883e7653e024ae4479e6de"
+checksum = "77957b295656769bb8ad2b6a6b09d897d94f05c41b069aede1fcdaa675eaea04"
+dependencies = [
+ "zerocopy",
+]
 
 [[package]]
 name = "precomputed-hash"
@@ -947,9 +950,9 @@ dependencies = [
 
 [[package]]
 name = "regex"
-version = "1.10.5"
+version = "1.10.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b91213439dad192326a0d7c6ee3955910425f441d7038e0d6933b0aec5c4517f"
+checksum = "4219d74c6b67a3654a9fbebc4b419e22126d13d2f3c4a07ee0cb61ff79a79619"
 dependencies = [
  "aho-corasick",
  "memchr",
@@ -991,7 +994,7 @@ dependencies = [
  "errno",
  "libc",
  "linux-raw-sys",
- "windows-sys",
+ "windows-sys 0.52.0",
 ]
 
 [[package]]
@@ -1049,11 +1052,12 @@ dependencies = [
 
 [[package]]
 name = "serde_json"
-version = "1.0.120"
+version = "1.0.122"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4e0d21c9a8cae1235ad58a00c11cb40d4b1e5c784f1ef2c537876ed6ffd8b7c5"
+checksum = "784b6203951c57ff748476b126ccb5e8e2959a5c19e5c617ab1956be3dbc68da"
 dependencies = [
  "itoa",
+ "memchr",
  "ryu",
  "serde",
 ]
@@ -1161,20 +1165,21 @@ dependencies = [
 
 [[package]]
 name = "target-lexicon"
-version = "0.12.15"
+version = "0.12.16"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4873307b7c257eddcb50c9bedf158eb669578359fb28428bef438fec8e6ba7c2"
+checksum = "61c41af27dd6d1e27b1b16b489db798443478cef1f06a660c96db617ba5de3b1"
 
 [[package]]
 name = "tempfile"
-version = "3.10.1"
+version = "3.11.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "85b77fafb263dd9d05cbeac119526425676db3784113aa9295c88498cbf8bff1"
+checksum = "b8fcd239983515c23a32fb82099f97d0b11b8c72f654ed659363a95c3dad7a53"
 dependencies = [
  "cfg-if",
  "fastrand",
+ "once_cell",
  "rustix",
- "windows-sys",
+ "windows-sys 0.52.0",
 ]
 
 [[package]]
@@ -1374,11 +1379,11 @@ checksum = "ac3b87c63620426dd9b991e5ce0329eff545bccbbb34f3be09ff6fb6ab51b7b6"
 
 [[package]]
 name = "winapi-util"
-version = "0.1.8"
+version = "0.1.9"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4d4cc384e1e73b93bafa6fb4f1df8c41695c8a91cf9c4c64358067d15a7b6c6b"
+checksum = "cf221c93e13a30d793f7645a0e7762c55d169dbb0a49671918a2319d289b10bb"
 dependencies = [
- "windows-sys",
+ "windows-sys 0.59.0",
 ]
 
 [[package]]
@@ -1396,6 +1401,15 @@ dependencies = [
  "windows-targets",
 ]
 
+[[package]]
+name = "windows-sys"
+version = "0.59.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1e38bc4d79ed67fd075bcc251a1c39b32a1776bbe92e5bef1f0bf1f8c531853b"
+dependencies = [
+ "windows-targets",
+]
+
 [[package]]
 name = "windows-targets"
 version = "0.52.6"
@@ -1475,6 +1489,7 @@ version = "0.7.35"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "1b9b4fd18abc82b8136838da5d50bae7bdea537c574d8dc1a34ed098d6c166f0"
 dependencies = [
+ "byteorder",
  "zerocopy-derive",
 ]
 

flake.nix

@@ -6,6 +6,7 @@
   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 {};
@@ -13,9 +14,24 @@
           ''
           mkdir -p $out/bin
           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 $out/bin/llvm-as-irrt
           '';
+        demo-linalg-stub = pkgs.rustPlatform.buildRustPackage {
+          name = "demo-linalg-stub";
+          src = ./nac3standalone/demo/linalg;
+          cargoLock = {
+            lockFile = ./nac3standalone/demo/linalg/Cargo.lock;
+          };
+          doCheck = false;
+        };
+        demo-linalg-stub32 = pkgs32.rustPlatform.buildRustPackage {
+          name = "demo-linalg-stub32";
+          src = ./nac3standalone/demo/linalg;
+          cargoLock = {
+            lockFile = ./nac3standalone/demo/linalg/Cargo.lock;
+          };
+          doCheck = false;
+        };
         nac3artiq = pkgs.python3Packages.toPythonModule (
           pkgs.rustPlatform.buildRustPackage rec {
             name = "nac3artiq";
@@ -24,7 +40,6 @@
             cargoLock = {
               lockFile = ./Cargo.lock;
             };
-            cargoTestFlags = [ "--features" "test" ];
             passthru.cargoLock = cargoLock;
             nativeBuildInputs = [ pkgs.python3 (pkgs.wrapClangMulti pkgs.llvmPackages_14.clang) llvm-tools-irrt pkgs.llvmPackages_14.llvm.out llvm-nac3 ];
             buildInputs = [ pkgs.python3 llvm-nac3 ];
@@ -34,7 +49,9 @@
               echo "Checking nac3standalone demos..."
               pushd nac3standalone/demo
               patchShebangs .
-              ./check_demos.sh
+              export DEMO_LINALG_STUB=${demo-linalg-stub}/lib/liblinalg.a
+              export DEMO_LINALG_STUB32=${demo-linalg-stub32}/lib/liblinalg.a
+              ./check_demos.sh -i686
               popd
               echo "Running Cargo tests..."
               cargoCheckHook
@@ -164,6 +181,11 @@
           pre-commit
           rustfmt
         ];
+        shellHook =
+          ''
+          export DEMO_LINALG_STUB=${packages.x86_64-linux.demo-linalg-stub}/lib/liblinalg.a
+          export DEMO_LINALG_STUB32=${packages.x86_64-linux.demo-linalg-stub32}/lib/liblinalg.a
+          '';
       };
       devShells.x86_64-linux.msys2 = pkgs.mkShell {
         name = "nac3-dev-shell-msys2";

nac3artiq/src/codegen.rs

@@ -386,7 +386,7 @@ fn gen_rpc_tag(
     } else {
         let ty_enum = ctx.unifier.get_ty(ty);
         match &*ty_enum {
-            TTuple { ty } => {
+            TTuple { ty, is_vararg_ctx: false } => {
                 buffer.push(b't');
                 buffer.push(ty.len() as u8);
                 for ty in ty {
@@ -700,6 +700,7 @@ pub fn attributes_writeback(
                     name: i.to_string().into(),
                     ty: *ty,
                     default_value: None,
+                    is_vararg: false,
                 })
                 .collect(),
             ret: ctx.primitives.none,

nac3artiq/src/lib.rs

@@ -265,7 +265,7 @@ impl Nac3 {
                     arg_names.len(),
                 ));
             }
-            for (i, FuncArg { ty, default_value, name }) in args.iter().enumerate() {
+            for (i, FuncArg { ty, default_value, name, .. }) in args.iter().enumerate() {
                 let in_name = match arg_names.get(i) {
                     Some(n) => n,
                     None if default_value.is_none() => {
@@ -869,6 +869,7 @@ impl Nac3 {
                         name: "t".into(),
                         ty: primitive.int64,
                         default_value: None,
+                        is_vararg: false,
                     }],
                     ret: primitive.none,
                     vars: VarMap::new(),
@@ -888,6 +889,7 @@ impl Nac3 {
                         name: "dt".into(),
                         ty: primitive.int64,
                         default_value: None,
+                        is_vararg: false,
                     }],
                     ret: primitive.none,
                     vars: VarMap::new(),

nac3artiq/src/symbol_resolver.rs

@@ -351,7 +351,7 @@ impl InnerResolver {
             Ok(Ok((ndarray, false)))
         } else if ty_id == self.primitive_ids.tuple {
             // do not handle type var param and concrete check here
-            Ok(Ok((unifier.add_ty(TypeEnum::TTuple { ty: vec![] }), false)))
+            Ok(Ok((unifier.add_ty(TypeEnum::TTuple { ty: vec![], is_vararg_ctx: false }), false)))
         } else if ty_id == self.primitive_ids.option {
             Ok(Ok((primitives.option, false)))
         } else if ty_id == self.primitive_ids.none {
@@ -555,7 +555,10 @@ impl InnerResolver {
                             Err(err) => return Ok(Err(err)),
                             _ => return Ok(Err("tuple type needs at least 1 type parameters".to_string()))
                         };
-                    Ok(Ok((unifier.add_ty(TypeEnum::TTuple { ty: args }), true)))
+                    Ok(Ok((
+                        unifier.add_ty(TypeEnum::TTuple { ty: args, is_vararg_ctx: false }),
+                        true,
+                    )))
                 }
                 TypeEnum::TObj { params, obj_id, .. } => {
                     let subst = {
@@ -797,7 +800,9 @@ impl InnerResolver {
                     .map(|elem| self.get_obj_type(py, elem, unifier, defs, primitives))
                     .collect();
                 let types = types?;
-                Ok(types.map(|types| unifier.add_ty(TypeEnum::TTuple { ty: types })))
+                Ok(types.map(|types| {
+                    unifier.add_ty(TypeEnum::TTuple { ty: types, is_vararg_ctx: false })
+                }))
             }
             // special handling for option type since its class member layout in python side
             // is special and cannot be mapped directly to a nac3 type as below
@@ -1203,7 +1208,9 @@ impl InnerResolver {
             Ok(Some(ndarray.as_pointer_value().into()))
         } else if ty_id == self.primitive_ids.tuple {
             let expected_ty_enum = ctx.unifier.get_ty_immutable(expected_ty);
-            let TypeEnum::TTuple { ty } = expected_ty_enum.as_ref() else { unreachable!() };
+            let TypeEnum::TTuple { ty, is_vararg_ctx: false } = expected_ty_enum.as_ref() else {
+                unreachable!()
+            };
 
             let tup_tys = ty.iter();
             let elements: &PyTuple = obj.downcast()?;

nac3core/Cargo.toml

@@ -1,6 +1,3 @@
-[features]
-test = []
-
 [package]
 name = "nac3core"
 version = "0.1.0"

nac3core/build.rs

@@ -3,34 +3,20 @@ use std::{
     env,
     fs::File,
     io::Write,
-    path::{Path, PathBuf},
+    path::Path,
     process::{Command, Stdio},
 };
 
-const CMD_IRRT_CLANG: &str = "clang-irrt";
-const CMD_IRRT_CLANG_TEST: &str = "clang-irrt-test";
-const CMD_IRRT_LLVM_AS: &str = "llvm-as-irrt";
-
-fn get_out_dir() -> PathBuf {
-    PathBuf::from(env::var("OUT_DIR").unwrap())
-}
-
-fn get_irrt_dir() -> &'static Path {
-    Path::new("irrt")
-}
-
-/// Compile `irrt.cpp` for use in `src/codegen`
-fn compile_irrt_cpp() {
-    let out_dir = get_out_dir();
-    let irrt_dir = get_irrt_dir();
+fn main() {
+    const FILE: &str = "src/codegen/irrt/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 irrt_cpp_path = irrt_dir.join("irrt.cpp");
     let flags: &[&str] = &[
         "--target=wasm32",
+        FILE,
         "-x",
         "c++",
         "-fno-discard-value-names",
@@ -47,16 +33,13 @@ fn compile_irrt_cpp() {
         "-Wextra",
         "-o",
         "-",
-        "-I",
-        irrt_dir.to_str().unwrap(),
-        irrt_cpp_path.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());
+    println!("cargo:rerun-if-changed={FILE}");
+    let out_dir = env::var("OUT_DIR").unwrap();
+    let out_path = Path::new(&out_dir);
 
-    // Compile IRRT and capture the LLVM IR output
-    let output = Command::new(CMD_IRRT_CLANG)
+    let output = Command::new("clang-irrt")
         .args(flags)
         .output()
         .map(|o| {
@@ -69,17 +52,7 @@ fn compile_irrt_cpp() {
     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();
+    let regex_filter = Regex::new(r"(?ms:^define.*?\}$)|(?m:^declare.*?$)").unwrap();
     for f in regex_filter.captures_iter(&output) {
         assert_eq!(f.len(), 1);
         filtered_output.push_str(&f[0]);
@@ -90,71 +63,20 @@ fn compile_irrt_cpp() {
         .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() {
-        let mut file = File::create(out_dir.join("irrt.ll")).unwrap();
+    println!("cargo:rerun-if-env-changed=DEBUG_DUMP_IRRT");
+    if env::var("DEBUG_DUMP_IRRT").is_ok() {
+        let mut file = File::create(out_path.join("irrt.ll")).unwrap();
         file.write_all(output.as_bytes()).unwrap();
-
-        let mut file = File::create(out_dir.join("irrt-filtered.ll")).unwrap();
+        let mut file = File::create(out_path.join("irrt-filtered.ll")).unwrap();
         file.write_all(filtered_output.as_bytes()).unwrap();
     }
 
-    // Assemble the emitted and filtered IR to .bc
-    // That .bc will be integrated into nac3core's codegen
-    let mut llvm_as = Command::new(CMD_IRRT_LLVM_AS)
+    let mut llvm_as = Command::new("llvm-as-irrt")
         .stdin(Stdio::piped())
         .arg("-o")
-        .arg(out_dir.join("irrt.bc"))
+        .arg(out_path.join("irrt.bc"))
         .spawn()
         .unwrap();
     llvm_as.stdin.as_mut().unwrap().write_all(filtered_output.as_bytes()).unwrap();
     assert!(llvm_as.wait().unwrap().success());
 }
-
-/// Compile `irrt_test.cpp` for testing
-fn compile_irrt_test_cpp() {
-    let out_dir = get_out_dir();
-    let irrt_dir = get_irrt_dir();
-
-    let exe_path = out_dir.join("irrt_test.out"); // Output path of the compiled test executable
-    let irrt_test_cpp_path = irrt_dir.join("irrt_test.cpp");
-    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(CMD_IRRT_CLANG_TEST)
-        .args(flags)
-        .output()
-        .map(|o| {
-            assert!(o.status.success(), "{}", std::str::from_utf8(&o.stderr).unwrap());
-            o
-        })
-        .unwrap();
-    println!("cargo:rerun-if-changed={}", irrt_dir.to_str().unwrap());
-}
-
-fn main() {
-    compile_irrt_cpp();
-
-    // 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_cpp();
-    }
-}

nac3core/irrt/irrt.cpp

@@ -1,10 +0,0 @@
-#define IRRT_DEFINE_TYPEDEF_INTS
-#include <irrt_everything.hpp>
-
-/*
- * All IRRT implementations.
- *
- * We don't have 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/int_defs.hpp

@@ -1,12 +0,0 @@
-#pragma once
-
-// This is made toggleable since `irrt_test.cpp` itself would include
-// headers that define these typedefs
-#ifdef IRRT_DEFINE_TYPEDEF_INTS
-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);
-#endif

nac3core/irrt/irrt/utils.hpp

@@ -1,21 +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;
-}
-
-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;
-}
-}  // namespace

nac3core/irrt/irrt_everything.hpp

@@ -1,5 +0,0 @@
-#pragma once
-
-#include <irrt/core.hpp>
-#include <irrt/int_defs.hpp>
-#include <irrt/utils.hpp>

nac3core/irrt/irrt_test.cpp

@@ -1,12 +0,0 @@
-// 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>
-#include <test/test_core.hpp>
-
-int main() {
-    test::core::run();
-    return 0;
-}

nac3core/irrt/test/includes.hpp

@@ -1,11 +0,0 @@
-#pragma once
-
-#include <cstdint>
-#include <cstdio>
-#include <cstdlib>
-#include <irrt_everything.hpp>
-#include <test/util.hpp>
-
-/*
-    Include this header for every test_*.cpp
-*/

nac3core/irrt/test/test_core.hpp

@@ -1,16 +0,0 @@
-#pragma once
-
-#include <test/includes.hpp>
-
-namespace test {
-namespace core {
-void test_int_exp() {
-    BEGIN_TEST();
-
-    assert_values_match(125, __nac3_int_exp_impl<int32_t>(5, 3));
-    assert_values_match(3125, __nac3_int_exp_impl<int32_t>(5, 5));
-}
-
-void run() { test_int_exp(); }
-}  // namespace core
-}  // namespace test

nac3core/irrt/test/util.hpp

@@ -1,116 +0,0 @@
-#pragma once
-
-#include <cstdio>
-#include <cstdlib>
-
-template <class T>
-void print_value(const T& value);
-
-template <>
-void print_value(const int8_t& value) {
-    printf("%d", value);
-}
-
-template <>
-void print_value(const int32_t& value) {
-    printf("%d", value);
-}
-
-template <>
-void print_value(const uint8_t& value) {
-    printf("%u", value);
-}
-
-template <>
-void print_value(const uint32_t& value) {
-    printf("%u", value);
-}
-
-template <>
-void print_value(const float& value) {
-    printf("%f", value);
-}
-
-template <>
-void print_value(const double& value) {
-    printf("%f", value);
-}
-
-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__)
-
-void test_fail() {
-    printf("[!] Test failed. Exiting with status code 1.\n");
-    exit(1);
-}
-
-template <typename T>
-void debug_print_array(int len, const T* as) {
-    printf("[");
-    for (int i = 0; i < len; i++) {
-        if (i != 0) printf(", ");
-        print_value(as[i]);
-    }
-    printf("]");
-}
-
-void print_assertion_passed(const char* file, int line) {
-    printf("[*] Assertion passed on %s:%d\n", file, line);
-}
-
-void print_assertion_failed(const char* file, int line) {
-    printf("[!] Assertion failed on %s:%d\n", file, line);
-}
-
-void __assert_true(const char* file, int line, bool cond) {
-    if (cond) {
-        print_assertion_passed(file, line);
-    } else {
-        print_assertion_failed(file, line);
-        test_fail();
-    }
-}
-
-#define assert_true(cond) __assert_true(__FILE__, __LINE__, cond)
-
-template <typename T>
-void __assert_arrays_match(const char* file, int line, int len,
-                           const T* expected, const T* got) {
-    if (arrays_match(len, expected, got)) {
-        print_assertion_passed(file, line);
-    } else {
-        print_assertion_failed(file, line);
-        printf("Expect = ");
-        debug_print_array(len, expected);
-        printf("\n");
-        printf("   Got = ");
-        debug_print_array(len, got);
-        printf("\n");
-        test_fail();
-    }
-}
-
-#define assert_arrays_match(len, expected, got) \
-    __assert_arrays_match(__FILE__, __LINE__, len, expected, got)
-
-template <typename T>
-void __assert_values_match(const char* file, int line, T expected, T got) {
-    if (expected == got) {
-        print_assertion_passed(file, line);
-    } else {
-        print_assertion_failed(file, line);
-        printf("Expect = ");
-        print_value(expected);
-        printf("\n");
-        printf("   Got = ");
-        print_value(got);
-        printf("\n");
-        test_fail();
-    }
-}
-
-#define assert_values_match(expected, got) \
-    __assert_values_match(__FILE__, __LINE__, expected, got)

nac3core/src/codegen/builtin_fns.rs

@@ -1,9 +1,11 @@
 use inkwell::types::BasicTypeEnum;
-use inkwell::values::BasicValueEnum;
+use inkwell::values::{BasicValue, BasicValueEnum, PointerValue};
 use inkwell::{FloatPredicate, IntPredicate, OptimizationLevel};
 use itertools::Itertools;
 
-use crate::codegen::classes::{NDArrayValue, ProxyValue, UntypedArrayLikeAccessor};
+use crate::codegen::classes::{
+    NDArrayValue, ProxyValue, UntypedArrayLikeAccessor, UntypedArrayLikeMutator,
+};
 use crate::codegen::numpy::ndarray_elementwise_unaryop_impl;
 use crate::codegen::stmt::gen_for_callback_incrementing;
 use crate::codegen::{extern_fns, irrt, llvm_intrinsics, numpy, CodeGenContext, CodeGenerator};
@@ -31,7 +33,6 @@ pub fn call_int32<'ctx, G: CodeGenerator + ?Sized>(
     let llvm_usize = generator.get_size_type(ctx.ctx);
 
     let (n_ty, n) = n;
-
     Ok(match n {
         BasicValueEnum::IntValue(n) if matches!(n.get_type().get_bit_width(), 1 | 8) => {
             debug_assert!(ctx.unifier.unioned(n_ty, ctx.primitives.bool));
@@ -602,7 +603,7 @@ pub fn call_ceil<'ctx, G: CodeGenerator + ?Sized>(
                 ret_elem_ty,
                 None,
                 NDArrayValue::from_ptr_val(n, llvm_usize, None),
-                |generator, ctx, val| call_floor(generator, ctx, (elem_ty, val), ret_elem_ty),
+                |generator, ctx, val| call_ceil(generator, ctx, (elem_ty, val), ret_elem_ty),
             )?;
 
             ndarray.as_base_value().into()
@@ -1836,3 +1837,501 @@ pub fn call_numpy_nextafter<'ctx, G: CodeGenerator + ?Sized>(
         _ => unsupported_type(ctx, FN_NAME, &[x1_ty, x2_ty]),
     })
 }
+
+/// Allocates a struct with the fields specified by `out_matrices` and returns a pointer to it
+fn build_output_struct<'ctx>(
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    out_matrices: Vec<BasicValueEnum<'ctx>>,
+) -> PointerValue<'ctx> {
+    let field_ty =
+        out_matrices.iter().map(BasicValueEnum::get_type).collect::<Vec<BasicTypeEnum>>();
+    let out_ty = ctx.ctx.struct_type(&field_ty, false);
+    let out_ptr = ctx.builder.build_alloca(out_ty, "").unwrap();
+
+    for (i, v) in out_matrices.into_iter().enumerate() {
+        unsafe {
+            let ptr = ctx
+                .builder
+                .build_in_bounds_gep(
+                    out_ptr,
+                    &[
+                        ctx.ctx.i32_type().const_zero(),
+                        ctx.ctx.i32_type().const_int(i as u64, false),
+                    ],
+                    "",
+                )
+                .unwrap();
+            ctx.builder.build_store(ptr, v).unwrap();
+        }
+    }
+    out_ptr
+}
+
+/// Invokes the `np_linalg_cholesky` linalg function
+pub fn call_np_linalg_cholesky<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    x1: (Type, BasicValueEnum<'ctx>),
+) -> Result<BasicValueEnum<'ctx>, String> {
+    const FN_NAME: &str = "np_linalg_cholesky";
+    let (x1_ty, x1) = x1;
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+
+    if let BasicValueEnum::PointerValue(n1) = x1 {
+        let (elem_ty, _) = unpack_ndarray_var_tys(&mut ctx.unifier, x1_ty);
+        let n1_elem_ty = ctx.get_llvm_type(generator, elem_ty);
+
+        let BasicTypeEnum::FloatType(_) = n1_elem_ty else {
+            unsupported_type(ctx, FN_NAME, &[x1_ty]);
+        };
+
+        let n1 = NDArrayValue::from_ptr_val(n1, llvm_usize, None);
+        let dim0 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_zero(), None)
+                .into_int_value()
+        };
+        let dim1 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_int(1, false), None)
+                .into_int_value()
+        };
+
+        let out = numpy::create_ndarray_const_shape(generator, ctx, elem_ty, &[dim0, dim1])
+            .unwrap()
+            .as_base_value()
+            .as_basic_value_enum();
+
+        extern_fns::call_np_linalg_cholesky(ctx, x1, out, None);
+        Ok(out)
+    } else {
+        unsupported_type(ctx, FN_NAME, &[x1_ty])
+    }
+}
+
+/// Invokes the `np_linalg_qr` linalg function
+pub fn call_np_linalg_qr<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    x1: (Type, BasicValueEnum<'ctx>),
+) -> Result<BasicValueEnum<'ctx>, String> {
+    const FN_NAME: &str = "np_linalg_qr";
+    let (x1_ty, x1) = x1;
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+
+    if let BasicValueEnum::PointerValue(n1) = x1 {
+        let (elem_ty, _) = unpack_ndarray_var_tys(&mut ctx.unifier, x1_ty);
+        let n1_elem_ty = ctx.get_llvm_type(generator, elem_ty);
+
+        let BasicTypeEnum::FloatType(_) = n1_elem_ty else {
+            unimplemented!("{FN_NAME} operates on float type NdArrays only");
+        };
+
+        let n1 = NDArrayValue::from_ptr_val(n1, llvm_usize, None);
+        let dim0 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_zero(), None)
+                .into_int_value()
+        };
+        let dim1 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_int(1, false), None)
+                .into_int_value()
+        };
+        let k = llvm_intrinsics::call_int_smin(ctx, dim0, dim1, None);
+
+        let out_q = numpy::create_ndarray_const_shape(generator, ctx, elem_ty, &[dim0, k])
+            .unwrap()
+            .as_base_value()
+            .as_basic_value_enum();
+        let out_r = numpy::create_ndarray_const_shape(generator, ctx, elem_ty, &[k, dim1])
+            .unwrap()
+            .as_base_value()
+            .as_basic_value_enum();
+
+        extern_fns::call_np_linalg_qr(ctx, x1, out_q, out_r, None);
+
+        let out_ptr = build_output_struct(ctx, vec![out_q, out_r]);
+
+        Ok(ctx.builder.build_load(out_ptr, "QR_Factorization_result").map(Into::into).unwrap())
+    } else {
+        unsupported_type(ctx, FN_NAME, &[x1_ty])
+    }
+}
+
+/// Invokes the `np_linalg_svd` linalg function
+pub fn call_np_linalg_svd<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    x1: (Type, BasicValueEnum<'ctx>),
+) -> Result<BasicValueEnum<'ctx>, String> {
+    const FN_NAME: &str = "np_linalg_svd";
+    let (x1_ty, x1) = x1;
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+
+    if let BasicValueEnum::PointerValue(n1) = x1 {
+        let (elem_ty, _) = unpack_ndarray_var_tys(&mut ctx.unifier, x1_ty);
+        let n1_elem_ty = ctx.get_llvm_type(generator, elem_ty);
+
+        let BasicTypeEnum::FloatType(_) = n1_elem_ty else {
+            unsupported_type(ctx, FN_NAME, &[x1_ty]);
+        };
+
+        let n1 = NDArrayValue::from_ptr_val(n1, llvm_usize, None);
+
+        let dim0 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_zero(), None)
+                .into_int_value()
+        };
+        let dim1 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_int(1, false), None)
+                .into_int_value()
+        };
+        let k = llvm_intrinsics::call_int_smin(ctx, dim0, dim1, None);
+
+        let out_u = numpy::create_ndarray_const_shape(generator, ctx, elem_ty, &[dim0, dim0])
+            .unwrap()
+            .as_base_value()
+            .as_basic_value_enum();
+        let out_s = numpy::create_ndarray_const_shape(generator, ctx, elem_ty, &[k])
+            .unwrap()
+            .as_base_value()
+            .as_basic_value_enum();
+        let out_vh = numpy::create_ndarray_const_shape(generator, ctx, elem_ty, &[dim1, dim1])
+            .unwrap()
+            .as_base_value()
+            .as_basic_value_enum();
+
+        extern_fns::call_np_linalg_svd(ctx, x1, out_u, out_s, out_vh, None);
+
+        let out_ptr = build_output_struct(ctx, vec![out_u, out_s, out_vh]);
+
+        Ok(ctx.builder.build_load(out_ptr, "SVD_Factorization_result").map(Into::into).unwrap())
+    } else {
+        unsupported_type(ctx, FN_NAME, &[x1_ty])
+    }
+}
+
+/// Invokes the `np_linalg_inv` linalg function
+pub fn call_np_linalg_inv<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    x1: (Type, BasicValueEnum<'ctx>),
+) -> Result<BasicValueEnum<'ctx>, String> {
+    const FN_NAME: &str = "np_linalg_inv";
+    let (x1_ty, x1) = x1;
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+
+    if let BasicValueEnum::PointerValue(n1) = x1 {
+        let (elem_ty, _) = unpack_ndarray_var_tys(&mut ctx.unifier, x1_ty);
+        let n1_elem_ty = ctx.get_llvm_type(generator, elem_ty);
+
+        let BasicTypeEnum::FloatType(_) = n1_elem_ty else {
+            unsupported_type(ctx, FN_NAME, &[x1_ty]);
+        };
+
+        let n1 = NDArrayValue::from_ptr_val(n1, llvm_usize, None);
+        let dim0 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_zero(), None)
+                .into_int_value()
+        };
+        let dim1 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_int(1, false), None)
+                .into_int_value()
+        };
+
+        let out = numpy::create_ndarray_const_shape(generator, ctx, elem_ty, &[dim0, dim1])
+            .unwrap()
+            .as_base_value()
+            .as_basic_value_enum();
+
+        extern_fns::call_np_linalg_inv(ctx, x1, out, None);
+        Ok(out)
+    } else {
+        unsupported_type(ctx, FN_NAME, &[x1_ty])
+    }
+}
+
+/// Invokes the `np_linalg_pinv` linalg function
+pub fn call_np_linalg_pinv<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    x1: (Type, BasicValueEnum<'ctx>),
+) -> Result<BasicValueEnum<'ctx>, String> {
+    const FN_NAME: &str = "np_linalg_pinv";
+    let (x1_ty, x1) = x1;
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+
+    if let BasicValueEnum::PointerValue(n1) = x1 {
+        let (elem_ty, _) = unpack_ndarray_var_tys(&mut ctx.unifier, x1_ty);
+        let n1_elem_ty = ctx.get_llvm_type(generator, elem_ty);
+
+        let BasicTypeEnum::FloatType(_) = n1_elem_ty else {
+            unsupported_type(ctx, FN_NAME, &[x1_ty]);
+        };
+
+        let n1 = NDArrayValue::from_ptr_val(n1, llvm_usize, None);
+
+        let dim0 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_zero(), None)
+                .into_int_value()
+        };
+        let dim1 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_int(1, false), None)
+                .into_int_value()
+        };
+
+        let out = numpy::create_ndarray_const_shape(generator, ctx, elem_ty, &[dim1, dim0])
+            .unwrap()
+            .as_base_value()
+            .as_basic_value_enum();
+
+        extern_fns::call_np_linalg_pinv(ctx, x1, out, None);
+        Ok(out)
+    } else {
+        unsupported_type(ctx, FN_NAME, &[x1_ty])
+    }
+}
+
+/// Invokes the `sp_linalg_lu` linalg function
+pub fn call_sp_linalg_lu<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    x1: (Type, BasicValueEnum<'ctx>),
+) -> Result<BasicValueEnum<'ctx>, String> {
+    const FN_NAME: &str = "sp_linalg_lu";
+    let (x1_ty, x1) = x1;
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+
+    if let BasicValueEnum::PointerValue(n1) = x1 {
+        let (elem_ty, _) = unpack_ndarray_var_tys(&mut ctx.unifier, x1_ty);
+        let n1_elem_ty = ctx.get_llvm_type(generator, elem_ty);
+
+        let BasicTypeEnum::FloatType(_) = n1_elem_ty else {
+            unsupported_type(ctx, FN_NAME, &[x1_ty]);
+        };
+
+        let n1 = NDArrayValue::from_ptr_val(n1, llvm_usize, None);
+
+        let dim0 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_zero(), None)
+                .into_int_value()
+        };
+        let dim1 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_int(1, false), None)
+                .into_int_value()
+        };
+        let k = llvm_intrinsics::call_int_smin(ctx, dim0, dim1, None);
+
+        let out_l = numpy::create_ndarray_const_shape(generator, ctx, elem_ty, &[dim0, k])
+            .unwrap()
+            .as_base_value()
+            .as_basic_value_enum();
+        let out_u = numpy::create_ndarray_const_shape(generator, ctx, elem_ty, &[k, dim1])
+            .unwrap()
+            .as_base_value()
+            .as_basic_value_enum();
+
+        extern_fns::call_sp_linalg_lu(ctx, x1, out_l, out_u, None);
+
+        let out_ptr = build_output_struct(ctx, vec![out_l, out_u]);
+        Ok(ctx.builder.build_load(out_ptr, "LU_Factorization_result").map(Into::into).unwrap())
+    } else {
+        unsupported_type(ctx, FN_NAME, &[x1_ty])
+    }
+}
+
+/// Invokes the `np_linalg_matrix_power` linalg function
+pub fn call_np_linalg_matrix_power<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    x1: (Type, BasicValueEnum<'ctx>),
+    x2: (Type, BasicValueEnum<'ctx>),
+) -> Result<BasicValueEnum<'ctx>, String> {
+    const FN_NAME: &str = "np_linalg_matrix_power";
+    let (x1_ty, x1) = x1;
+    let (x2_ty, x2) = x2;
+    let x2 = call_float(generator, ctx, (x2_ty, x2)).unwrap();
+
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+    if let (BasicValueEnum::PointerValue(n1), BasicValueEnum::FloatValue(n2)) = (x1, x2) {
+        let (elem_ty, _) = unpack_ndarray_var_tys(&mut ctx.unifier, x1_ty);
+        let n1_elem_ty = ctx.get_llvm_type(generator, elem_ty);
+
+        let BasicTypeEnum::FloatType(_) = n1_elem_ty else {
+            unsupported_type(ctx, FN_NAME, &[x1_ty, x2_ty]);
+        };
+
+        let n1 = NDArrayValue::from_ptr_val(n1, llvm_usize, None);
+        // Changing second parameter to a `NDArray` for uniformity in function call
+        let n2_array = numpy::create_ndarray_const_shape(
+            generator,
+            ctx,
+            elem_ty,
+            &[llvm_usize.const_int(1, false)],
+        )
+        .unwrap();
+        unsafe {
+            n2_array.data().set_unchecked(
+                ctx,
+                generator,
+                &llvm_usize.const_zero(),
+                n2.as_basic_value_enum(),
+            );
+        };
+        let n2_array = n2_array.as_base_value().as_basic_value_enum();
+
+        let outdim0 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_zero(), None)
+                .into_int_value()
+        };
+        let outdim1 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_int(1, false), None)
+                .into_int_value()
+        };
+
+        let out = numpy::create_ndarray_const_shape(generator, ctx, elem_ty, &[outdim0, outdim1])
+            .unwrap()
+            .as_base_value()
+            .as_basic_value_enum();
+
+        extern_fns::call_np_linalg_matrix_power(ctx, x1, n2_array, out, None);
+        Ok(out)
+    } else {
+        unsupported_type(ctx, FN_NAME, &[x1_ty, x2_ty])
+    }
+}
+
+/// Invokes the `np_linalg_det` linalg function
+pub fn call_np_linalg_det<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    x1: (Type, BasicValueEnum<'ctx>),
+) -> Result<BasicValueEnum<'ctx>, String> {
+    const FN_NAME: &str = "np_linalg_matrix_power";
+    let (x1_ty, x1) = x1;
+
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+    if let BasicValueEnum::PointerValue(_) = x1 {
+        let (elem_ty, _) = unpack_ndarray_var_tys(&mut ctx.unifier, x1_ty);
+        let n1_elem_ty = ctx.get_llvm_type(generator, elem_ty);
+
+        let BasicTypeEnum::FloatType(_) = n1_elem_ty else {
+            unsupported_type(ctx, FN_NAME, &[x1_ty]);
+        };
+
+        // Changing second parameter to a `NDArray` for uniformity in function call
+        let out = numpy::create_ndarray_const_shape(
+            generator,
+            ctx,
+            elem_ty,
+            &[llvm_usize.const_int(1, false)],
+        )
+        .unwrap();
+        extern_fns::call_np_linalg_det(ctx, x1, out.as_base_value().as_basic_value_enum(), None);
+        let res =
+            unsafe { out.data().get_unchecked(ctx, generator, &llvm_usize.const_zero(), None) };
+        Ok(res)
+    } else {
+        unsupported_type(ctx, FN_NAME, &[x1_ty])
+    }
+}
+
+/// Invokes the `sp_linalg_schur` linalg function
+pub fn call_sp_linalg_schur<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    x1: (Type, BasicValueEnum<'ctx>),
+) -> Result<BasicValueEnum<'ctx>, String> {
+    const FN_NAME: &str = "sp_linalg_schur";
+    let (x1_ty, x1) = x1;
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+
+    if let BasicValueEnum::PointerValue(n1) = x1 {
+        let (elem_ty, _) = unpack_ndarray_var_tys(&mut ctx.unifier, x1_ty);
+        let n1_elem_ty = ctx.get_llvm_type(generator, elem_ty);
+
+        let BasicTypeEnum::FloatType(_) = n1_elem_ty else {
+            unsupported_type(ctx, FN_NAME, &[x1_ty]);
+        };
+
+        let n1 = NDArrayValue::from_ptr_val(n1, llvm_usize, None);
+
+        let dim0 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_zero(), None)
+                .into_int_value()
+        };
+        let out_t = numpy::create_ndarray_const_shape(generator, ctx, elem_ty, &[dim0, dim0])
+            .unwrap()
+            .as_base_value()
+            .as_basic_value_enum();
+        let out_z = numpy::create_ndarray_const_shape(generator, ctx, elem_ty, &[dim0, dim0])
+            .unwrap()
+            .as_base_value()
+            .as_basic_value_enum();
+
+        extern_fns::call_sp_linalg_schur(ctx, x1, out_t, out_z, None);
+
+        let out_ptr = build_output_struct(ctx, vec![out_t, out_z]);
+        Ok(ctx.builder.build_load(out_ptr, "Schur_Factorization_result").map(Into::into).unwrap())
+    } else {
+        unsupported_type(ctx, FN_NAME, &[x1_ty])
+    }
+}
+
+/// Invokes the `sp_linalg_hessenberg` linalg function
+pub fn call_sp_linalg_hessenberg<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    x1: (Type, BasicValueEnum<'ctx>),
+) -> Result<BasicValueEnum<'ctx>, String> {
+    const FN_NAME: &str = "sp_linalg_hessenberg";
+    let (x1_ty, x1) = x1;
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+
+    if let BasicValueEnum::PointerValue(n1) = x1 {
+        let (elem_ty, _) = unpack_ndarray_var_tys(&mut ctx.unifier, x1_ty);
+        let n1_elem_ty = ctx.get_llvm_type(generator, elem_ty);
+
+        let BasicTypeEnum::FloatType(_) = n1_elem_ty else {
+            unsupported_type(ctx, FN_NAME, &[x1_ty]);
+        };
+
+        let n1 = NDArrayValue::from_ptr_val(n1, llvm_usize, None);
+
+        let dim0 = unsafe {
+            n1.dim_sizes()
+                .get_unchecked(ctx, generator, &llvm_usize.const_zero(), None)
+                .into_int_value()
+        };
+        let out_h = numpy::create_ndarray_const_shape(generator, ctx, elem_ty, &[dim0, dim0])
+            .unwrap()
+            .as_base_value()
+            .as_basic_value_enum();
+        let out_q = numpy::create_ndarray_const_shape(generator, ctx, elem_ty, &[dim0, dim0])
+            .unwrap()
+            .as_base_value()
+            .as_basic_value_enum();
+        extern_fns::call_sp_linalg_hessenberg(ctx, x1, out_h, out_q, None);
+
+        let out_ptr = build_output_struct(ctx, vec![out_h, out_q]);
+        Ok(ctx
+            .builder
+            .build_load(out_ptr, "Hessenberg_decomposition_result")
+            .map(Into::into)
+            .unwrap())
+    } else {
+        unsupported_type(ctx, FN_NAME, &[x1_ty])
+    }
+}

nac3core/src/codegen/concrete_type.rs

@@ -25,6 +25,7 @@ pub struct ConcreteFuncArg {
     pub name: StrRef,
     pub ty: ConcreteType,
     pub default_value: Option<SymbolValue>,
+    pub is_vararg: bool,
 }
 
 #[derive(Clone, Debug)]
@@ -46,6 +47,7 @@ pub enum ConcreteTypeEnum {
     TPrimitive(Primitive),
     TTuple {
         ty: Vec<ConcreteType>,
+        is_vararg_ctx: bool,
     },
     TObj {
         obj_id: DefinitionId,
@@ -102,8 +104,16 @@ impl ConcreteTypeStore {
                 .iter()
                 .map(|arg| ConcreteFuncArg {
                     name: arg.name,
-                    ty: self.from_unifier_type(unifier, primitives, arg.ty, cache),
+                    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)
+                    },
                     default_value: arg.default_value.clone(),
+                    is_vararg: arg.is_vararg,
                 })
                 .collect(),
             ret: self.from_unifier_type(unifier, primitives, signature.ret, cache),
@@ -158,11 +168,12 @@ impl ConcreteTypeStore {
             cache.insert(ty, None);
             let ty_enum = unifier.get_ty(ty);
             let result = match &*ty_enum {
-                TypeEnum::TTuple { ty } => ConcreteTypeEnum::TTuple {
+                TypeEnum::TTuple { ty, is_vararg_ctx } => 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,
@@ -248,11 +259,12 @@ impl ConcreteTypeStore {
                 *cache.get_mut(&cty).unwrap() = Some(ty);
                 return ty;
             }
-            ConcreteTypeEnum::TTuple { ty } => TypeEnum::TTuple {
+            ConcreteTypeEnum::TTuple { ty, is_vararg_ctx } => 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) }
@@ -277,6 +289,7 @@ 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/expr.rs

@@ -1,5 +1,3 @@
-use std::{collections::HashMap, convert::TryInto, iter::once, iter::zip};
-
 use crate::{
     codegen::{
         classes::{
@@ -7,7 +5,7 @@ use crate::{
             ProxyValue, RangeValue, TypedArrayLikeAccessor, UntypedArrayLikeAccessor,
         },
         concrete_type::{ConcreteFuncArg, ConcreteTypeEnum, ConcreteTypeStore},
-        gen_in_range_check, get_llvm_abi_type, get_llvm_type,
+        gen_in_range_check, get_llvm_abi_type, get_llvm_type, get_va_count_arg_name,
         irrt::*,
         llvm_intrinsics::{
             call_expect, call_float_floor, call_float_pow, call_float_powi, call_int_smax,
@@ -42,6 +40,8 @@ use nac3parser::ast::{
     self, Boolop, Cmpop, Comprehension, Constant, Expr, ExprKind, Location, Operator, StrRef,
     Unaryop,
 };
+use std::iter::{repeat, repeat_with};
+use std::{collections::HashMap, convert::TryInto, iter::once, iter::zip};
 
 pub fn get_subst_key(
     unifier: &mut Unifier,
@@ -201,7 +201,7 @@ impl<'ctx, 'a> CodeGenContext<'ctx, 'a> {
     /// See [`get_llvm_type`].
     pub fn get_llvm_type<G: CodeGenerator + ?Sized>(
         &mut self,
-        generator: &mut G,
+        generator: &G,
         ty: Type,
     ) -> BasicTypeEnum<'ctx> {
         get_llvm_type(
@@ -218,7 +218,7 @@ impl<'ctx, 'a> CodeGenContext<'ctx, 'a> {
     /// See [`get_llvm_abi_type`].
     pub fn get_llvm_abi_type<G: CodeGenerator + ?Sized>(
         &mut self,
-        generator: &mut G,
+        generator: &G,
         ty: Type,
     ) -> BasicTypeEnum<'ctx> {
         get_llvm_abi_type(
@@ -267,13 +267,16 @@ impl<'ctx, 'a> CodeGenContext<'ctx, 'a> {
             }
             Constant::Tuple(v) => {
                 let ty = self.unifier.get_ty(ty);
-                let types =
-                    if let TypeEnum::TTuple { ty } = &*ty { ty.clone() } else { unreachable!() };
+                let (types, is_vararg_ctx) = if let TypeEnum::TTuple { ty, is_vararg_ctx } = &*ty {
+                    (ty.clone(), *is_vararg_ctx)
+                } else {
+                    unreachable!()
+                };
                 let values = zip(types, v.iter())
                     .map_while(|(ty, v)| self.gen_const(generator, v, ty))
                     .collect_vec();
 
-                if values.len() == v.len() {
+                if is_vararg_ctx || values.len() == v.len() {
                     let types = values.iter().map(BasicValueEnum::get_type).collect_vec();
                     let ty = self.ctx.struct_type(&types, false);
                     Some(ty.const_named_struct(&values).into())
@@ -514,16 +517,19 @@ impl<'ctx, 'a> CodeGenContext<'ctx, 'a> {
                 }
             }
         }
+
         let params = if loc_params.is_empty() { params } else { &loc_params };
         let params = fun
             .get_type()
             .get_param_types()
             .into_iter()
+            .map(Some)
+            .chain(repeat(None))
             .zip(params.iter())
             .map(|(ty, val)| match (ty, val.get_type()) {
-                (BasicTypeEnum::PointerType(arg_ty), BasicTypeEnum::PointerType(val_ty))
+                (Some(BasicTypeEnum::PointerType(arg_ty)), BasicTypeEnum::PointerType(val_ty))
                     if {
-                        ty != val.get_type()
+                        ty.unwrap() != val.get_type()
                             && arg_ty.get_element_type().is_struct_type()
                             && val_ty.get_element_type().is_struct_type()
                     } =>
@@ -533,6 +539,7 @@ impl<'ctx, 'a> CodeGenContext<'ctx, 'a> {
                 _ => *val,
             })
             .collect_vec();
+
         let result = if let Some(target) = self.unwind_target {
             let current = self.builder.get_insert_block().unwrap().get_parent().unwrap();
             let then_block = self.ctx.append_basic_block(current, &format!("after.{call_name}"));
@@ -552,6 +559,7 @@ impl<'ctx, 'a> CodeGenContext<'ctx, 'a> {
                 .map(Either::left)
                 .unwrap()
         };
+
         if let Some(slot) = return_slot {
             Some(self.builder.build_load(slot, call_name).unwrap())
         } else {
@@ -726,13 +734,41 @@ pub fn gen_func_instance<'ctx>(
         .collect();
 
     let mut signature = store.from_signature(&mut ctx.unifier, &ctx.primitives, sign, &mut cache);
+    let ConcreteTypeEnum::TFunc { args, .. } = &mut signature else { unreachable!() };
 
     if let Some(obj) = &obj {
         let zelf = store.from_unifier_type(&mut ctx.unifier, &ctx.primitives, obj.0, &mut cache);
-        let ConcreteTypeEnum::TFunc { args, .. } = &mut signature else { unreachable!() };
 
-        args.insert(0, ConcreteFuncArg { name: "self".into(), ty: zelf, default_value: None });
+        args.insert(
+            0,
+            ConcreteFuncArg {
+                name: "self".into(),
+                ty: zelf,
+                default_value: None,
+                is_vararg: false,
+            },
+        );
     }
+
+    if let Some(vararg_arg) = sign.args.iter().find(|arg| arg.is_vararg) {
+        let va_count_arg = get_va_count_arg_name(vararg_arg.name);
+
+        args.insert(
+            args.len() - 1,
+            ConcreteFuncArg {
+                name: va_count_arg,
+                ty: store.from_unifier_type(
+                    &mut ctx.unifier,
+                    &ctx.primitives,
+                    ctx.primitives.usize(),
+                    &mut cache,
+                ),
+                default_value: None,
+                is_vararg: false,
+            },
+        );
+    }
+
     let signature = store.add_cty(signature);
 
     ctx.registry.add_task(CodeGenTask {
@@ -757,11 +793,17 @@ pub fn gen_call<'ctx, G: CodeGenerator>(
     fun: (&FunSignature, DefinitionId),
     params: Vec<(Option<StrRef>, ValueEnum<'ctx>)>,
 ) -> Result<Option<BasicValueEnum<'ctx>>, String> {
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+
     let definition = ctx.top_level.definitions.read().get(fun.1 .0).cloned().unwrap();
     let id;
     let key;
     let param_vals;
     let is_extern;
+    let vararg_arg;
+
+    // Ensure that the function object only contains up to 1 vararg parameter
+    debug_assert!(fun.0.args.iter().filter(|arg| arg.is_vararg).count() <= 1);
 
     let symbol = {
         // make sure this lock guard is dropped at the end of this scope...
@@ -777,22 +819,72 @@ pub fn gen_call<'ctx, G: CodeGenerator>(
                     return callback.run(ctx, obj, fun, params, generator);
                 }
                 is_extern = instance_to_stmt.is_empty();
+                vararg_arg = fun.0.args.iter().find(|arg| arg.is_vararg);
                 let old_key = ctx.get_subst_key(obj.as_ref().map(|a| a.0), fun.0, None);
                 let mut keys = fun.0.args.clone();
-                let mut mapping = HashMap::new();
+                let mut mapping = HashMap::<_, Vec<ValueEnum>>::new();
+
                 for (key, value) in params {
-                    mapping.insert(key.unwrap_or_else(|| keys.remove(0).name), value);
+                    // Find the matching argument
+                    let matching_param = fun
+                        .0
+                        .args
+                        .iter()
+                        .find_or_last(|p| key.is_some_and(|k| k == p.name))
+                        .unwrap();
+                    if matching_param.is_vararg {
+                        if key.is_none() && !keys.is_empty() {
+                            keys.remove(0);
+                        }
+
+                        // vararg is lowered into two arguments - va_count and `...`
+                        // Handle va_count first, for each argument encountered we increment it by 1
+                        let va_count = get_va_count_arg_name(matching_param.name);
+                        if let Some(params) = mapping.get_mut(&va_count) {
+                            debug_assert_eq!(params.len(), 1);
+
+                            let param = params[0]
+                                .clone()
+                                .to_basic_value_enum(ctx, generator, ctx.primitives.usize())?
+                                .into_int_value();
+                            params[0] = param.const_add(llvm_usize.const_int(1, false)).into();
+                        } else {
+                            mapping.insert(va_count, vec![llvm_usize.const_int(1, false).into()]);
+                        }
+
+                        if let Some(param) = mapping.get_mut(&matching_param.name) {
+                            param.push(value);
+                        } else {
+                            mapping.insert(key.unwrap_or(matching_param.name), vec![value]);
+                        }
+                    } else {
+                        mapping.insert(key.unwrap_or_else(|| keys.remove(0).name), vec![value]);
+                    }
                 }
+
                 // default value handling
                 for k in keys {
                     if mapping.contains_key(&k.name) {
                         continue;
                     }
-                    mapping.insert(
-                        k.name,
-                        ctx.gen_symbol_val(generator, &k.default_value.unwrap(), k.ty).into(),
-                    );
+
+                    if k.is_vararg {
+                        mapping.insert(
+                            get_va_count_arg_name(k.name),
+                            vec![llvm_usize.const_zero().into()],
+                        );
+
+                        mapping.insert(k.name, Vec::default());
+                    } else {
+                        mapping.insert(
+                            k.name,
+                            vec![ctx
+                                .gen_symbol_val(generator, &k.default_value.unwrap(), k.ty)
+                                .into()],
+                        );
+                    }
                 }
+
                 // reorder the parameters
                 let mut real_params = fun
                     .0
@@ -801,13 +893,24 @@ pub fn gen_call<'ctx, G: CodeGenerator>(
                     .map(|arg| (mapping.remove(&arg.name).unwrap(), arg.ty))
                     .collect_vec();
                 if let Some(obj) = &obj {
-                    real_params.insert(0, (obj.1.clone(), obj.0));
+                    real_params.insert(0, (vec![obj.1.clone()], obj.0));
                 }
+                if let Some(vararg) = vararg_arg {
+                    let vararg_arg_name = get_va_count_arg_name(vararg.name);
+
+                    real_params.insert(
+                        real_params.len() - 1,
+                        (mapping[&vararg_arg_name].clone(), ctx.primitives.usize()),
+                    );
+                }
+
                 let static_params = real_params
                     .iter()
                     .enumerate()
                     .filter_map(|(i, (v, _))| {
-                        if let ValueEnum::Static(s) = v {
+                        if v.len() != 1 {
+                            None
+                        } else if let ValueEnum::Static(s) = &v[0] {
                             Some((i, s.clone()))
                         } else {
                             None
@@ -837,8 +940,13 @@ pub fn gen_call<'ctx, G: CodeGenerator>(
                 };
                 param_vals = real_params
                     .into_iter()
-                    .map(|(p, t)| p.to_basic_value_enum(ctx, generator, t))
-                    .collect::<Result<Vec<_>, String>>()?;
+                    .map(|(ps, t)| {
+                        ps.into_iter().map(|p| p.to_basic_value_enum(ctx, generator, t)).collect()
+                    })
+                    .collect::<Result<Vec<Vec<_>>, _>>()?
+                    .into_iter()
+                    .flatten()
+                    .collect::<Vec<_>>();
                 instance_to_symbol.get(&key).cloned().ok_or_else(String::new)
             }
             TopLevelDef::Class { .. } => {
@@ -852,7 +960,10 @@ pub fn gen_call<'ctx, G: CodeGenerator>(
     let fun_val = ctx.module.get_function(&symbol).unwrap_or_else(|| {
         let mut args = fun.0.args.clone();
         if let Some(obj) = &obj {
-            args.insert(0, FuncArg { name: "self".into(), ty: obj.0, default_value: None });
+            args.insert(
+                0,
+                FuncArg { name: "self".into(), ty: obj.0, default_value: None, is_vararg: false },
+            );
         }
         let ret_type = if ctx.unifier.unioned(fun.0.ret, ctx.primitives.none) {
             None
@@ -864,6 +975,7 @@ pub fn gen_call<'ctx, G: CodeGenerator>(
         let mut params = args
             .iter()
             .enumerate()
+            .filter(|(_, arg)| !arg.is_vararg)
             .map(|(i, arg)| {
                 match ctx.get_llvm_abi_type(generator, arg.ty) {
                     BasicTypeEnum::StructType(ty) if is_extern => {
@@ -878,9 +990,13 @@ pub fn gen_call<'ctx, G: CodeGenerator>(
         if has_sret {
             params.insert(0, ret_type.unwrap().ptr_type(AddressSpace::default()).into());
         }
+        let is_vararg = args.iter().any(|arg| arg.is_vararg);
+        if is_vararg {
+            params.push(generator.get_size_type(ctx.ctx).into());
+        }
         let fun_ty = match ret_type {
-            Some(ret_type) if !has_sret => ret_type.fn_type(&params, false),
-            _ => ctx.ctx.void_type().fn_type(&params, false),
+            Some(ret_type) if !has_sret => ret_type.fn_type(&params, is_vararg),
+            _ => ctx.ctx.void_type().fn_type(&params, is_vararg),
         };
         let fun_val = ctx.module.add_function(&symbol, fun_ty, None);
         let offset = if has_sret {
@@ -912,13 +1028,16 @@ pub fn gen_call<'ctx, G: CodeGenerator>(
     });
 
     // Convert boolean parameter values into i1
+    let vararg_ty = vararg_arg.map(|vararg| ctx.get_llvm_abi_type(generator, vararg.ty));
     let param_vals = fun_val
         .get_params()
         .iter()
+        .map(BasicValueEnum::get_type)
+        .chain(repeat_with(|| vararg_ty.unwrap()))
         .zip(param_vals)
         .map(|(p, v)| {
-            if p.is_int_value() && v.is_int_value() {
-                let expected_ty = p.into_int_value().get_type();
+            if p.is_int_type() && v.is_int_value() {
+                let expected_ty = p.into_int_type();
                 let param_val = v.into_int_value();
 
                 if expected_ty.get_bit_width() == 1 && param_val.get_type().get_bit_width() != 1 {
@@ -995,8 +1114,10 @@ pub fn gen_comprehension<'ctx, G: CodeGenerator>(
     ctx.builder.position_at_end(init_bb);
 
     let Comprehension { target, iter, ifs, .. } = &generators[0];
+
+    let iter_ty = iter.custom.unwrap();
     let iter_val = if let Some(v) = generator.gen_expr(ctx, iter)? {
-        v.to_basic_value_enum(ctx, generator, iter.custom.unwrap())?
+        v.to_basic_value_enum(ctx, generator, iter_ty)?
     } else {
         for bb in [test_bb, body_bb, cont_bb] {
             ctx.builder.position_at_end(bb);
@@ -1014,96 +1135,120 @@ pub fn gen_comprehension<'ctx, G: CodeGenerator>(
     ctx.builder.build_store(index, zero_size_t).unwrap();
 
     let elem_ty = ctx.get_llvm_type(generator, elt.custom.unwrap());
-    let is_range = ctx.unifier.unioned(iter.custom.unwrap(), ctx.primitives.range);
     let list;
 
-    if is_range {
-        let iter_val = RangeValue::from_ptr_val(iter_val.into_pointer_value(), Some("range"));
-        let (start, stop, step) = destructure_range(ctx, iter_val);
-        let diff = ctx.builder.build_int_sub(stop, start, "diff").unwrap();
-        // add 1 to the length as the value is rounded to zero
-        // the length may be 1 more than the actual length if the division is exact, but the
-        // length is a upper bound only anyway so it does not matter.
-        let length = ctx.builder.build_int_signed_div(diff, step, "div").unwrap();
-        let length = ctx.builder.build_int_add(length, int32.const_int(1, false), "add1").unwrap();
-        // in case length is non-positive
-        let is_valid =
-            ctx.builder.build_int_compare(IntPredicate::SGT, length, zero_32, "check").unwrap();
+    match &*ctx.unifier.get_ty(iter_ty) {
+        TypeEnum::TObj { obj_id, .. }
+            if *obj_id == ctx.primitives.range.obj_id(&ctx.unifier).unwrap() =>
+        {
+            let iter_val = RangeValue::from_ptr_val(iter_val.into_pointer_value(), Some("range"));
+            let (start, stop, step) = destructure_range(ctx, iter_val);
+            let diff = ctx.builder.build_int_sub(stop, start, "diff").unwrap();
+            // add 1 to the length as the value is rounded to zero
+            // the length may be 1 more than the actual length if the division is exact, but the
+            // length is a upper bound only anyway so it does not matter.
+            let length = ctx.builder.build_int_signed_div(diff, step, "div").unwrap();
+            let length =
+                ctx.builder.build_int_add(length, int32.const_int(1, false), "add1").unwrap();
+            // in case length is non-positive
+            let is_valid =
+                ctx.builder.build_int_compare(IntPredicate::SGT, length, zero_32, "check").unwrap();
 
-        let list_alloc_size = ctx
-            .builder
-            .build_select(
-                is_valid,
-                ctx.builder.build_int_z_extend_or_bit_cast(length, size_t, "z_ext_len").unwrap(),
-                zero_size_t,
-                "listcomp.alloc_size",
-            )
-            .unwrap();
-        list = allocate_list(
-            generator,
-            ctx,
-            Some(elem_ty),
-            list_alloc_size.into_int_value(),
-            Some("listcomp.addr"),
-        );
+            let list_alloc_size = ctx
+                .builder
+                .build_select(
+                    is_valid,
+                    ctx.builder
+                        .build_int_z_extend_or_bit_cast(length, size_t, "z_ext_len")
+                        .unwrap(),
+                    zero_size_t,
+                    "listcomp.alloc_size",
+                )
+                .unwrap();
+            list = allocate_list(
+                generator,
+                ctx,
+                Some(elem_ty),
+                list_alloc_size.into_int_value(),
+                Some("listcomp.addr"),
+            );
 
-        let i = generator.gen_store_target(ctx, target, Some("i.addr"))?.unwrap();
-        ctx.builder
-            .build_store(i, ctx.builder.build_int_sub(start, step, "start_init").unwrap())
-            .unwrap();
+            let i = generator.gen_store_target(ctx, target, Some("i.addr"))?.unwrap();
+            ctx.builder
+                .build_store(i, ctx.builder.build_int_sub(start, step, "start_init").unwrap())
+                .unwrap();
 
-        ctx.builder
-            .build_conditional_branch(gen_in_range_check(ctx, start, stop, step), test_bb, cont_bb)
-            .unwrap();
+            ctx.builder
+                .build_conditional_branch(
+                    gen_in_range_check(ctx, start, stop, step),
+                    test_bb,
+                    cont_bb,
+                )
+                .unwrap();
 
-        ctx.builder.position_at_end(test_bb);
-        // add and test
-        let tmp = ctx
-            .builder
-            .build_int_add(
-                ctx.builder.build_load(i, "i").map(BasicValueEnum::into_int_value).unwrap(),
-                step,
-                "start_loop",
-            )
-            .unwrap();
-        ctx.builder.build_store(i, tmp).unwrap();
-        ctx.builder
-            .build_conditional_branch(gen_in_range_check(ctx, tmp, stop, step), body_bb, cont_bb)
-            .unwrap();
+            ctx.builder.position_at_end(test_bb);
+            // add and test
+            let tmp = ctx
+                .builder
+                .build_int_add(
+                    ctx.builder.build_load(i, "i").map(BasicValueEnum::into_int_value).unwrap(),
+                    step,
+                    "start_loop",
+                )
+                .unwrap();
+            ctx.builder.build_store(i, tmp).unwrap();
+            ctx.builder
+                .build_conditional_branch(
+                    gen_in_range_check(ctx, tmp, stop, step),
+                    body_bb,
+                    cont_bb,
+                )
+                .unwrap();
 
-        ctx.builder.position_at_end(body_bb);
-    } else {
-        let length = ctx
-            .build_gep_and_load(
-                iter_val.into_pointer_value(),
-                &[zero_size_t, int32.const_int(1, false)],
-                Some("length"),
-            )
-            .into_int_value();
-        list = allocate_list(generator, ctx, Some(elem_ty), length, Some("listcomp"));
+            ctx.builder.position_at_end(body_bb);
+        }
+        TypeEnum::TObj { obj_id, .. }
+            if *obj_id == ctx.primitives.list.obj_id(&ctx.unifier).unwrap() =>
+        {
+            let length = ctx
+                .build_gep_and_load(
+                    iter_val.into_pointer_value(),
+                    &[zero_size_t, int32.const_int(1, false)],
+                    Some("length"),
+                )
+                .into_int_value();
+            list = allocate_list(generator, ctx, Some(elem_ty), length, Some("listcomp"));
 
-        let counter = generator.gen_var_alloc(ctx, size_t.into(), Some("counter.addr"))?;
-        // counter = -1
-        ctx.builder.build_store(counter, size_t.const_all_ones()).unwrap();
-        ctx.builder.build_unconditional_branch(test_bb).unwrap();
+            let counter = generator.gen_var_alloc(ctx, size_t.into(), Some("counter.addr"))?;
+            // counter = -1
+            ctx.builder.build_store(counter, size_t.const_all_ones()).unwrap();
+            ctx.builder.build_unconditional_branch(test_bb).unwrap();
 
-        ctx.builder.position_at_end(test_bb);
-        let tmp = ctx.builder.build_load(counter, "i").map(BasicValueEnum::into_int_value).unwrap();
-        let tmp = ctx.builder.build_int_add(tmp, size_t.const_int(1, false), "inc").unwrap();
-        ctx.builder.build_store(counter, tmp).unwrap();
-        let cmp = ctx.builder.build_int_compare(IntPredicate::SLT, tmp, length, "cmp").unwrap();
-        ctx.builder.build_conditional_branch(cmp, body_bb, cont_bb).unwrap();
+            ctx.builder.position_at_end(test_bb);
+            let tmp =
+                ctx.builder.build_load(counter, "i").map(BasicValueEnum::into_int_value).unwrap();
+            let tmp = ctx.builder.build_int_add(tmp, size_t.const_int(1, false), "inc").unwrap();
+            ctx.builder.build_store(counter, tmp).unwrap();
+            let cmp = ctx.builder.build_int_compare(IntPredicate::SLT, tmp, length, "cmp").unwrap();
+            ctx.builder.build_conditional_branch(cmp, body_bb, cont_bb).unwrap();
 
-        ctx.builder.position_at_end(body_bb);
-        let arr_ptr = ctx
-            .build_gep_and_load(
-                iter_val.into_pointer_value(),
-                &[zero_size_t, zero_32],
-                Some("arr.addr"),
-            )
-            .into_pointer_value();
-        let val = ctx.build_gep_and_load(arr_ptr, &[tmp], Some("val"));
-        generator.gen_assign(ctx, target, val.into())?;
+            ctx.builder.position_at_end(body_bb);
+            let arr_ptr = ctx
+                .build_gep_and_load(
+                    iter_val.into_pointer_value(),
+                    &[zero_size_t, zero_32],
+                    Some("arr.addr"),
+                )
+                .into_pointer_value();
+            let val = ctx.build_gep_and_load(arr_ptr, &[tmp], Some("val"));
+            generator.gen_assign(ctx, target, val.into(), elt.custom.unwrap())?;
+        }
+        _ => {
+            panic!(
+                "unsupported list comprehension iterator type: {}",
+                ctx.unifier.stringify(iter_ty)
+            );
+        }
     }
 
     // Emits the content of `cont_bb`

nac3core/src/codegen/extern_fns.rs

@@ -130,3 +130,62 @@ pub fn call_ldexp<'ctx>(
         .map(Either::unwrap_left)
         .unwrap()
 }
+
+/// 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:literal`: 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, $extern_fn:literal, 2) => {
+        generate_linalg_extern_fn!($fn_name, $extern_fn, mat1, mat2);
+    };
+    ($fn_name:ident, $extern_fn:literal, 3) => {
+        generate_linalg_extern_fn!($fn_name, $extern_fn, mat1, mat2, mat3);
+    };
+    ($fn_name:ident, $extern_fn:literal, 4) => {
+        generate_linalg_extern_fn!($fn_name, $extern_fn, mat1, mat2, mat3, mat4);
+    };
+    ($fn_name:ident, $extern_fn:literal $(,$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 = $extern_fn;
+            let extern_fn = ctx.module.get_function(FN_NAME).unwrap_or_else(|| {
+                let fn_type = ctx.ctx.void_type().fn_type(&[$($input_matrix.get_type().into()),*], false);
+
+                let func = ctx.module.add_function(FN_NAME, fn_type, None);
+                for attr in ["mustprogress", "nofree", "nounwind", "willreturn", "writeonly"] {
+                    func.add_attribute(
+                        AttributeLoc::Function,
+                        ctx.ctx.create_enum_attribute(Attribute::get_named_enum_kind_id(attr), 0),
+                    );
+                }
+                func
+            });
+
+            ctx.builder.build_call(extern_fn, &[$($input_matrix.into(),)*], name.unwrap_or_default()).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);

nac3core/src/codegen/generator.rs

@@ -123,11 +123,45 @@ 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)
+        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)
     }
 
     /// Generate code for a while expression.

nac3core/src/codegen/irrt/irrt.cpp

@@ -1,17 +1,27 @@
-#pragma once
-
-#include <irrt/int_defs.hpp>
-#include <irrt/utils.hpp>
+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);
 
 // NDArray indices are always `uint32_t`.
 using NDIndex = uint32_t;
-// The type of an index or a value describing the length of a
-// range/slice is always `int32_t`.
+// The type of an index or a value describing the length of a range/slice is always `int32_t`.
 using SliceIndex = int32_t;
 
 namespace {
-// adapted from GNU Scientific Library:
-// https://git.savannah.gnu.org/cgit/gsl.git/tree/sys/pow_int.c
+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;
+}
+
+// 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) {
@@ -28,8 +38,12 @@ T __nac3_int_exp_impl(T base, T exp) {
 }
 
 template <typename SizeT>
-SizeT __nac3_ndarray_calc_size_impl(const SizeT* list_data, SizeT list_len,
-                                    SizeT begin_idx, SizeT end_idx) {
+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;
@@ -42,8 +56,12 @@ SizeT __nac3_ndarray_calc_size_impl(const SizeT* list_data, SizeT list_len,
 }
 
 template <typename SizeT>
-void __nac3_ndarray_calc_nd_indices_impl(SizeT index, const SizeT* dims,
-                                         SizeT num_dims, NDIndex* idxs) {
+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;
@@ -54,9 +72,12 @@ void __nac3_ndarray_calc_nd_indices_impl(SizeT index, const SizeT* dims,
 }
 
 template <typename SizeT>
-SizeT __nac3_ndarray_flatten_index_impl(const SizeT* dims, SizeT num_dims,
-                                        const NDIndex* indices,
-                                        SizeT num_indices) {
+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) {
@@ -72,17 +93,18 @@ SizeT __nac3_ndarray_flatten_index_impl(const SizeT* dims, SizeT num_dims,
 }
 
 template <typename SizeT>
-void __nac3_ndarray_calc_broadcast_impl(const SizeT* lhs_dims, SizeT lhs_ndims,
-                                        const SizeT* rhs_dims, SizeT rhs_ndims,
-                                        SizeT* out_dims) {
+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;
-
+        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) {
@@ -102,10 +124,12 @@ void __nac3_ndarray_calc_broadcast_impl(const SizeT* lhs_dims, SizeT lhs_ndims,
 }
 
 template <typename SizeT>
-void __nac3_ndarray_calc_broadcast_idx_impl(const SizeT* src_dims,
-                                            SizeT src_ndims,
-                                            const NDIndex* in_idx,
-                                            NDIndex* out_idx) {
+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];
@@ -114,15 +138,15 @@ void __nac3_ndarray_calc_broadcast_idx_impl(const SizeT* src_dims,
 }  // namespace
 
 extern "C" {
-#define DEF_nac3_int_exp_(T)                   \
-    T __nac3_int_exp_##T(T base, T exp) {      \
-        return __nac3_int_exp_impl(base, exp); \
+#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);
+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) {
@@ -136,8 +160,11 @@ SliceIndex __nac3_slice_index_bound(SliceIndex i, const SliceIndex len) {
     return i;
 }
 
-SliceIndex __nac3_range_slice_len(const SliceIndex start, const SliceIndex end,
-                                  const SliceIndex step) {
+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;
@@ -153,52 +180,62 @@ SliceIndex __nac3_range_slice_len(const SliceIndex start, const SliceIndex end,
 // - 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)
+// 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
-     */
+    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 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;
+        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);
+            __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);
+            __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));
+        (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));
+        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 (;
+        (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);
@@ -207,26 +244,30 @@ SliceIndex __nac3_list_slice_assign_var_size(
         } 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);
+            /* 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 + size + size + size);
+            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_isinf(double x) {
+    return __builtin_isinf(x);
+}
 
-int32_t __nac3_isnan(double x) { return __builtin_isnan(x); }
+int32_t __nac3_isnan(double x) {
+    return __builtin_isnan(x);
+}
 
 double tgamma(double arg);
 
@@ -279,71 +320,95 @@ double __nac3_j0(double x) {
     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);
+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);
+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) {
+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) {
+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);
+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);
+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_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_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_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_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);
 }
 }  // extern "C"

nac3core/src/codegen/irrt/mod.rs

@@ -1,7 +1,5 @@
 use crate::typecheck::typedef::Type;
 
-mod test;
-
 use super::{
     classes::{
         ArrayLikeIndexer, ArrayLikeValue, ArraySliceValue, ListValue, NDArrayValue,

nac3core/src/codegen/irrt/test.rs

@@ -1,26 +0,0 @@
-#[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

@@ -35,6 +35,40 @@ fn get_float_intrinsic_repr(ctx: &Context, ft: FloatType) -> &'static str {
     unreachable!()
 }
 
+/// 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";
+
+    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_start`](https://llvm.org/docs/LangRef.html#llvm-va-start-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();
+}
+
 /// Invokes the [`llvm.stacksave`](https://llvm.org/docs/LangRef.html#llvm-stacksave-intrinsic)
 /// intrinsic.
 pub fn call_stacksave<'ctx>(

nac3core/src/codegen/mod.rs

@@ -444,7 +444,7 @@ pub struct CodeGenTask {
 fn get_llvm_type<'ctx, G: CodeGenerator + ?Sized>(
     ctx: &'ctx Context,
     module: &Module<'ctx>,
-    generator: &mut G,
+    generator: &G,
     unifier: &mut Unifier,
     top_level: &TopLevelContext,
     type_cache: &mut HashMap<Type, BasicTypeEnum<'ctx>>,
@@ -538,8 +538,10 @@ fn get_llvm_type<'ctx, G: CodeGenerator + ?Sized>(
                 };
                 return ty;
             }
-            TTuple { ty } => {
+            TTuple { ty, is_vararg_ctx } => {
                 // 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| {
@@ -569,7 +571,7 @@ fn get_llvm_type<'ctx, G: CodeGenerator + ?Sized>(
 fn get_llvm_abi_type<'ctx, G: CodeGenerator + ?Sized>(
     ctx: &'ctx Context,
     module: &Module<'ctx>,
-    generator: &mut G,
+    generator: &G,
     unifier: &mut Unifier,
     top_level: &TopLevelContext,
     type_cache: &mut HashMap<Type, BasicTypeEnum<'ctx>>,
@@ -607,6 +609,40 @@ 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,
@@ -718,6 +754,7 @@ 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),
@@ -740,7 +777,10 @@ pub fn gen_func_impl<
     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,
@@ -759,9 +799,12 @@ 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, false),
-        _ => context.void_type().fn_type(&params, false),
+        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()),
     };
 
     let symbol = &task.symbol_name;
@@ -791,7 +834,9 @@ pub fn gen_func_impl<
 
     let mut var_assignment = HashMap::new();
     let offset = u32::from(has_sret);
-    for (n, arg) in args.iter().enumerate() {
+
+    // Store non-vararg argument values into local variables
+    for (n, arg) in args.iter().enumerate().filter(|(_, arg)| !arg.is_vararg) {
         let param = fn_val.get_nth_param((n as u32) + offset).unwrap();
         let local_type = get_llvm_type(
             context,
@@ -824,6 +869,8 @@ 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 {
@@ -1046,3 +1093,9 @@ 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()
+}

nac3core/src/codegen/numpy.rs

@@ -26,12 +26,15 @@ use crate::{
         typedef::{FunSignature, Type, TypeEnum},
     },
 };
-use inkwell::types::{AnyTypeEnum, BasicTypeEnum, PointerType};
 use inkwell::{
     types::BasicType,
     values::{BasicValueEnum, IntValue, PointerValue},
     AddressSpace, IntPredicate, OptimizationLevel,
 };
+use inkwell::{
+    types::{AnyTypeEnum, BasicTypeEnum, PointerType},
+    values::BasicValue,
+};
 use nac3parser::ast::{Operator, StrRef};
 
 /// Creates an uninitialized `NDArray` instance.
@@ -159,7 +162,7 @@ where
 ///
 /// * `elem_ty` - The element type of the `NDArray`.
 /// * `shape` - The shape of the `NDArray`, represented am array of [`IntValue`]s.
-fn create_ndarray_const_shape<'ctx, G: CodeGenerator + ?Sized>(
+pub fn create_ndarray_const_shape<'ctx, G: CodeGenerator + ?Sized>(
     generator: &mut G,
     ctx: &mut CodeGenContext<'ctx, '_>,
     elem_ty: Type,
@@ -2026,3 +2029,493 @@ pub fn gen_ndarray_fill<'ctx>(
 
     Ok(())
 }
+
+/// Generates LLVM IR for `ndarray.transpose`.
+pub fn ndarray_transpose<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    x1: (Type, BasicValueEnum<'ctx>),
+) -> Result<BasicValueEnum<'ctx>, String> {
+    const FN_NAME: &str = "ndarray_transpose";
+    let (x1_ty, x1) = x1;
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+
+    if let BasicValueEnum::PointerValue(n1) = x1 {
+        let (elem_ty, _) = unpack_ndarray_var_tys(&mut ctx.unifier, x1_ty);
+        let n1 = NDArrayValue::from_ptr_val(n1, llvm_usize, None);
+        let n_sz = call_ndarray_calc_size(generator, ctx, &n1.dim_sizes(), (None, None));
+
+        // Dimensions are reversed in the transposed array
+        let out = create_ndarray_dyn_shape(
+            generator,
+            ctx,
+            elem_ty,
+            &n1,
+            |_, ctx, n| Ok(n.load_ndims(ctx)),
+            |generator, ctx, n, idx| {
+                let new_idx = ctx.builder.build_int_sub(n.load_ndims(ctx), idx, "").unwrap();
+                let new_idx = ctx
+                    .builder
+                    .build_int_sub(new_idx, new_idx.get_type().const_int(1, false), "")
+                    .unwrap();
+                unsafe { Ok(n.dim_sizes().get_typed_unchecked(ctx, generator, &new_idx, None)) }
+            },
+        )
+        .unwrap();
+
+        gen_for_callback_incrementing(
+            generator,
+            ctx,
+            None,
+            llvm_usize.const_zero(),
+            (n_sz, false),
+            |generator, ctx, _, idx| {
+                let elem = unsafe { n1.data().get_unchecked(ctx, generator, &idx, None) };
+
+                let new_idx = generator.gen_var_alloc(ctx, llvm_usize.into(), None)?;
+                let rem_idx = generator.gen_var_alloc(ctx, llvm_usize.into(), None)?;
+                ctx.builder.build_store(new_idx, llvm_usize.const_zero()).unwrap();
+                ctx.builder.build_store(rem_idx, idx).unwrap();
+
+                // Incrementally calculate the new index in the transposed array
+                // For each index, we first decompose it into the n-dims and use those to reconstruct the new index
+                // The formula used for indexing is:
+                // idx = dim_n * ( ... (dim2 * (dim0 * dim1) + dim1) + dim2 ... ) + dim_n
+                gen_for_callback_incrementing(
+                    generator,
+                    ctx,
+                    None,
+                    llvm_usize.const_zero(),
+                    (n1.load_ndims(ctx), false),
+                    |generator, ctx, _, ndim| {
+                        let ndim_rev =
+                            ctx.builder.build_int_sub(n1.load_ndims(ctx), ndim, "").unwrap();
+                        let ndim_rev = ctx
+                            .builder
+                            .build_int_sub(ndim_rev, llvm_usize.const_int(1, false), "")
+                            .unwrap();
+                        let dim = unsafe {
+                            n1.dim_sizes().get_typed_unchecked(ctx, generator, &ndim_rev, None)
+                        };
+
+                        let rem_idx_val =
+                            ctx.builder.build_load(rem_idx, "").unwrap().into_int_value();
+                        let new_idx_val =
+                            ctx.builder.build_load(new_idx, "").unwrap().into_int_value();
+
+                        let add_component =
+                            ctx.builder.build_int_unsigned_rem(rem_idx_val, dim, "").unwrap();
+                        let rem_idx_val =
+                            ctx.builder.build_int_unsigned_div(rem_idx_val, dim, "").unwrap();
+
+                        let new_idx_val = ctx.builder.build_int_mul(new_idx_val, dim, "").unwrap();
+                        let new_idx_val =
+                            ctx.builder.build_int_add(new_idx_val, add_component, "").unwrap();
+
+                        ctx.builder.build_store(rem_idx, rem_idx_val).unwrap();
+                        ctx.builder.build_store(new_idx, new_idx_val).unwrap();
+
+                        Ok(())
+                    },
+                    llvm_usize.const_int(1, false),
+                )?;
+
+                let new_idx_val = ctx.builder.build_load(new_idx, "").unwrap().into_int_value();
+                unsafe { out.data().set_unchecked(ctx, generator, &new_idx_val, elem) };
+                Ok(())
+            },
+            llvm_usize.const_int(1, false),
+        )?;
+
+        Ok(out.as_base_value().into())
+    } else {
+        unreachable!(
+            "{FN_NAME}() not supported for '{}'",
+            format!("'{}'", ctx.unifier.stringify(x1_ty))
+        )
+    }
+}
+
+/// LLVM-typed implementation for generating the implementation for `ndarray.reshape`.
+///
+/// * `x1` - `NDArray` to reshape.
+/// * `shape` - The `shape` parameter used to construct the new `NDArray`.
+/// Just like numpy, the `shape` argument can be:
+///   1. A list of `int32`;   e.g., `np.reshape(arr, [600, -1, 3])`
+///   2. A tuple of `int32`;  e.g., `np.reshape(arr, (-1, 800, 3))`
+///   3. A scalar `int32`;     e.g., `np.reshape(arr, 3)`
+/// Note that unlike other generating functions, one of the dimesions in the shape can be negative
+pub fn ndarray_reshape<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    x1: (Type, BasicValueEnum<'ctx>),
+    shape: (Type, BasicValueEnum<'ctx>),
+) -> Result<BasicValueEnum<'ctx>, String> {
+    const FN_NAME: &str = "ndarray_reshape";
+    let (x1_ty, x1) = x1;
+    let (_, shape) = shape;
+
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+
+    if let BasicValueEnum::PointerValue(n1) = x1 {
+        let (elem_ty, _) = unpack_ndarray_var_tys(&mut ctx.unifier, x1_ty);
+        let n1 = NDArrayValue::from_ptr_val(n1, llvm_usize, None);
+        let n_sz = call_ndarray_calc_size(generator, ctx, &n1.dim_sizes(), (None, None));
+
+        let acc = generator.gen_var_alloc(ctx, llvm_usize.into(), None)?;
+        let num_neg = generator.gen_var_alloc(ctx, llvm_usize.into(), None)?;
+        ctx.builder.build_store(acc, llvm_usize.const_int(1, false)).unwrap();
+        ctx.builder.build_store(num_neg, llvm_usize.const_zero()).unwrap();
+
+        let out = match shape {
+            BasicValueEnum::PointerValue(shape_list_ptr)
+                if ListValue::is_instance(shape_list_ptr, llvm_usize).is_ok() =>
+            {
+                // 1. A list of ints; e.g., `np.reshape(arr, [int64(600), int64(800, -1])`
+
+                let shape_list = ListValue::from_ptr_val(shape_list_ptr, llvm_usize, None);
+                // Check for -1 in dimensions
+                gen_for_callback_incrementing(
+                    generator,
+                    ctx,
+                    None,
+                    llvm_usize.const_zero(),
+                    (shape_list.load_size(ctx, None), false),
+                    |generator, ctx, _, idx| {
+                        let ele =
+                            shape_list.data().get(ctx, generator, &idx, None).into_int_value();
+                        let ele = ctx.builder.build_int_s_extend(ele, llvm_usize, "").unwrap();
+
+                        gen_if_else_expr_callback(
+                            generator,
+                            ctx,
+                            |_, ctx| {
+                                Ok(ctx
+                                    .builder
+                                    .build_int_compare(
+                                        IntPredicate::SLT,
+                                        ele,
+                                        llvm_usize.const_zero(),
+                                        "",
+                                    )
+                                    .unwrap())
+                            },
+                            |_, ctx| -> Result<Option<IntValue>, String> {
+                                let num_neg_value =
+                                    ctx.builder.build_load(num_neg, "").unwrap().into_int_value();
+                                let num_neg_value = ctx
+                                    .builder
+                                    .build_int_add(
+                                        num_neg_value,
+                                        llvm_usize.const_int(1, false),
+                                        "",
+                                    )
+                                    .unwrap();
+                                ctx.builder.build_store(num_neg, num_neg_value).unwrap();
+                                Ok(None)
+                            },
+                            |_, ctx| {
+                                let acc_value =
+                                    ctx.builder.build_load(acc, "").unwrap().into_int_value();
+                                let acc_value =
+                                    ctx.builder.build_int_mul(acc_value, ele, "").unwrap();
+                                ctx.builder.build_store(acc, acc_value).unwrap();
+                                Ok(None)
+                            },
+                        )?;
+                        Ok(())
+                    },
+                    llvm_usize.const_int(1, false),
+                )?;
+                let acc_val = ctx.builder.build_load(acc, "").unwrap().into_int_value();
+                let rem = ctx.builder.build_int_unsigned_div(n_sz, acc_val, "").unwrap();
+                // Generate the output shape by filling -1 with `rem`
+                create_ndarray_dyn_shape(
+                    generator,
+                    ctx,
+                    elem_ty,
+                    &shape_list,
+                    |_, ctx, _| Ok(shape_list.load_size(ctx, None)),
+                    |generator, ctx, shape_list, idx| {
+                        let dim =
+                            shape_list.data().get(ctx, generator, &idx, None).into_int_value();
+                        let dim = ctx.builder.build_int_s_extend(dim, llvm_usize, "").unwrap();
+
+                        Ok(gen_if_else_expr_callback(
+                            generator,
+                            ctx,
+                            |_, ctx| {
+                                Ok(ctx
+                                    .builder
+                                    .build_int_compare(
+                                        IntPredicate::SLT,
+                                        dim,
+                                        llvm_usize.const_zero(),
+                                        "",
+                                    )
+                                    .unwrap())
+                            },
+                            |_, _| Ok(Some(rem)),
+                            |_, _| Ok(Some(dim)),
+                        )?
+                        .unwrap()
+                        .into_int_value())
+                    },
+                )
+            }
+            BasicValueEnum::StructValue(shape_tuple) => {
+                // 2. A tuple of `int32`;  e.g., `np.reshape(arr, (-1, 800, 3))`
+
+                let ndims = shape_tuple.get_type().count_fields();
+                // Check for -1 in dims
+                for dim_i in 0..ndims {
+                    let dim = ctx
+                        .builder
+                        .build_extract_value(shape_tuple, dim_i, "")
+                        .unwrap()
+                        .into_int_value();
+                    let dim = ctx.builder.build_int_s_extend(dim, llvm_usize, "").unwrap();
+
+                    gen_if_else_expr_callback(
+                        generator,
+                        ctx,
+                        |_, ctx| {
+                            Ok(ctx
+                                .builder
+                                .build_int_compare(
+                                    IntPredicate::SLT,
+                                    dim,
+                                    llvm_usize.const_zero(),
+                                    "",
+                                )
+                                .unwrap())
+                        },
+                        |_, ctx| -> Result<Option<IntValue>, String> {
+                            let num_negs =
+                                ctx.builder.build_load(num_neg, "").unwrap().into_int_value();
+                            let num_negs = ctx
+                                .builder
+                                .build_int_add(num_negs, llvm_usize.const_int(1, false), "")
+                                .unwrap();
+                            ctx.builder.build_store(num_neg, num_negs).unwrap();
+                            Ok(None)
+                        },
+                        |_, ctx| {
+                            let acc_val = ctx.builder.build_load(acc, "").unwrap().into_int_value();
+                            let acc_val = ctx.builder.build_int_mul(acc_val, dim, "").unwrap();
+                            ctx.builder.build_store(acc, acc_val).unwrap();
+                            Ok(None)
+                        },
+                    )?;
+                }
+
+                let acc_val = ctx.builder.build_load(acc, "").unwrap().into_int_value();
+                let rem = ctx.builder.build_int_unsigned_div(n_sz, acc_val, "").unwrap();
+                let mut shape = Vec::with_capacity(ndims as usize);
+
+                // Reconstruct shape filling negatives with rem
+                for dim_i in 0..ndims {
+                    let dim = ctx
+                        .builder
+                        .build_extract_value(shape_tuple, dim_i, "")
+                        .unwrap()
+                        .into_int_value();
+                    let dim = ctx.builder.build_int_s_extend(dim, llvm_usize, "").unwrap();
+
+                    let dim = gen_if_else_expr_callback(
+                        generator,
+                        ctx,
+                        |_, ctx| {
+                            Ok(ctx
+                                .builder
+                                .build_int_compare(
+                                    IntPredicate::SLT,
+                                    dim,
+                                    llvm_usize.const_zero(),
+                                    "",
+                                )
+                                .unwrap())
+                        },
+                        |_, _| Ok(Some(rem)),
+                        |_, _| Ok(Some(dim)),
+                    )?
+                    .unwrap()
+                    .into_int_value();
+                    shape.push(dim);
+                }
+                create_ndarray_const_shape(generator, ctx, elem_ty, shape.as_slice())
+            }
+            BasicValueEnum::IntValue(shape_int) => {
+                // 3. A scalar `int32`;     e.g., `np.reshape(arr, 3)`
+                let shape_int = gen_if_else_expr_callback(
+                    generator,
+                    ctx,
+                    |_, ctx| {
+                        Ok(ctx
+                            .builder
+                            .build_int_compare(
+                                IntPredicate::SLT,
+                                shape_int,
+                                llvm_usize.const_zero(),
+                                "",
+                            )
+                            .unwrap())
+                    },
+                    |_, _| Ok(Some(n_sz)),
+                    |_, ctx| {
+                        Ok(Some(ctx.builder.build_int_s_extend(shape_int, llvm_usize, "").unwrap()))
+                    },
+                )?
+                .unwrap()
+                .into_int_value();
+                create_ndarray_const_shape(generator, ctx, elem_ty, &[shape_int])
+            }
+            _ => unreachable!(),
+        }
+        .unwrap();
+
+        // Only allow one dimension to be negative
+        let num_negs = ctx.builder.build_load(num_neg, "").unwrap().into_int_value();
+        ctx.make_assert(
+            generator,
+            ctx.builder
+                .build_int_compare(IntPredicate::ULT, num_negs, llvm_usize.const_int(2, false), "")
+                .unwrap(),
+            "0:ValueError",
+            "can only specify one unknown dimension",
+            [None, None, None],
+            ctx.current_loc,
+        );
+
+        // The new shape must be compatible with the old shape
+        let out_sz = call_ndarray_calc_size(generator, ctx, &out.dim_sizes(), (None, None));
+        ctx.make_assert(
+            generator,
+            ctx.builder.build_int_compare(IntPredicate::EQ, out_sz, n_sz, "").unwrap(),
+            "0:ValueError",
+            "cannot reshape array of size {0} into provided shape of size {1}",
+            [Some(n_sz), Some(out_sz), None],
+            ctx.current_loc,
+        );
+
+        gen_for_callback_incrementing(
+            generator,
+            ctx,
+            None,
+            llvm_usize.const_zero(),
+            (n_sz, false),
+            |generator, ctx, _, idx| {
+                let elem = unsafe { n1.data().get_unchecked(ctx, generator, &idx, None) };
+                unsafe { out.data().set_unchecked(ctx, generator, &idx, elem) };
+                Ok(())
+            },
+            llvm_usize.const_int(1, false),
+        )?;
+
+        Ok(out.as_base_value().into())
+    } else {
+        unreachable!(
+            "{FN_NAME}() not supported for '{}'",
+            format!("'{}'", ctx.unifier.stringify(x1_ty))
+        )
+    }
+}
+
+/// Generates LLVM IR for `ndarray.dot`.
+/// Calculate inner product of two vectors or literals
+/// For matrix multiplication use `np_matmul`
+///
+/// The input `NDArray` are flattened and treated as 1D
+/// The operation is equivalent to `np.dot(arr1.ravel(), arr2.ravel())`
+pub fn ndarray_dot<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    x1: (Type, BasicValueEnum<'ctx>),
+    x2: (Type, BasicValueEnum<'ctx>),
+) -> Result<BasicValueEnum<'ctx>, String> {
+    const FN_NAME: &str = "ndarray_dot";
+    let (x1_ty, x1) = x1;
+    let (_, x2) = x2;
+
+    let llvm_usize = generator.get_size_type(ctx.ctx);
+
+    match (x1, x2) {
+        (BasicValueEnum::PointerValue(n1), BasicValueEnum::PointerValue(n2)) => {
+            let n1 = NDArrayValue::from_ptr_val(n1, llvm_usize, None);
+            let n2 = NDArrayValue::from_ptr_val(n2, llvm_usize, None);
+
+            let n1_sz = call_ndarray_calc_size(generator, ctx, &n1.dim_sizes(), (None, None));
+            let n2_sz = call_ndarray_calc_size(generator, ctx, &n1.dim_sizes(), (None, None));
+
+            ctx.make_assert(
+                generator,
+                ctx.builder.build_int_compare(IntPredicate::EQ, n1_sz, n2_sz, "").unwrap(),
+                "0:ValueError",
+                "shapes ({0}), ({1}) not aligned",
+                [Some(n1_sz), Some(n2_sz), None],
+                ctx.current_loc,
+            );
+
+            let identity =
+                unsafe { n1.data().get_unchecked(ctx, generator, &llvm_usize.const_zero(), None) };
+            let acc = ctx.builder.build_alloca(identity.get_type(), "").unwrap();
+            ctx.builder.build_store(acc, identity.get_type().const_zero()).unwrap();
+
+            gen_for_callback_incrementing(
+                generator,
+                ctx,
+                None,
+                llvm_usize.const_zero(),
+                (n1_sz, false),
+                |generator, ctx, _, idx| {
+                    let elem1 = unsafe { n1.data().get_unchecked(ctx, generator, &idx, None) };
+                    let elem2 = unsafe { n2.data().get_unchecked(ctx, generator, &idx, None) };
+
+                    let product = match elem1 {
+                        BasicValueEnum::IntValue(e1) => ctx
+                            .builder
+                            .build_int_mul(e1, elem2.into_int_value(), "")
+                            .unwrap()
+                            .as_basic_value_enum(),
+                        BasicValueEnum::FloatValue(e1) => ctx
+                            .builder
+                            .build_float_mul(e1, elem2.into_float_value(), "")
+                            .unwrap()
+                            .as_basic_value_enum(),
+                        _ => unreachable!(),
+                    };
+                    let acc_val = ctx.builder.build_load(acc, "").unwrap();
+                    let acc_val = match acc_val {
+                        BasicValueEnum::IntValue(e1) => ctx
+                            .builder
+                            .build_int_add(e1, product.into_int_value(), "")
+                            .unwrap()
+                            .as_basic_value_enum(),
+                        BasicValueEnum::FloatValue(e1) => ctx
+                            .builder
+                            .build_float_add(e1, product.into_float_value(), "")
+                            .unwrap()
+                            .as_basic_value_enum(),
+                        _ => unreachable!(),
+                    };
+                    ctx.builder.build_store(acc, acc_val).unwrap();
+
+                    Ok(())
+                },
+                llvm_usize.const_int(1, false),
+            )?;
+            let acc_val = ctx.builder.build_load(acc, "").unwrap();
+            Ok(acc_val)
+        }
+        (BasicValueEnum::IntValue(e1), BasicValueEnum::IntValue(e2)) => {
+            Ok(ctx.builder.build_int_mul(e1, e2, "").unwrap().as_basic_value_enum())
+        }
+        (BasicValueEnum::FloatValue(e1), BasicValueEnum::FloatValue(e2)) => {
+            Ok(ctx.builder.build_float_mul(e1, e2, "").unwrap().as_basic_value_enum())
+        }
+        _ => unreachable!(
+            "{FN_NAME}() not supported for '{}'",
+            format!("'{}'", ctx.unifier.stringify(x1_ty))
+        ),
+    }
+}

nac3core/src/codegen/stmt.rs

@@ -10,10 +10,10 @@ use crate::{
         expr::gen_binop_expr,
         gen_in_range_check,
     },
-    toplevel::{helper::PrimDef, numpy::unpack_ndarray_var_tys, DefinitionId, TopLevelDef},
+    toplevel::{DefinitionId, TopLevelDef},
     typecheck::{
         magic_methods::Binop,
-        typedef::{FunSignature, Type, TypeEnum},
+        typedef::{iter_type_vars, FunSignature, Type, TypeEnum},
     },
 };
 use inkwell::{
@@ -23,10 +23,10 @@ use inkwell::{
     values::{BasicValue, BasicValueEnum, FunctionValue, IntValue, PointerValue},
     IntPredicate,
 };
+use itertools::{izip, Itertools};
 use nac3parser::ast::{
     Constant, ExcepthandlerKind, Expr, ExprKind, Location, Stmt, StmtKind, StrRef,
 };
-use std::convert::TryFrom;
 
 /// See [`CodeGenerator::gen_var_alloc`].
 pub fn gen_var<'ctx>(
@@ -97,8 +97,6 @@ pub fn gen_store_target<'ctx, G: CodeGenerator>(
     pattern: &Expr<Option<Type>>,
     name: Option<&str>,
 ) -> Result<Option<PointerValue<'ctx>>, String> {
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-
     // very similar to gen_expr, but we don't do an extra load at the end
     // and we flatten nested tuples
     Ok(Some(match &pattern.node {
@@ -137,65 +135,6 @@ pub fn gen_store_target<'ctx, G: CodeGenerator>(
             }
             .unwrap()
         }
-        ExprKind::Subscript { value, slice, .. } => {
-            match ctx.unifier.get_ty_immutable(value.custom.unwrap()).as_ref() {
-                TypeEnum::TObj { obj_id, .. } if *obj_id == PrimDef::List.id() => {
-                    let v = generator
-                        .gen_expr(ctx, value)?
-                        .unwrap()
-                        .to_basic_value_enum(ctx, generator, value.custom.unwrap())?
-                        .into_pointer_value();
-                    let v = ListValue::from_ptr_val(v, llvm_usize, None);
-                    let len = v.load_size(ctx, Some("len"));
-                    let raw_index = generator
-                        .gen_expr(ctx, slice)?
-                        .unwrap()
-                        .to_basic_value_enum(ctx, generator, slice.custom.unwrap())?
-                        .into_int_value();
-                    let raw_index = ctx
-                        .builder
-                        .build_int_s_extend(raw_index, generator.get_size_type(ctx.ctx), "sext")
-                        .unwrap();
-                    // handle negative index
-                    let is_negative = ctx
-                        .builder
-                        .build_int_compare(
-                            IntPredicate::SLT,
-                            raw_index,
-                            generator.get_size_type(ctx.ctx).const_zero(),
-                            "is_neg",
-                        )
-                        .unwrap();
-                    let adjusted = ctx.builder.build_int_add(raw_index, len, "adjusted").unwrap();
-                    let index = ctx
-                        .builder
-                        .build_select(is_negative, adjusted, raw_index, "index")
-                        .map(BasicValueEnum::into_int_value)
-                        .unwrap();
-                    // unsigned less than is enough, because negative index after adjustment is
-                    // bigger than the length (for unsigned cmp)
-                    let bound_check = ctx
-                        .builder
-                        .build_int_compare(IntPredicate::ULT, index, len, "inbound")
-                        .unwrap();
-                    ctx.make_assert(
-                        generator,
-                        bound_check,
-                        "0:IndexError",
-                        "index {0} out of bounds 0:{1}",
-                        [Some(raw_index), Some(len), None],
-                        slice.location,
-                    );
-                    v.data().ptr_offset(ctx, generator, &index, name)
-                }
-
-                TypeEnum::TObj { obj_id, .. } if *obj_id == PrimDef::NDArray.id() => {
-                    todo!()
-                }
-
-                _ => unreachable!(),
-            }
-        }
         _ => unreachable!(),
     }))
 }
@@ -206,70 +145,20 @@ pub fn gen_assign<'ctx, G: CodeGenerator>(
     ctx: &mut CodeGenContext<'ctx, '_>,
     target: &Expr<Option<Type>>,
     value: ValueEnum<'ctx>,
+    value_ty: Type,
 ) -> Result<(), String> {
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-
+    // See https://docs.python.org/3/reference/simple_stmts.html#assignment-statements.
     match &target.node {
-        ExprKind::Tuple { elts, .. } => {
-            let BasicValueEnum::StructValue(v) =
-                value.to_basic_value_enum(ctx, generator, target.custom.unwrap())?
-            else {
-                unreachable!()
-            };
-
-            for (i, elt) in elts.iter().enumerate() {
-                let v = ctx
-                    .builder
-                    .build_extract_value(v, u32::try_from(i).unwrap(), "struct_elem")
-                    .unwrap();
-                generator.gen_assign(ctx, elt, v.into())?;
-            }
+        ExprKind::Subscript { value: target, slice: key, .. } => {
+            // Handle "slicing" or "subscription"
+            generator.gen_setitem(ctx, target, key, value, value_ty)?;
         }
-        ExprKind::Subscript { value: ls, slice, .. }
-            if matches!(&slice.node, ExprKind::Slice { .. }) =>
-        {
-            let ExprKind::Slice { lower, upper, step } = &slice.node else { unreachable!() };
-
-            let ls = generator
-                .gen_expr(ctx, ls)?
-                .unwrap()
-                .to_basic_value_enum(ctx, generator, ls.custom.unwrap())?
-                .into_pointer_value();
-            let ls = ListValue::from_ptr_val(ls, llvm_usize, None);
-            let Some((start, end, step)) =
-                handle_slice_indices(lower, upper, step, ctx, generator, ls.load_size(ctx, None))?
-            else {
-                return Ok(());
-            };
-            let value = value
-                .to_basic_value_enum(ctx, generator, target.custom.unwrap())?
-                .into_pointer_value();
-            let value = ListValue::from_ptr_val(value, llvm_usize, None);
-            let ty = match &*ctx.unifier.get_ty_immutable(target.custom.unwrap()) {
-                TypeEnum::TObj { obj_id, params, .. } if *obj_id == PrimDef::List.id() => {
-                    *params.iter().next().unwrap().1
-                }
-                TypeEnum::TObj { obj_id, .. } if *obj_id == PrimDef::NDArray.id() => {
-                    unpack_ndarray_var_tys(&mut ctx.unifier, target.custom.unwrap()).0
-                }
-                _ => unreachable!(),
-            };
-
-            let ty = ctx.get_llvm_type(generator, ty);
-            let Some(src_ind) = handle_slice_indices(
-                &None,
-                &None,
-                &None,
-                ctx,
-                generator,
-                value.load_size(ctx, None),
-            )?
-            else {
-                return Ok(());
-            };
-            list_slice_assignment(generator, ctx, ty, ls, (start, end, step), value, src_ind);
+        ExprKind::Tuple { elts, .. } | ExprKind::List { elts, .. } => {
+            // Fold on `"[" [target_list] "]"` and `"(" [target_list] ")"`
+            generator.gen_assign_target_list(ctx, elts, value, value_ty)?;
         }
         _ => {
+            // Handle attribute and direct variable assignments.
             let name = if let ExprKind::Name { id, .. } = &target.node {
                 format!("{id}.addr")
             } else {
@@ -293,6 +182,234 @@ pub fn gen_assign<'ctx, G: CodeGenerator>(
     Ok(())
 }
 
+/// See [`CodeGenerator::gen_assign_target_list`].
+pub fn gen_assign_target_list<'ctx, G: CodeGenerator>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    targets: &Vec<Expr<Option<Type>>>,
+    value: ValueEnum<'ctx>,
+    value_ty: Type,
+) -> Result<(), String> {
+    // Deconstruct the tuple `value`
+    let BasicValueEnum::StructValue(tuple) = value.to_basic_value_enum(ctx, generator, value_ty)?
+    else {
+        unreachable!()
+    };
+
+    // NOTE: Currently, RHS's type is forced to be a Tuple by the type inferencer.
+    let TypeEnum::TTuple { ty: tuple_tys, .. } = &*ctx.unifier.get_ty(value_ty) else {
+        unreachable!();
+    };
+
+    assert_eq!(tuple.get_type().count_fields() as usize, tuple_tys.len());
+
+    let tuple = (0..tuple.get_type().count_fields())
+        .map(|i| ctx.builder.build_extract_value(tuple, i, "item").unwrap())
+        .collect_vec();
+
+    // Find the starred target if it exists.
+    let mut starred_target_index: Option<usize> = None; // Index of the "starred" target. If it exists, there may only be one.
+    for (i, target) in targets.iter().enumerate() {
+        if matches!(target.node, ExprKind::Starred { .. }) {
+            assert!(starred_target_index.is_none()); // The typechecker ensures this
+            starred_target_index = Some(i);
+        }
+    }
+
+    if let Some(starred_target_index) = starred_target_index {
+        assert!(tuple_tys.len() >= targets.len() - 1); // The typechecker ensures this
+
+        let a = starred_target_index; // Number of RHS values before the starred target
+        let b = tuple_tys.len() - (targets.len() - 1 - starred_target_index); // Number of RHS values after the starred target
+                                                                              // Thus `tuple[a..b]` is assigned to the starred target.
+
+        // Handle assignment before the starred target
+        for (target, val, val_ty) in
+            izip!(&targets[..starred_target_index], &tuple[..a], &tuple_tys[..a])
+        {
+            generator.gen_assign(ctx, target, ValueEnum::Dynamic(*val), *val_ty)?;
+        }
+
+        // Handle assignment to the starred target
+        if let ExprKind::Starred { value: target, .. } = &targets[starred_target_index].node {
+            let vals = &tuple[a..b];
+            let val_tys = &tuple_tys[a..b];
+
+            // Create a sub-tuple from `value` for the starred target.
+            let sub_tuple_ty = ctx
+                .ctx
+                .struct_type(&vals.iter().map(BasicValueEnum::get_type).collect_vec(), false);
+            let psub_tuple_val =
+                ctx.builder.build_alloca(sub_tuple_ty, "starred_target_value_ptr").unwrap();
+            for (i, val) in vals.iter().enumerate() {
+                let pitem = ctx
+                    .builder
+                    .build_struct_gep(psub_tuple_val, i as u32, "starred_target_value_item")
+                    .unwrap();
+                ctx.builder.build_store(pitem, *val).unwrap();
+            }
+            let sub_tuple_val =
+                ctx.builder.build_load(psub_tuple_val, "starred_target_value").unwrap();
+
+            // Create the typechecker type of the sub-tuple
+            let sub_tuple_ty =
+                ctx.unifier.add_ty(TypeEnum::TTuple { ty: val_tys.to_vec(), is_vararg_ctx: false });
+
+            // Now assign with that sub-tuple to the starred target.
+            generator.gen_assign(ctx, target, ValueEnum::Dynamic(sub_tuple_val), sub_tuple_ty)?;
+        } else {
+            unreachable!() // The typechecker ensures this
+        }
+
+        // Handle assignment after the starred target
+        for (target, val, val_ty) in
+            izip!(&targets[starred_target_index + 1..], &tuple[b..], &tuple_tys[b..])
+        {
+            generator.gen_assign(ctx, target, ValueEnum::Dynamic(*val), *val_ty)?;
+        }
+    } else {
+        assert_eq!(tuple_tys.len(), targets.len()); // The typechecker ensures this
+
+        for (target, val, val_ty) in izip!(targets, tuple, tuple_tys) {
+            generator.gen_assign(ctx, target, ValueEnum::Dynamic(val), *val_ty)?;
+        }
+    }
+    Ok(())
+}
+
+/// See [`CodeGenerator::gen_setitem`].
+pub fn gen_setitem<'ctx, G: CodeGenerator>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    target: &Expr<Option<Type>>,
+    key: &Expr<Option<Type>>,
+    value: ValueEnum<'ctx>,
+    value_ty: Type,
+) -> Result<(), String> {
+    let target_ty = target.custom.unwrap();
+    let key_ty = key.custom.unwrap();
+
+    match &*ctx.unifier.get_ty(target_ty) {
+        TypeEnum::TObj { obj_id, params: list_params, .. }
+            if *obj_id == ctx.primitives.list.obj_id(&ctx.unifier).unwrap() =>
+        {
+            // Handle list item assignment
+            let llvm_usize = generator.get_size_type(ctx.ctx);
+            let target_item_ty = iter_type_vars(list_params).next().unwrap().ty;
+
+            let target = generator
+                .gen_expr(ctx, target)?
+                .unwrap()
+                .to_basic_value_enum(ctx, generator, target_ty)?
+                .into_pointer_value();
+            let target = ListValue::from_ptr_val(target, llvm_usize, None);
+
+            if let ExprKind::Slice { .. } = &key.node {
+                // Handle assigning to a slice
+                let ExprKind::Slice { lower, upper, step } = &key.node else { unreachable!() };
+                let Some((start, end, step)) = handle_slice_indices(
+                    lower,
+                    upper,
+                    step,
+                    ctx,
+                    generator,
+                    target.load_size(ctx, None),
+                )?
+                else {
+                    return Ok(());
+                };
+
+                let value =
+                    value.to_basic_value_enum(ctx, generator, value_ty)?.into_pointer_value();
+                let value = ListValue::from_ptr_val(value, llvm_usize, None);
+
+                let target_item_ty = ctx.get_llvm_type(generator, target_item_ty);
+                let Some(src_ind) = handle_slice_indices(
+                    &None,
+                    &None,
+                    &None,
+                    ctx,
+                    generator,
+                    value.load_size(ctx, None),
+                )?
+                else {
+                    return Ok(());
+                };
+                list_slice_assignment(
+                    generator,
+                    ctx,
+                    target_item_ty,
+                    target,
+                    (start, end, step),
+                    value,
+                    src_ind,
+                );
+            } else {
+                // Handle assigning to an index
+                let len = target.load_size(ctx, Some("len"));
+
+                let index = generator
+                    .gen_expr(ctx, key)?
+                    .unwrap()
+                    .to_basic_value_enum(ctx, generator, key_ty)?
+                    .into_int_value();
+                let index = ctx
+                    .builder
+                    .build_int_s_extend(index, generator.get_size_type(ctx.ctx), "sext")
+                    .unwrap();
+
+                // handle negative index
+                let is_negative = ctx
+                    .builder
+                    .build_int_compare(
+                        IntPredicate::SLT,
+                        index,
+                        generator.get_size_type(ctx.ctx).const_zero(),
+                        "is_neg",
+                    )
+                    .unwrap();
+                let adjusted = ctx.builder.build_int_add(index, len, "adjusted").unwrap();
+                let index = ctx
+                    .builder
+                    .build_select(is_negative, adjusted, index, "index")
+                    .map(BasicValueEnum::into_int_value)
+                    .unwrap();
+
+                // unsigned less than is enough, because negative index after adjustment is
+                // bigger than the length (for unsigned cmp)
+                let bound_check = ctx
+                    .builder
+                    .build_int_compare(IntPredicate::ULT, index, len, "inbound")
+                    .unwrap();
+                ctx.make_assert(
+                    generator,
+                    bound_check,
+                    "0:IndexError",
+                    "index {0} out of bounds 0:{1}",
+                    [Some(index), Some(len), None],
+                    key.location,
+                );
+
+                // Write value to index on list
+                let item_ptr =
+                    target.data().ptr_offset(ctx, generator, &index, Some("list_item_ptr"));
+                let value = value.to_basic_value_enum(ctx, generator, value_ty)?;
+                ctx.builder.build_store(item_ptr, value).unwrap();
+            }
+        }
+        TypeEnum::TObj { obj_id, .. }
+            if *obj_id == ctx.primitives.ndarray.obj_id(&ctx.unifier).unwrap() =>
+        {
+            // Handle NDArray item assignment
+            todo!("ndarray subscript assignment is not yet implemented");
+        }
+        _ => {
+            panic!("encountered unknown target type: {}", ctx.unifier.stringify(target_ty));
+        }
+    }
+    Ok(())
+}
+
 /// See [`CodeGenerator::gen_for`].
 pub fn gen_for<G: CodeGenerator>(
     generator: &mut G,
@@ -315,9 +432,6 @@ pub fn gen_for<G: CodeGenerator>(
     let orelse_bb =
         if orelse.is_empty() { cont_bb } else { ctx.ctx.append_basic_block(current, "for.orelse") };
 
-    // Whether the iterable is a range() expression
-    let is_iterable_range_expr = ctx.unifier.unioned(iter.custom.unwrap(), ctx.primitives.range);
-
     // The BB containing the increment expression
     let incr_bb = ctx.ctx.append_basic_block(current, "for.incr");
     // The BB containing the loop condition check
@@ -326,113 +440,132 @@ pub fn gen_for<G: CodeGenerator>(
     // store loop bb information and restore it later
     let loop_bb = ctx.loop_target.replace((incr_bb, cont_bb));
 
+    let iter_ty = iter.custom.unwrap();
     let iter_val = if let Some(v) = generator.gen_expr(ctx, iter)? {
-        v.to_basic_value_enum(ctx, generator, iter.custom.unwrap())?
+        v.to_basic_value_enum(ctx, generator, iter_ty)?
     } else {
         return Ok(());
     };
-    if is_iterable_range_expr {
-        let iter_val = RangeValue::from_ptr_val(iter_val.into_pointer_value(), Some("range"));
-        // Internal variable for loop; Cannot be assigned
-        let i = generator.gen_var_alloc(ctx, int32.into(), Some("for.i.addr"))?;
-        // Variable declared in "target" expression of the loop; Can be reassigned *or* shadowed
-        let Some(target_i) = generator.gen_store_target(ctx, target, Some("for.target.addr"))?
-        else {
-            unreachable!()
-        };
-        let (start, stop, step) = destructure_range(ctx, iter_val);
-
-        ctx.builder.build_store(i, start).unwrap();
-
-        // Check "If step is zero, ValueError is raised."
-        let rangenez =
-            ctx.builder.build_int_compare(IntPredicate::NE, step, int32.const_zero(), "").unwrap();
-        ctx.make_assert(
-            generator,
-            rangenez,
-            "ValueError",
-            "range() arg 3 must not be zero",
-            [None, None, None],
-            ctx.current_loc,
-        );
-        ctx.builder.build_unconditional_branch(cond_bb).unwrap();
 
+    match &*ctx.unifier.get_ty(iter_ty) {
+        TypeEnum::TObj { obj_id, .. }
+            if *obj_id == ctx.primitives.range.obj_id(&ctx.unifier).unwrap() =>
         {
-            ctx.builder.position_at_end(cond_bb);
-            ctx.builder
-                .build_conditional_branch(
-                    gen_in_range_check(
-                        ctx,
-                        ctx.builder.build_load(i, "").map(BasicValueEnum::into_int_value).unwrap(),
-                        stop,
-                        step,
-                    ),
-                    body_bb,
-                    orelse_bb,
+            let iter_val = RangeValue::from_ptr_val(iter_val.into_pointer_value(), Some("range"));
+            // Internal variable for loop; Cannot be assigned
+            let i = generator.gen_var_alloc(ctx, int32.into(), Some("for.i.addr"))?;
+            // Variable declared in "target" expression of the loop; Can be reassigned *or* shadowed
+            let Some(target_i) =
+                generator.gen_store_target(ctx, target, Some("for.target.addr"))?
+            else {
+                unreachable!()
+            };
+            let (start, stop, step) = destructure_range(ctx, iter_val);
+
+            ctx.builder.build_store(i, start).unwrap();
+
+            // Check "If step is zero, ValueError is raised."
+            let rangenez = ctx
+                .builder
+                .build_int_compare(IntPredicate::NE, step, int32.const_zero(), "")
+                .unwrap();
+            ctx.make_assert(
+                generator,
+                rangenez,
+                "ValueError",
+                "range() arg 3 must not be zero",
+                [None, None, None],
+                ctx.current_loc,
+            );
+            ctx.builder.build_unconditional_branch(cond_bb).unwrap();
+
+            {
+                ctx.builder.position_at_end(cond_bb);
+                ctx.builder
+                    .build_conditional_branch(
+                        gen_in_range_check(
+                            ctx,
+                            ctx.builder
+                                .build_load(i, "")
+                                .map(BasicValueEnum::into_int_value)
+                                .unwrap(),
+                            stop,
+                            step,
+                        ),
+                        body_bb,
+                        orelse_bb,
+                    )
+                    .unwrap();
+            }
+
+            ctx.builder.position_at_end(incr_bb);
+            let next_i = ctx
+                .builder
+                .build_int_add(
+                    ctx.builder.build_load(i, "").map(BasicValueEnum::into_int_value).unwrap(),
+                    step,
+                    "inc",
                 )
                 .unwrap();
+            ctx.builder.build_store(i, next_i).unwrap();
+            ctx.builder.build_unconditional_branch(cond_bb).unwrap();
+
+            ctx.builder.position_at_end(body_bb);
+            ctx.builder
+                .build_store(
+                    target_i,
+                    ctx.builder.build_load(i, "").map(BasicValueEnum::into_int_value).unwrap(),
+                )
+                .unwrap();
+            generator.gen_block(ctx, body.iter())?;
         }
+        TypeEnum::TObj { obj_id, params: list_params, .. }
+            if *obj_id == ctx.primitives.list.obj_id(&ctx.unifier).unwrap() =>
+        {
+            let index_addr = generator.gen_var_alloc(ctx, size_t.into(), Some("for.index.addr"))?;
+            ctx.builder.build_store(index_addr, size_t.const_zero()).unwrap();
+            let len = ctx
+                .build_gep_and_load(
+                    iter_val.into_pointer_value(),
+                    &[zero, int32.const_int(1, false)],
+                    Some("len"),
+                )
+                .into_int_value();
+            ctx.builder.build_unconditional_branch(cond_bb).unwrap();
 
-        ctx.builder.position_at_end(incr_bb);
-        let next_i = ctx
-            .builder
-            .build_int_add(
-                ctx.builder.build_load(i, "").map(BasicValueEnum::into_int_value).unwrap(),
-                step,
-                "inc",
-            )
-            .unwrap();
-        ctx.builder.build_store(i, next_i).unwrap();
-        ctx.builder.build_unconditional_branch(cond_bb).unwrap();
+            ctx.builder.position_at_end(cond_bb);
+            let index = ctx
+                .builder
+                .build_load(index_addr, "for.index")
+                .map(BasicValueEnum::into_int_value)
+                .unwrap();
+            let cmp = ctx.builder.build_int_compare(IntPredicate::SLT, index, len, "cond").unwrap();
+            ctx.builder.build_conditional_branch(cmp, body_bb, orelse_bb).unwrap();
 
-        ctx.builder.position_at_end(body_bb);
-        ctx.builder
-            .build_store(
-                target_i,
-                ctx.builder.build_load(i, "").map(BasicValueEnum::into_int_value).unwrap(),
-            )
-            .unwrap();
-        generator.gen_block(ctx, body.iter())?;
-    } else {
-        let index_addr = generator.gen_var_alloc(ctx, size_t.into(), Some("for.index.addr"))?;
-        ctx.builder.build_store(index_addr, size_t.const_zero()).unwrap();
-        let len = ctx
-            .build_gep_and_load(
-                iter_val.into_pointer_value(),
-                &[zero, int32.const_int(1, false)],
-                Some("len"),
-            )
-            .into_int_value();
-        ctx.builder.build_unconditional_branch(cond_bb).unwrap();
+            ctx.builder.position_at_end(incr_bb);
+            let index =
+                ctx.builder.build_load(index_addr, "").map(BasicValueEnum::into_int_value).unwrap();
+            let inc = ctx.builder.build_int_add(index, size_t.const_int(1, true), "inc").unwrap();
+            ctx.builder.build_store(index_addr, inc).unwrap();
+            ctx.builder.build_unconditional_branch(cond_bb).unwrap();
 
-        ctx.builder.position_at_end(cond_bb);
-        let index = ctx
-            .builder
-            .build_load(index_addr, "for.index")
-            .map(BasicValueEnum::into_int_value)
-            .unwrap();
-        let cmp = ctx.builder.build_int_compare(IntPredicate::SLT, index, len, "cond").unwrap();
-        ctx.builder.build_conditional_branch(cmp, body_bb, orelse_bb).unwrap();
-
-        ctx.builder.position_at_end(incr_bb);
-        let index =
-            ctx.builder.build_load(index_addr, "").map(BasicValueEnum::into_int_value).unwrap();
-        let inc = ctx.builder.build_int_add(index, size_t.const_int(1, true), "inc").unwrap();
-        ctx.builder.build_store(index_addr, inc).unwrap();
-        ctx.builder.build_unconditional_branch(cond_bb).unwrap();
-
-        ctx.builder.position_at_end(body_bb);
-        let arr_ptr = ctx
-            .build_gep_and_load(iter_val.into_pointer_value(), &[zero, zero], Some("arr.addr"))
-            .into_pointer_value();
-        let index = ctx
-            .builder
-            .build_load(index_addr, "for.index")
-            .map(BasicValueEnum::into_int_value)
-            .unwrap();
-        let val = ctx.build_gep_and_load(arr_ptr, &[index], Some("val"));
-        generator.gen_assign(ctx, target, val.into())?;
-        generator.gen_block(ctx, body.iter())?;
+            ctx.builder.position_at_end(body_bb);
+            let arr_ptr = ctx
+                .build_gep_and_load(iter_val.into_pointer_value(), &[zero, zero], Some("arr.addr"))
+                .into_pointer_value();
+            let index = ctx
+                .builder
+                .build_load(index_addr, "for.index")
+                .map(BasicValueEnum::into_int_value)
+                .unwrap();
+            let val = ctx.build_gep_and_load(arr_ptr, &[index], Some("val"));
+            let val_ty = iter_type_vars(list_params).next().unwrap().ty;
+            generator.gen_assign(ctx, target, val.into(), val_ty)?;
+            generator.gen_block(ctx, body.iter())?;
+        }
+        _ => {
+            panic!("unsupported for loop iterator type: {}", ctx.unifier.stringify(iter_ty));
+        }
     }
 
     for (k, (_, _, counter)) in &var_assignment {
@@ -1588,14 +1721,14 @@ pub fn gen_stmt<G: CodeGenerator>(
         }
         StmtKind::AnnAssign { target, value, .. } => {
             if let Some(value) = value {
-                let Some(value) = generator.gen_expr(ctx, value)? else { return Ok(()) };
-                generator.gen_assign(ctx, target, value)?;
+                let Some(value_enum) = generator.gen_expr(ctx, value)? else { return Ok(()) };
+                generator.gen_assign(ctx, target, value_enum, value.custom.unwrap())?;
             }
         }
         StmtKind::Assign { targets, value, .. } => {
-            let Some(value) = generator.gen_expr(ctx, value)? else { return Ok(()) };
+            let Some(value_enum) = generator.gen_expr(ctx, value)? else { return Ok(()) };
             for target in targets {
-                generator.gen_assign(ctx, target, value.clone())?;
+                generator.gen_assign(ctx, target, value_enum.clone(), value.custom.unwrap())?;
             }
         }
         StmtKind::Continue { .. } => {
@@ -1609,15 +1742,16 @@ pub fn gen_stmt<G: CodeGenerator>(
         StmtKind::For { .. } => generator.gen_for(ctx, stmt)?,
         StmtKind::With { .. } => generator.gen_with(ctx, stmt)?,
         StmtKind::AugAssign { target, op, value, .. } => {
-            let value = gen_binop_expr(
+            let value_enum = gen_binop_expr(
                 generator,
                 ctx,
                 target,
                 Binop::aug_assign(*op),
                 value,
                 stmt.location,
-            )?;
-            generator.gen_assign(ctx, target, value.unwrap())?;
+            )?
+            .unwrap();
+            generator.gen_assign(ctx, target, value_enum, value.custom.unwrap())?;
         }
         StmtKind::Try { .. } => gen_try(generator, ctx, stmt)?,
         StmtKind::Raise { exc, .. } => {

nac3core/src/codegen/test.rs

@@ -109,8 +109,18 @@ fn test_primitives() {
     let threads = vec![DefaultCodeGenerator::new("test".into(), 32).into()];
     let signature = FunSignature {
         args: vec![
-            FuncArg { name: "a".into(), ty: primitives.int32, default_value: None },
-            FuncArg { name: "b".into(), ty: primitives.int32, default_value: None },
+            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,
+            },
         ],
         ret: primitives.int32,
         vars: VarMap::new(),
@@ -255,7 +265,12 @@ fn test_simple_call() {
     unifier.top_level = Some(top_level.clone());
 
     let signature = FunSignature {
-        args: vec![FuncArg { name: "a".into(), ty: primitives.int32, default_value: None }],
+        args: vec![FuncArg {
+            name: "a".into(),
+            ty: primitives.int32,
+            default_value: None,
+            is_vararg: false,
+        }],
         ret: primitives.int32,
         vars: VarMap::new(),
     };

nac3core/src/symbol_resolver.rs

@@ -78,14 +78,14 @@ impl SymbolValue {
             }
             Constant::Tuple(t) => {
                 let expected_ty = unifier.get_ty(expected_ty);
-                let TypeEnum::TTuple { ty } = expected_ty.as_ref() else {
+                let TypeEnum::TTuple { ty, is_vararg_ctx } = expected_ty.as_ref() else {
                     return Err(format!(
                         "Expected {:?}, but got Tuple",
                         expected_ty.get_type_name()
                     ));
                 };
 
-                assert_eq!(ty.len(), t.len());
+                assert!(*is_vararg_ctx || ty.len() == t.len());
 
                 let elems = t
                     .iter()
@@ -155,7 +155,7 @@ impl SymbolValue {
             SymbolValue::Bool(_) => primitives.bool,
             SymbolValue::Tuple(vs) => {
                 let vs_tys = vs.iter().map(|v| v.get_type(primitives, unifier)).collect::<Vec<_>>();
-                unifier.add_ty(TypeEnum::TTuple { ty: vs_tys })
+                unifier.add_ty(TypeEnum::TTuple { ty: vs_tys, is_vararg_ctx: false })
             }
             SymbolValue::OptionSome(_) | SymbolValue::OptionNone => primitives.option,
         }
@@ -482,7 +482,7 @@ pub fn parse_type_annotation<T>(
                         parse_type_annotation(resolver, top_level_defs, unifier, primitives, elt)
                     })
                     .collect::<Result<Vec<_>, _>>()?;
-                Ok(unifier.add_ty(TypeEnum::TTuple { ty }))
+                Ok(unifier.add_ty(TypeEnum::TTuple { ty, is_vararg_ctx: false }))
             } else {
                 Err(HashSet::from(["Expected multiple elements for tuple".into()]))
             }

nac3core/src/toplevel/builtins.rs

@@ -45,10 +45,26 @@ pub fn get_exn_constructor(
             name: "msg".into(),
             ty: string,
             default_value: Some(SymbolValue::Str(String::new())),
+            is_vararg: false,
+        },
+        FuncArg {
+            name: "param0".into(),
+            ty: int64,
+            default_value: Some(SymbolValue::I64(0)),
+            is_vararg: false,
+        },
+        FuncArg {
+            name: "param1".into(),
+            ty: int64,
+            default_value: Some(SymbolValue::I64(0)),
+            is_vararg: false,
+        },
+        FuncArg {
+            name: "param2".into(),
+            ty: int64,
+            default_value: Some(SymbolValue::I64(0)),
+            is_vararg: false,
         },
-        FuncArg { name: "param0".into(), ty: int64, default_value: Some(SymbolValue::I64(0)) },
-        FuncArg { name: "param1".into(), ty: int64, default_value: Some(SymbolValue::I64(0)) },
-        FuncArg { name: "param2".into(), ty: int64, default_value: Some(SymbolValue::I64(0)) },
     ];
     let exn_type = unifier.add_ty(TypeEnum::TObj {
         obj_id: DefinitionId(class_id),
@@ -114,7 +130,12 @@ fn create_fn_by_codegen(
         signature: unifier.add_ty(TypeEnum::TFunc(FunSignature {
             args: param_ty
                 .iter()
-                .map(|p| FuncArg { name: p.1.into(), ty: p.0, default_value: None })
+                .map(|p| FuncArg {
+                    name: p.1.into(),
+                    ty: p.0,
+                    default_value: None,
+                    is_vararg: false,
+                })
                 .collect(),
             ret: ret_ty,
             vars: var_map.clone(),
@@ -556,6 +577,22 @@ impl<'a> BuiltinBuilder<'a> {
             | PrimDef::FunNpLdExp
             | PrimDef::FunNpHypot
             | PrimDef::FunNpNextAfter => self.build_np_2ary_function(prim),
+
+            PrimDef::FunNpTranspose | PrimDef::FunNpReshape => {
+                self.build_np_sp_ndarray_function(prim)
+            }
+
+            PrimDef::FunNpDot
+            | PrimDef::FunNpLinalgCholesky
+            | PrimDef::FunNpLinalgQr
+            | PrimDef::FunNpLinalgSvd
+            | PrimDef::FunNpLinalgInv
+            | PrimDef::FunNpLinalgPinv
+            | PrimDef::FunNpLinalgMatrixPower
+            | PrimDef::FunNpLinalgDet
+            | PrimDef::FunSpLinalgLu
+            | PrimDef::FunSpLinalgSchur
+            | PrimDef::FunSpLinalgHessenberg => self.build_linalg_methods(prim),
         };
 
         if cfg!(debug_assertions) {
@@ -613,17 +650,24 @@ impl<'a> BuiltinBuilder<'a> {
         let make_ctor_signature = |unifier: &mut Unifier| {
             unifier.add_ty(TypeEnum::TFunc(FunSignature {
                 args: vec![
-                    FuncArg { name: "start".into(), ty: int32, default_value: None },
+                    FuncArg {
+                        name: "start".into(),
+                        ty: int32,
+                        default_value: None,
+                        is_vararg: false,
+                    },
                     FuncArg {
                         name: "stop".into(),
                         ty: int32,
                         // placeholder
                         default_value: Some(SymbolValue::I32(0)),
+                        is_vararg: false,
                     },
                     FuncArg {
                         name: "step".into(),
                         ty: int32,
                         default_value: Some(SymbolValue::I32(1)),
+                        is_vararg: false,
                     },
                 ],
                 ret: range,
@@ -879,6 +923,7 @@ impl<'a> BuiltinBuilder<'a> {
                         name: "n".into(),
                         ty: self.option_tvar.ty,
                         default_value: None,
+                        is_vararg: false,
                     }],
                     ret: self.primitives.option,
                     vars: into_var_map([self.option_tvar]),
@@ -1013,6 +1058,7 @@ impl<'a> BuiltinBuilder<'a> {
                     name: "n".into(),
                     ty: self.num_or_ndarray_ty.ty,
                     default_value: None,
+                    is_vararg: false,
                 }],
                 ret: self.num_or_ndarray_ty.ty,
                 vars: self.num_or_ndarray_var_map.clone(),
@@ -1232,16 +1278,23 @@ impl<'a> BuiltinBuilder<'a> {
                     simple_name: prim.simple_name().into(),
                     signature: self.unifier.add_ty(TypeEnum::TFunc(FunSignature {
                         args: vec![
-                            FuncArg { name: "object".into(), ty: tv.ty, default_value: None },
+                            FuncArg {
+                                name: "object".into(),
+                                ty: tv.ty,
+                                default_value: None,
+                                is_vararg: false,
+                            },
                             FuncArg {
                                 name: "copy".into(),
                                 ty: bool,
                                 default_value: Some(SymbolValue::Bool(true)),
+                                is_vararg: false,
                             },
                             FuncArg {
                                 name: "ndmin".into(),
                                 ty: int32,
                                 default_value: Some(SymbolValue::U32(0)),
+                                is_vararg: false,
                             },
                         ],
                         ret: ndarray,
@@ -1283,17 +1336,24 @@ impl<'a> BuiltinBuilder<'a> {
                     simple_name: prim.simple_name().into(),
                     signature: self.unifier.add_ty(TypeEnum::TFunc(FunSignature {
                         args: vec![
-                            FuncArg { name: "N".into(), ty: int32, default_value: None },
+                            FuncArg {
+                                name: "N".into(),
+                                ty: int32,
+                                default_value: None,
+                                is_vararg: false,
+                            },
                             // TODO(Derppening): Default values current do not work?
                             FuncArg {
                                 name: "M".into(),
                                 ty: int32,
                                 default_value: Some(SymbolValue::OptionNone),
+                                is_vararg: false,
                             },
                             FuncArg {
                                 name: "k".into(),
                                 ty: int32,
                                 default_value: Some(SymbolValue::I32(0)),
+                                is_vararg: false,
                             },
                         ],
                         ret: self.ndarray_float_2d,
@@ -1337,7 +1397,12 @@ impl<'a> BuiltinBuilder<'a> {
             name: prim.name().into(),
             simple_name: prim.simple_name().into(),
             signature: self.unifier.add_ty(TypeEnum::TFunc(FunSignature {
-                args: vec![FuncArg { name: "s".into(), ty: str, default_value: None }],
+                args: vec![FuncArg {
+                    name: "s".into(),
+                    ty: str,
+                    default_value: None,
+                    is_vararg: false,
+                }],
                 ret: str,
                 vars: VarMap::default(),
             })),
@@ -1423,7 +1488,12 @@ impl<'a> BuiltinBuilder<'a> {
             name: prim.name().into(),
             simple_name: prim.simple_name().into(),
             signature: self.unifier.add_ty(TypeEnum::TFunc(FunSignature {
-                args: vec![FuncArg { name: "ls".into(), ty: arg_ty.ty, default_value: None }],
+                args: vec![FuncArg {
+                    name: "ls".into(),
+                    ty: arg_ty.ty,
+                    default_value: None,
+                    is_vararg: false,
+                }],
                 ret: int32,
                 vars: into_var_map([tvar, arg_ty]),
             })),
@@ -1528,8 +1598,18 @@ impl<'a> BuiltinBuilder<'a> {
             simple_name: prim.simple_name().into(),
             signature: self.unifier.add_ty(TypeEnum::TFunc(FunSignature {
                 args: vec![
-                    FuncArg { name: "m".into(), ty: self.num_ty.ty, default_value: None },
-                    FuncArg { name: "n".into(), ty: self.num_ty.ty, default_value: None },
+                    FuncArg {
+                        name: "m".into(),
+                        ty: self.num_ty.ty,
+                        default_value: None,
+                        is_vararg: false,
+                    },
+                    FuncArg {
+                        name: "n".into(),
+                        ty: self.num_ty.ty,
+                        default_value: None,
+                        is_vararg: false,
+                    },
                 ],
                 ret: self.num_ty.ty,
                 vars: self.num_var_map.clone(),
@@ -1611,7 +1691,12 @@ impl<'a> BuiltinBuilder<'a> {
             signature: self.unifier.add_ty(TypeEnum::TFunc(FunSignature {
                 args: param_ty
                     .iter()
-                    .map(|p| FuncArg { name: p.1.into(), ty: p.0, default_value: None })
+                    .map(|p| FuncArg {
+                        name: p.1.into(),
+                        ty: p.0,
+                        default_value: None,
+                        is_vararg: false,
+                    })
                     .collect(),
                 ret: ret_ty.ty,
                 vars: into_var_map([x1_ty, x2_ty, ret_ty]),
@@ -1652,6 +1737,7 @@ impl<'a> BuiltinBuilder<'a> {
                     name: "n".into(),
                     ty: self.num_or_ndarray_ty.ty,
                     default_value: None,
+                    is_vararg: false,
                 }],
                 ret: self.num_or_ndarray_ty.ty,
                 vars: self.num_or_ndarray_var_map.clone(),
@@ -1840,7 +1926,12 @@ impl<'a> BuiltinBuilder<'a> {
             signature: self.unifier.add_ty(TypeEnum::TFunc(FunSignature {
                 args: param_ty
                     .iter()
-                    .map(|p| FuncArg { name: p.1.into(), ty: p.0, default_value: None })
+                    .map(|p| FuncArg {
+                        name: p.1.into(),
+                        ty: p.0,
+                        default_value: None,
+                        is_vararg: false,
+                    })
                     .collect(),
                 ret: ret_ty.ty,
                 vars: into_var_map([x1_ty, x2_ty, ret_ty]),
@@ -1874,6 +1965,207 @@ impl<'a> BuiltinBuilder<'a> {
         }
     }
 
+    /// Build np/sp functions that take as input `NDArray` only
+    fn build_np_sp_ndarray_function(&mut self, prim: PrimDef) -> TopLevelDef {
+        debug_assert_prim_is_allowed(prim, &[PrimDef::FunNpTranspose, PrimDef::FunNpReshape]);
+
+        match prim {
+            PrimDef::FunNpTranspose => {
+                let ndarray_ty = self.unifier.get_fresh_var_with_range(
+                    &[self.ndarray_num_ty],
+                    Some("T".into()),
+                    None,
+                );
+                create_fn_by_codegen(
+                    self.unifier,
+                    &into_var_map([ndarray_ty]),
+                    prim.name(),
+                    ndarray_ty.ty,
+                    &[(ndarray_ty.ty, "x")],
+                    Box::new(move |ctx, _, fun, args, generator| {
+                        let arg_ty = fun.0.args[0].ty;
+                        let arg_val =
+                            args[0].1.clone().to_basic_value_enum(ctx, generator, arg_ty)?;
+                        Ok(Some(ndarray_transpose(generator, ctx, (arg_ty, arg_val))?))
+                    }),
+                )
+            }
+
+            // NOTE: on `ndarray_factory_fn_shape_arg_tvar` and
+            // the `param_ty` for `create_fn_by_codegen`.
+            //
+            // Similar to `build_ndarray_from_shape_factory_function` we delegate the responsibility of typechecking
+            // to [`typecheck::type_inferencer::Inferencer::fold_numpy_function_call_shape_argument`],
+            // and use a dummy [`TypeVar`] `ndarray_factory_fn_shape_arg_tvar` as a placeholder for `param_ty`.
+            PrimDef::FunNpReshape => create_fn_by_codegen(
+                self.unifier,
+                &VarMap::new(),
+                prim.name(),
+                self.ndarray_num_ty,
+                &[(self.ndarray_num_ty, "x"), (self.ndarray_factory_fn_shape_arg_tvar.ty, "shape")],
+                Box::new(move |ctx, _, fun, args, generator| {
+                    let x1_ty = fun.0.args[0].ty;
+                    let x1_val = args[0].1.clone().to_basic_value_enum(ctx, generator, x1_ty)?;
+                    let x2_ty = fun.0.args[1].ty;
+                    let x2_val = args[1].1.clone().to_basic_value_enum(ctx, generator, x2_ty)?;
+                    Ok(Some(ndarray_reshape(generator, ctx, (x1_ty, x1_val), (x2_ty, x2_val))?))
+                }),
+            ),
+
+            _ => unreachable!(),
+        }
+    }
+
+    /// Build `np_linalg` and `sp_linalg` functions
+    ///
+    /// The input to these functions must be floating point `NDArray`
+    fn build_linalg_methods(&mut self, prim: PrimDef) -> TopLevelDef {
+        debug_assert_prim_is_allowed(
+            prim,
+            &[
+                PrimDef::FunNpDot,
+                PrimDef::FunNpLinalgCholesky,
+                PrimDef::FunNpLinalgQr,
+                PrimDef::FunNpLinalgSvd,
+                PrimDef::FunNpLinalgInv,
+                PrimDef::FunNpLinalgPinv,
+                PrimDef::FunNpLinalgMatrixPower,
+                PrimDef::FunNpLinalgDet,
+                PrimDef::FunSpLinalgLu,
+                PrimDef::FunSpLinalgSchur,
+                PrimDef::FunSpLinalgHessenberg,
+            ],
+        );
+
+        match prim {
+            PrimDef::FunNpDot => create_fn_by_codegen(
+                self.unifier,
+                &self.num_or_ndarray_var_map,
+                prim.name(),
+                self.num_ty.ty,
+                &[(self.num_or_ndarray_ty.ty, "x1"), (self.num_or_ndarray_ty.ty, "x2")],
+                Box::new(move |ctx, _, fun, args, generator| {
+                    let x1_ty = fun.0.args[0].ty;
+                    let x1_val = args[0].1.clone().to_basic_value_enum(ctx, generator, x1_ty)?;
+                    let x2_ty = fun.0.args[1].ty;
+                    let x2_val = args[1].1.clone().to_basic_value_enum(ctx, generator, x2_ty)?;
+
+                    Ok(Some(ndarray_dot(generator, ctx, (x1_ty, x1_val), (x2_ty, x2_val))?))
+                }),
+            ),
+
+            PrimDef::FunNpLinalgCholesky | PrimDef::FunNpLinalgInv | PrimDef::FunNpLinalgPinv => {
+                create_fn_by_codegen(
+                    self.unifier,
+                    &VarMap::new(),
+                    prim.name(),
+                    self.ndarray_float_2d,
+                    &[(self.ndarray_float_2d, "x1")],
+                    Box::new(move |ctx, _, fun, args, generator| {
+                        let x1_ty = fun.0.args[0].ty;
+                        let x1_val =
+                            args[0].1.clone().to_basic_value_enum(ctx, generator, x1_ty)?;
+
+                        let func = match prim {
+                            PrimDef::FunNpLinalgCholesky => builtin_fns::call_np_linalg_cholesky,
+                            PrimDef::FunNpLinalgInv => builtin_fns::call_np_linalg_inv,
+                            PrimDef::FunNpLinalgPinv => builtin_fns::call_np_linalg_pinv,
+                            _ => unreachable!(),
+                        };
+                        Ok(Some(func(generator, ctx, (x1_ty, x1_val))?))
+                    }),
+                )
+            }
+
+            PrimDef::FunNpLinalgQr
+            | PrimDef::FunSpLinalgLu
+            | PrimDef::FunSpLinalgSchur
+            | PrimDef::FunSpLinalgHessenberg => {
+                let ret_ty = self.unifier.add_ty(TypeEnum::TTuple {
+                    ty: vec![self.ndarray_float_2d, self.ndarray_float_2d],
+                    is_vararg_ctx: false,
+                });
+                create_fn_by_codegen(
+                    self.unifier,
+                    &VarMap::new(),
+                    prim.name(),
+                    ret_ty,
+                    &[(self.ndarray_float_2d, "x1")],
+                    Box::new(move |ctx, _, fun, args, generator| {
+                        let x1_ty = fun.0.args[0].ty;
+                        let x1_val =
+                            args[0].1.clone().to_basic_value_enum(ctx, generator, x1_ty)?;
+
+                        let func = match prim {
+                            PrimDef::FunNpLinalgQr => builtin_fns::call_np_linalg_qr,
+                            PrimDef::FunSpLinalgLu => builtin_fns::call_sp_linalg_lu,
+                            PrimDef::FunSpLinalgSchur => builtin_fns::call_sp_linalg_schur,
+                            PrimDef::FunSpLinalgHessenberg => {
+                                builtin_fns::call_sp_linalg_hessenberg
+                            }
+                            _ => unreachable!(),
+                        };
+                        Ok(Some(func(generator, ctx, (x1_ty, x1_val))?))
+                    }),
+                )
+            }
+
+            PrimDef::FunNpLinalgSvd => {
+                let ret_ty = self.unifier.add_ty(TypeEnum::TTuple {
+                    ty: vec![self.ndarray_float_2d, self.ndarray_float, self.ndarray_float_2d],
+                    is_vararg_ctx: false,
+                });
+                create_fn_by_codegen(
+                    self.unifier,
+                    &VarMap::new(),
+                    prim.name(),
+                    ret_ty,
+                    &[(self.ndarray_float_2d, "x1")],
+                    Box::new(move |ctx, _, fun, args, generator| {
+                        let x1_ty = fun.0.args[0].ty;
+                        let x1_val =
+                            args[0].1.clone().to_basic_value_enum(ctx, generator, x1_ty)?;
+
+                        Ok(Some(builtin_fns::call_np_linalg_svd(generator, ctx, (x1_ty, x1_val))?))
+                    }),
+                )
+            }
+            PrimDef::FunNpLinalgMatrixPower => create_fn_by_codegen(
+                self.unifier,
+                &VarMap::new(),
+                prim.name(),
+                self.ndarray_float_2d,
+                &[(self.ndarray_float_2d, "x1"), (self.primitives.int32, "power")],
+                Box::new(move |ctx, _, fun, args, generator| {
+                    let x1_ty = fun.0.args[0].ty;
+                    let x1_val = args[0].1.clone().to_basic_value_enum(ctx, generator, x1_ty)?;
+                    let x2_ty = fun.0.args[1].ty;
+                    let x2_val = args[1].1.clone().to_basic_value_enum(ctx, generator, x2_ty)?;
+
+                    Ok(Some(builtin_fns::call_np_linalg_matrix_power(
+                        generator,
+                        ctx,
+                        (x1_ty, x1_val),
+                        (x2_ty, x2_val),
+                    )?))
+                }),
+            ),
+            PrimDef::FunNpLinalgDet => create_fn_by_codegen(
+                self.unifier,
+                &VarMap::new(),
+                prim.name(),
+                self.primitives.float,
+                &[(self.ndarray_float_2d, "x1")],
+                Box::new(move |ctx, _, fun, args, generator| {
+                    let x1_ty = fun.0.args[0].ty;
+                    let x1_val = args[0].1.clone().to_basic_value_enum(ctx, generator, x1_ty)?;
+                    Ok(Some(builtin_fns::call_np_linalg_det(generator, ctx, (x1_ty, x1_val))?))
+                }),
+            ),
+            _ => unreachable!(),
+        }
+    }
+
     fn create_method(prim: PrimDef, method_ty: Type) -> (StrRef, Type, DefinitionId) {
         (prim.simple_name().into(), method_ty, prim.id())
     }

nac3core/src/toplevel/composer.rs

@@ -860,7 +860,73 @@ impl TopLevelComposer {
             let resolver = &**resolver;
 
             let mut function_var_map = VarMap::new();
-            let arg_types = {
+
+            let vararg = args
+                .vararg
+                .as_ref()
+                .map(|vararg| -> Result<_, HashSet<String>> {
+                    let vararg = vararg.as_ref();
+
+                    let annotation = vararg
+                        .node
+                        .annotation
+                        .as_ref()
+                        .ok_or_else(|| {
+                            HashSet::from([format!(
+                                "function parameter `{}` needs type annotation at {}",
+                                vararg.node.arg, vararg.location
+                            )])
+                        })?
+                        .as_ref();
+
+                    let type_annotation = parse_ast_to_type_annotation_kinds(
+                        resolver,
+                        temp_def_list.as_slice(),
+                        unifier,
+                        primitives_store,
+                        annotation,
+                        // NOTE: since only class need this, for function
+                        // it should be fine to be empty map
+                        HashMap::new(),
+                    )?;
+
+                    let type_vars_within =
+                        get_type_var_contained_in_type_annotation(&type_annotation)
+                            .into_iter()
+                            .map(|x| -> Result<TypeVar, HashSet<String>> {
+                                let TypeAnnotation::TypeVar(ty) = x else {
+                                    unreachable!("must be type var annotation kind")
+                                };
+
+                                let id = Self::get_var_id(ty, unifier)?;
+                                Ok(TypeVar { id, ty })
+                            })
+                            .collect::<Result<Vec<_>, _>>()?;
+                    for var in type_vars_within {
+                        if let Some(prev_ty) = function_var_map.insert(var.id, var.ty) {
+                            // if already have the type inserted, make sure they are the same thing
+                            assert_eq!(prev_ty, var.ty);
+                        }
+                    }
+
+                    let ty = get_type_from_type_annotation_kinds(
+                        temp_def_list.as_ref(),
+                        unifier,
+                        primitives_store,
+                        &type_annotation,
+                        &mut None,
+                    )?;
+
+                    Ok(FuncArg {
+                        name: vararg.node.arg,
+                        ty,
+                        default_value: Some(SymbolValue::Tuple(Vec::default())),
+                        is_vararg: true,
+                    })
+                })
+                .transpose()?;
+
+            let mut arg_types = {
                 // make sure no duplicate parameter
                 let mut defined_parameter_name: HashSet<_> = HashSet::new();
                 for x in &args.args {
@@ -961,11 +1027,18 @@ impl TopLevelComposer {
                                     v
                                 }),
                             },
+                            is_vararg: false,
                         })
                     })
                     .collect::<Result<Vec<_>, _>>()?
             };
 
+            if let Some(vararg) = vararg {
+                arg_types.push(vararg);
+            };
+
+            let arg_types = arg_types;
+
             let return_ty = {
                 if let Some(returns) = returns {
                     let return_ty_annotation = {
@@ -1217,6 +1290,7 @@ impl TopLevelComposer {
                                             })
                                         }
                                     },
+                                    is_vararg: false,
                                 };
                                 // push the dummy type and the type annotation
                                 // into the list for later unification
@@ -1642,21 +1716,25 @@ impl TopLevelComposer {
                                 name: "msg".into(),
                                 ty: string,
                                 default_value: Some(SymbolValue::Str(String::new())),
+                                is_vararg: false,
                             },
                             FuncArg {
                                 name: "param0".into(),
                                 ty: int64,
                                 default_value: Some(SymbolValue::I64(0)),
+                                is_vararg: false,
                             },
                             FuncArg {
                                 name: "param1".into(),
                                 ty: int64,
                                 default_value: Some(SymbolValue::I64(0)),
+                                is_vararg: false,
                             },
                             FuncArg {
                                 name: "param2".into(),
                                 ty: int64,
                                 default_value: Some(SymbolValue::I64(0)),
+                                is_vararg: false,
                             },
                         ],
                         ret: self_type,
@@ -1866,6 +1944,7 @@ impl TopLevelComposer {
                                 name: a.name,
                                 ty: unifier.subst(a.ty, &subst).unwrap_or(a.ty),
                                 default_value: a.default_value.clone(),
+                                is_vararg: false,
                             })
                             .collect_vec()
                     };

nac3core/src/toplevel/helper.rs

@@ -99,6 +99,21 @@ pub enum PrimDef {
     FunNpLdExp,
     FunNpHypot,
     FunNpNextAfter,
+    FunNpTranspose,
+    FunNpReshape,
+
+    // Linalg functions
+    FunNpDot,
+    FunNpLinalgCholesky,
+    FunNpLinalgQr,
+    FunNpLinalgSvd,
+    FunNpLinalgInv,
+    FunNpLinalgPinv,
+    FunNpLinalgMatrixPower,
+    FunNpLinalgDet,
+    FunSpLinalgLu,
+    FunSpLinalgSchur,
+    FunSpLinalgHessenberg,
 
     // Miscellaneous Python & NAC3 functions
     FunInt32,
@@ -270,6 +285,21 @@ impl PrimDef {
             PrimDef::FunNpLdExp => fun("np_ldexp", None),
             PrimDef::FunNpHypot => fun("np_hypot", None),
             PrimDef::FunNpNextAfter => fun("np_nextafter", None),
+            PrimDef::FunNpTranspose => fun("np_transpose", None),
+            PrimDef::FunNpReshape => fun("np_reshape", None),
+
+            // Linalg functions
+            PrimDef::FunNpDot => fun("np_dot", None),
+            PrimDef::FunNpLinalgCholesky => fun("np_linalg_cholesky", None),
+            PrimDef::FunNpLinalgQr => fun("np_linalg_qr", None),
+            PrimDef::FunNpLinalgSvd => fun("np_linalg_svd", None),
+            PrimDef::FunNpLinalgInv => fun("np_linalg_inv", None),
+            PrimDef::FunNpLinalgPinv => fun("np_linalg_pinv", None),
+            PrimDef::FunNpLinalgMatrixPower => fun("np_linalg_matrix_power", None),
+            PrimDef::FunNpLinalgDet => fun("np_linalg_det", None),
+            PrimDef::FunSpLinalgLu => fun("sp_linalg_lu", None),
+            PrimDef::FunSpLinalgSchur => fun("sp_linalg_schur", None),
+            PrimDef::FunSpLinalgHessenberg => fun("sp_linalg_hessenberg", None),
 
             // Miscellaneous Python & NAC3 functions
             PrimDef::FunInt32 => fun("int32", None),
@@ -475,6 +505,7 @@ impl TopLevelComposer {
                 name: "value".into(),
                 ty: ndarray_dtype_tvar.ty,
                 default_value: None,
+                is_vararg: false,
             }],
             ret: none,
             vars: into_var_map([ndarray_dtype_tvar, ndarray_ndims_tvar]),

nac3core/src/toplevel/snapshots/nac3core__toplevel__test__test_analyze__generic_class.snap

@@ -5,7 +5,7 @@ expression: res_vec
 [
     "Class {\nname: \"Generic_A\",\nancestors: [\"Generic_A[V]\", \"B\"],\nfields: [\"aa\", \"a\"],\nmethods: [(\"__init__\", \"fn[[], none]\"), (\"foo\", \"fn[[b:T], none]\"), (\"fun\", \"fn[[a:int32], V]\")],\ntype_vars: [\"V\"]\n}\n",
     "Function {\nname: \"Generic_A.__init__\",\nsig: \"fn[[], none]\",\nvar_id: []\n}\n",
-    "Function {\nname: \"Generic_A.fun\",\nsig: \"fn[[a:int32], V]\",\nvar_id: [TypeVarId(245)]\n}\n",
+    "Function {\nname: \"Generic_A.fun\",\nsig: \"fn[[a:int32], V]\",\nvar_id: [TypeVarId(246)]\n}\n",
     "Class {\nname: \"B\",\nancestors: [\"B\"],\nfields: [\"aa\"],\nmethods: [(\"__init__\", \"fn[[], none]\"), (\"foo\", \"fn[[b:T], none]\")],\ntype_vars: []\n}\n",
     "Function {\nname: \"B.__init__\",\nsig: \"fn[[], none]\",\nvar_id: []\n}\n",
     "Function {\nname: \"B.foo\",\nsig: \"fn[[b:T], none]\",\nvar_id: []\n}\n",

nac3core/src/toplevel/snapshots/nac3core__toplevel__test__test_analyze__inheritance_override.snap

@@ -7,7 +7,7 @@ expression: res_vec
     "Function {\nname: \"A.__init__\",\nsig: \"fn[[t:T], none]\",\nvar_id: []\n}\n",
     "Function {\nname: \"A.fun\",\nsig: \"fn[[a:int32, b:T], list[virtual[B[bool]]]]\",\nvar_id: []\n}\n",
     "Function {\nname: \"A.foo\",\nsig: \"fn[[c:C], none]\",\nvar_id: []\n}\n",
-    "Class {\nname: \"B\",\nancestors: [\"B[typevar234]\", \"A[float]\"],\nfields: [\"a\", \"b\", \"c\", \"d\"],\nmethods: [(\"__init__\", \"fn[[], none]\"), (\"fun\", \"fn[[a:int32, b:T], list[virtual[B[bool]]]]\"), (\"foo\", \"fn[[c:C], none]\")],\ntype_vars: [\"typevar234\"]\n}\n",
+    "Class {\nname: \"B\",\nancestors: [\"B[typevar235]\", \"A[float]\"],\nfields: [\"a\", \"b\", \"c\", \"d\"],\nmethods: [(\"__init__\", \"fn[[], none]\"), (\"fun\", \"fn[[a:int32, b:T], list[virtual[B[bool]]]]\"), (\"foo\", \"fn[[c:C], none]\")],\ntype_vars: [\"typevar235\"]\n}\n",
     "Function {\nname: \"B.__init__\",\nsig: \"fn[[], none]\",\nvar_id: []\n}\n",
     "Function {\nname: \"B.fun\",\nsig: \"fn[[a:int32, b:T], list[virtual[B[bool]]]]\",\nvar_id: []\n}\n",
     "Class {\nname: \"C\",\nancestors: [\"C\", \"B[bool]\", \"A[float]\"],\nfields: [\"a\", \"b\", \"c\", \"d\", \"e\"],\nmethods: [(\"__init__\", \"fn[[], none]\"), (\"fun\", \"fn[[a:int32, b:T], list[virtual[B[bool]]]]\"), (\"foo\", \"fn[[c:C], none]\")],\ntype_vars: []\n}\n",

nac3core/src/toplevel/snapshots/nac3core__toplevel__test__test_analyze__list_tuple_generic.snap

@@ -5,8 +5,8 @@ expression: res_vec
 [
     "Function {\nname: \"foo\",\nsig: \"fn[[a:list[int32], b:tuple[T, float]], A[B, bool]]\",\nvar_id: []\n}\n",
     "Class {\nname: \"A\",\nancestors: [\"A[T, V]\"],\nfields: [\"a\", \"b\"],\nmethods: [(\"__init__\", \"fn[[v:V], none]\"), (\"fun\", \"fn[[a:T], V]\")],\ntype_vars: [\"T\", \"V\"]\n}\n",
-    "Function {\nname: \"A.__init__\",\nsig: \"fn[[v:V], none]\",\nvar_id: [TypeVarId(247)]\n}\n",
-    "Function {\nname: \"A.fun\",\nsig: \"fn[[a:T], V]\",\nvar_id: [TypeVarId(252)]\n}\n",
+    "Function {\nname: \"A.__init__\",\nsig: \"fn[[v:V], none]\",\nvar_id: [TypeVarId(248)]\n}\n",
+    "Function {\nname: \"A.fun\",\nsig: \"fn[[a:T], V]\",\nvar_id: [TypeVarId(253)]\n}\n",
     "Function {\nname: \"gfun\",\nsig: \"fn[[a:A[list[float], int32]], none]\",\nvar_id: []\n}\n",
     "Class {\nname: \"B\",\nancestors: [\"B\"],\nfields: [],\nmethods: [(\"__init__\", \"fn[[], none]\")],\ntype_vars: []\n}\n",
     "Function {\nname: \"B.__init__\",\nsig: \"fn[[], none]\",\nvar_id: []\n}\n",

nac3core/src/toplevel/snapshots/nac3core__toplevel__test__test_analyze__self1.snap

@@ -3,7 +3,7 @@ source: nac3core/src/toplevel/test.rs
 expression: res_vec
 ---
 [
-    "Class {\nname: \"A\",\nancestors: [\"A[typevar233, typevar234]\"],\nfields: [\"a\", \"b\"],\nmethods: [(\"__init__\", \"fn[[a:A[float, bool], b:B], none]\"), (\"fun\", \"fn[[a:A[float, bool]], A[bool, int32]]\")],\ntype_vars: [\"typevar233\", \"typevar234\"]\n}\n",
+    "Class {\nname: \"A\",\nancestors: [\"A[typevar234, typevar235]\"],\nfields: [\"a\", \"b\"],\nmethods: [(\"__init__\", \"fn[[a:A[float, bool], b:B], none]\"), (\"fun\", \"fn[[a:A[float, bool]], A[bool, int32]]\")],\ntype_vars: [\"typevar234\", \"typevar235\"]\n}\n",
     "Function {\nname: \"A.__init__\",\nsig: \"fn[[a:A[float, bool], b:B], none]\",\nvar_id: []\n}\n",
     "Function {\nname: \"A.fun\",\nsig: \"fn[[a:A[float, bool]], A[bool, int32]]\",\nvar_id: []\n}\n",
     "Class {\nname: \"B\",\nancestors: [\"B\", \"A[int64, bool]\"],\nfields: [\"a\", \"b\"],\nmethods: [(\"__init__\", \"fn[[], none]\"), (\"fun\", \"fn[[a:A[float, bool]], A[bool, int32]]\"), (\"foo\", \"fn[[b:B], B]\"), (\"bar\", \"fn[[a:A[list[B], int32]], tuple[A[virtual[A[B, int32]], bool], B]]\")],\ntype_vars: []\n}\n",

nac3core/src/toplevel/snapshots/nac3core__toplevel__test__test_analyze__simple_class_compose.snap

@@ -6,12 +6,12 @@ expression: res_vec
     "Class {\nname: \"A\",\nancestors: [\"A\"],\nfields: [\"a\"],\nmethods: [(\"__init__\", \"fn[[], none]\"), (\"fun\", \"fn[[b:B], none]\"), (\"foo\", \"fn[[a:T, b:V], none]\")],\ntype_vars: []\n}\n",
     "Function {\nname: \"A.__init__\",\nsig: \"fn[[], none]\",\nvar_id: []\n}\n",
     "Function {\nname: \"A.fun\",\nsig: \"fn[[b:B], none]\",\nvar_id: []\n}\n",
-    "Function {\nname: \"A.foo\",\nsig: \"fn[[a:T, b:V], none]\",\nvar_id: [TypeVarId(253)]\n}\n",
+    "Function {\nname: \"A.foo\",\nsig: \"fn[[a:T, b:V], none]\",\nvar_id: [TypeVarId(254)]\n}\n",
     "Class {\nname: \"B\",\nancestors: [\"B\", \"C\", \"A\"],\nfields: [\"a\"],\nmethods: [(\"__init__\", \"fn[[], none]\"), (\"fun\", \"fn[[b:B], none]\"), (\"foo\", \"fn[[a:T, b:V], none]\")],\ntype_vars: []\n}\n",
     "Function {\nname: \"B.__init__\",\nsig: \"fn[[], none]\",\nvar_id: []\n}\n",
     "Class {\nname: \"C\",\nancestors: [\"C\", \"A\"],\nfields: [\"a\"],\nmethods: [(\"__init__\", \"fn[[], none]\"), (\"fun\", \"fn[[b:B], none]\"), (\"foo\", \"fn[[a:T, b:V], none]\")],\ntype_vars: []\n}\n",
     "Function {\nname: \"C.__init__\",\nsig: \"fn[[], none]\",\nvar_id: []\n}\n",
     "Function {\nname: \"C.fun\",\nsig: \"fn[[b:B], none]\",\nvar_id: []\n}\n",
     "Function {\nname: \"foo\",\nsig: \"fn[[a:A], none]\",\nvar_id: []\n}\n",
-    "Function {\nname: \"ff\",\nsig: \"fn[[a:T], V]\",\nvar_id: [TypeVarId(261)]\n}\n",
+    "Function {\nname: \"ff\",\nsig: \"fn[[a:T], V]\",\nvar_id: [TypeVarId(262)]\n}\n",
 ]

nac3core/src/toplevel/type_annotation.rs

@@ -552,7 +552,7 @@ pub fn get_type_from_type_annotation_kinds(
                     )
                 })
                 .collect::<Result<Vec<_>, _>>()?;
-            Ok(unifier.add_ty(TypeEnum::TTuple { ty: tys }))
+            Ok(unifier.add_ty(TypeEnum::TTuple { ty: tys, is_vararg_ctx: false }))
         }
     }
 }

nac3core/src/typecheck/function_check.rs

@@ -34,13 +34,18 @@ impl<'a> Inferencer<'a> {
                 self.should_have_value(pattern)?;
                 Ok(())
             }
-            ExprKind::Tuple { elts, .. } => {
+            ExprKind::List { elts, .. } | ExprKind::Tuple { elts, .. } => {
                 for elt in elts {
                     self.check_pattern(elt, defined_identifiers)?;
                     self.should_have_value(elt)?;
                 }
                 Ok(())
             }
+            ExprKind::Starred { value, .. } => {
+                self.check_pattern(value, defined_identifiers)?;
+                self.should_have_value(value)?;
+                Ok(())
+            }
             ExprKind::Subscript { value, slice, .. } => {
                 self.check_expr(value, defined_identifiers)?;
                 self.should_have_value(value)?;
@@ -218,7 +223,7 @@ impl<'a> Inferencer<'a> {
             ]
             .iter()
             .any(|allowed_ty| self.unifier.unioned(ret_ty, *allowed_ty)),
-            TypeEnum::TTuple { ty } => ty.iter().all(|t| self.check_return_value_ty(*t)),
+            TypeEnum::TTuple { ty, .. } => ty.iter().all(|t| self.check_return_value_ty(*t)),
             _ => false,
         }
     }

nac3core/src/typecheck/magic_methods.rs

@@ -197,6 +197,7 @@ pub fn impl_binop(
                             ty: other_ty,
                             default_value: None,
                             name: "other".into(),
+                            is_vararg: false,
                         }],
                     })),
                     false,
@@ -261,6 +262,7 @@ pub fn impl_cmpop(
                             ty: other_ty,
                             default_value: None,
                             name: "other".into(),
+                            is_vararg: false,
                         }],
                     })),
                     false,

nac3core/src/typecheck/type_error.rs

@@ -183,9 +183,10 @@ impl<'a> Display for DisplayTypeError<'a> {
                         }
                         result
                     }
-                    (TypeEnum::TTuple { ty: ty1 }, TypeEnum::TTuple { ty: ty2 })
-                        if ty1.len() != ty2.len() =>
-                    {
+                    (
+                        TypeEnum::TTuple { ty: ty1, is_vararg_ctx: is_vararg1 },
+                        TypeEnum::TTuple { ty: ty2, is_vararg_ctx: is_vararg2 },
+                    ) if !is_vararg1 && !is_vararg2 && ty1.len() != ty2.len() => {
                         let t1 = self.unifier.stringify_with_notes(*t1, &mut notes);
                         let t2 = self.unifier.stringify_with_notes(*t2, &mut notes);
                         write!(f, "Tuple length mismatch: got {t1} and {t2}")

nac3core/src/typecheck/type_inferencer/test.rs

@@ -83,7 +83,12 @@ impl TestEnvironment {
         });
         with_fields(&mut unifier, int32, |unifier, fields| {
             let add_ty = unifier.add_ty(TypeEnum::TFunc(FunSignature {
-                args: vec![FuncArg { name: "other".into(), ty: int32, default_value: None }],
+                args: vec![FuncArg {
+                    name: "other".into(),
+                    ty: int32,
+                    default_value: None,
+                    is_vararg: false,
+                }],
                 ret: int32,
                 vars: VarMap::new(),
             }));
@@ -224,7 +229,12 @@ impl TestEnvironment {
         });
         with_fields(&mut unifier, int32, |unifier, fields| {
             let add_ty = unifier.add_ty(TypeEnum::TFunc(FunSignature {
-                args: vec![FuncArg { name: "other".into(), ty: int32, default_value: None }],
+                args: vec![FuncArg {
+                    name: "other".into(),
+                    ty: int32,
+                    default_value: None,
+                    is_vararg: false,
+                }],
                 ret: int32,
                 vars: VarMap::new(),
             }));

nac3core/src/typecheck/typedef/mod.rs

@@ -1,15 +1,14 @@
 use indexmap::IndexMap;
-use itertools::Itertools;
+use itertools::{repeat_n, Itertools};
+use nac3parser::ast::{Cmpop, Location, StrRef, Unaryop};
 use std::cell::RefCell;
 use std::collections::HashMap;
 use std::fmt::{self, Display};
-use std::iter::zip;
+use std::iter::{repeat, zip};
 use std::rc::Rc;
 use std::sync::{Arc, Mutex};
 use std::{borrow::Cow, collections::HashSet};
 
-use nac3parser::ast::{Cmpop, Location, StrRef, Unaryop};
-
 use super::magic_methods::Binop;
 use super::type_error::{TypeError, TypeErrorKind};
 use super::unification_table::{UnificationKey, UnificationTable};
@@ -115,6 +114,7 @@ pub struct FuncArg {
     pub name: StrRef,
     pub ty: Type,
     pub default_value: Option<SymbolValue>,
+    pub is_vararg: bool,
 }
 
 impl FuncArg {
@@ -233,6 +233,12 @@ pub enum TypeEnum {
     TTuple {
         /// The types of elements present in this tuple.
         ty: Vec<Type>,
+
+        /// Whether this tuple is used in a vararg context.
+        ///
+        /// If `true`, `ty` must only contain one type, and the tuple is assumed to contain any
+        /// number of `ty`-typed values.
+        is_vararg_ctx: bool,
     },
 
     /// An object type.
@@ -527,7 +533,7 @@ impl Unifier {
             TypeEnum::TVirtual { ty } => self.get_instantiations(*ty).map(|ty| {
                 ty.iter().map(|&ty| self.add_ty(TypeEnum::TVirtual { ty })).collect_vec()
             }),
-            TypeEnum::TTuple { ty } => {
+            TypeEnum::TTuple { ty, is_vararg_ctx } => {
                 let tuples = ty
                     .iter()
                     .map(|ty| self.get_instantiations(*ty).unwrap_or_else(|| vec![*ty]))
@@ -537,7 +543,12 @@ impl Unifier {
                     None
                 } else {
                     Some(
-                        tuples.into_iter().map(|ty| self.add_ty(TypeEnum::TTuple { ty })).collect(),
+                        tuples
+                            .into_iter()
+                            .map(|ty| {
+                                self.add_ty(TypeEnum::TTuple { ty, is_vararg_ctx: *is_vararg_ctx })
+                            })
+                            .collect(),
                     )
                 }
             }
@@ -581,7 +592,7 @@ impl Unifier {
             TVar { .. } => allowed_typevars.iter().any(|b| self.unification_table.unioned(a, *b)),
             TCall { .. } => false,
             TVirtual { ty } => self.is_concrete(*ty, allowed_typevars),
-            TTuple { ty } => ty.iter().all(|ty| self.is_concrete(*ty, allowed_typevars)),
+            TTuple { ty, .. } => ty.iter().all(|ty| self.is_concrete(*ty, allowed_typevars)),
             TObj { params: vars, .. } => {
                 vars.values().all(|ty| self.is_concrete(*ty, allowed_typevars))
             }
@@ -649,6 +660,7 @@ impl Unifier {
 
         // Get details about the function signature/parameters.
         let num_params = signature.args.len();
+        let is_vararg = signature.args.iter().any(|arg| arg.is_vararg);
 
         // Force the type vars in `b` and `signature' to be up-to-date.
         let b = self.instantiate_fun(b, signature);
@@ -737,7 +749,7 @@ impl Unifier {
                 };
 
                 // Check for "too many arguments"
-                if num_params < posargs.len() {
+                if !is_vararg && num_params < posargs.len() {
                     let expected_min_count =
                         signature.args.iter().filter(|param| param.is_required()).count();
                     let expected_max_count = num_params;
@@ -770,6 +782,19 @@ impl Unifier {
                     type_check_arg(param.name, param.ty, arg_ty)?;
                 }
 
+                if is_vararg {
+                    debug_assert!(!signature.args.is_empty());
+
+                    let vararg_args = posargs.iter().skip(signature.args.len());
+                    let vararg_param = signature.args.last().unwrap();
+
+                    for (&arg_ty, param) in zip(vararg_args, repeat(vararg_param)) {
+                        // `param_info` for this argument would've already been marked as supplied
+                        // during non-vararg posarg typecheck
+                        type_check_arg(param.name, param.ty, arg_ty)?;
+                    }
+                }
+
                 // Now consume all keyword arguments and typecheck them.
                 for (&param_name, &arg_ty) in kwargs {
                     // We will also use this opportunity to check if this keyword argument is "legal".
@@ -959,7 +984,10 @@ impl Unifier {
                 self.unify_impl(x, b, false)?;
                 self.set_a_to_b(a, x);
             }
-            (TVar { fields: Some(fields), range, is_const_generic: false, .. }, TTuple { ty }) => {
+            (
+                TVar { fields: Some(fields), range, is_const_generic: false, .. },
+                TTuple { ty, .. },
+            ) => {
                 let len = i32::try_from(ty.len()).unwrap();
                 for (k, v) in fields {
                     match *k {
@@ -1056,15 +1084,47 @@ impl Unifier {
                 self.set_a_to_b(a, b);
             }
 
-            (TTuple { ty: ty1 }, TTuple { ty: ty2 }) => {
-                if ty1.len() != ty2.len() {
-                    return Err(TypeError::new(TypeErrorKind::IncompatibleTypes(a, b), None));
-                }
-                for (x, y) in ty1.iter().zip(ty2.iter()) {
-                    if self.unify_impl(*x, *y, false).is_err() {
-                        return Err(TypeError::new(TypeErrorKind::IncompatibleTypes(a, b), None));
+            (
+                TTuple { ty: ty1, is_vararg_ctx: is_vararg1 },
+                TTuple { ty: ty2, is_vararg_ctx: is_vararg2 },
+            ) => {
+                // Rules for Tuples:
+                // - ty1: is_vararg && ty2: is_vararg -> ty1[0] == ty2[0]
+                // - ty1: is_vararg && ty2: !is_vararg -> type error (not enough info to infer the correct number of arguments)
+                // - ty1: !is_vararg && ty2: is_vararg -> ty1[..] == ty2[0]
+                // - ty1: !is_vararg && ty2: !is_vararg -> ty1.len() == ty2.len() && ty1[i] == ty2[i]
+
+                debug_assert!(!is_vararg1 || ty1.len() == 1);
+                debug_assert!(!is_vararg2 || ty2.len() == 1);
+
+                match (*is_vararg1, *is_vararg2) {
+                    (true, true) => {
+                        if self.unify_impl(ty1[0], ty2[0], false).is_err() {
+                            return Self::incompatible_types(a, b);
+                        }
+                    }
+                    (true, false) => return Self::incompatible_types(a, b),
+
+                    (false, true) => {
+                        for y in ty2 {
+                            if self.unify_impl(ty1[0], *y, false).is_err() {
+                                return Self::incompatible_types(a, b);
+                            }
+                        }
+                    }
+                    (false, false) => {
+                        if ty1.len() != ty2.len() {
+                            return Self::incompatible_types(a, b);
+                        }
+
+                        for (x, y) in ty1.iter().zip(ty2.iter()) {
+                            if self.unify_impl(*x, *y, false).is_err() {
+                                return Self::incompatible_types(a, b);
+                            }
+                        }
                     }
                 }
+
                 self.set_a_to_b(a, b);
             }
             (TVar { fields: Some(map), range, .. }, TObj { obj_id, fields, params }) => {
@@ -1307,10 +1367,22 @@ impl Unifier {
             TypeEnum::TLiteral { values, .. } => {
                 format!("const({})", values.iter().map(|v| format!("{v:?}")).join(", "))
             }
-            TypeEnum::TTuple { ty } => {
-                let mut fields =
-                    ty.iter().map(|v| self.internal_stringify(*v, obj_to_name, var_to_name, notes));
-                format!("tuple[{}]", fields.join(", "))
+            TypeEnum::TTuple { ty, is_vararg_ctx } => {
+                if *is_vararg_ctx {
+                    debug_assert_eq!(ty.len(), 1);
+                    let field = self.internal_stringify(
+                        *ty.iter().next().unwrap(),
+                        obj_to_name,
+                        var_to_name,
+                        notes,
+                    );
+                    format!("tuple[*{field}]")
+                } else {
+                    let mut fields = ty
+                        .iter()
+                        .map(|v| self.internal_stringify(*v, obj_to_name, var_to_name, notes));
+                    format!("tuple[{}]", fields.join(", "))
+                }
             }
             TypeEnum::TVirtual { ty } => {
                 format!(
@@ -1335,17 +1407,21 @@ impl Unifier {
                     .args
                     .iter()
                     .map(|arg| {
+                        let vararg_prefix = if arg.is_vararg { "*" } else { "" };
+
                         if let Some(dv) = &arg.default_value {
                             format!(
-                                "{}:{}={}",
+                                "{}:{}{}={}",
                                 arg.name,
+                                vararg_prefix,
                                 self.internal_stringify(arg.ty, obj_to_name, var_to_name, notes),
                                 dv
                             )
                         } else {
                             format!(
-                                "{}:{}",
+                                "{}:{}{}",
                                 arg.name,
+                                vararg_prefix,
                                 self.internal_stringify(arg.ty, obj_to_name, var_to_name, notes)
                             )
                         }
@@ -1431,7 +1507,7 @@ impl Unifier {
         match &*ty {
             TypeEnum::TRigidVar { .. } | TypeEnum::TLiteral { .. } => None,
             TypeEnum::TVar { id, .. } => mapping.get(id).copied(),
-            TypeEnum::TTuple { ty } => {
+            TypeEnum::TTuple { ty, is_vararg_ctx } => {
                 let mut new_ty = Cow::from(ty);
                 for (i, t) in ty.iter().enumerate() {
                     if let Some(t1) = self.subst_impl(*t, mapping, cache) {
@@ -1439,7 +1515,10 @@ impl Unifier {
                     }
                 }
                 if matches!(new_ty, Cow::Owned(_)) {
-                    Some(self.add_ty(TypeEnum::TTuple { ty: new_ty.into_owned() }))
+                    Some(self.add_ty(TypeEnum::TTuple {
+                        ty: new_ty.into_owned(),
+                        is_vararg_ctx: *is_vararg_ctx,
+                    }))
                 } else {
                     None
                 }
@@ -1599,16 +1678,37 @@ impl Unifier {
                 }
             }
             (TVar { range, .. }, _) => self.check_var_compatibility(b, range).or(Err(())),
-            (TTuple { ty: ty1 }, TTuple { ty: ty2 }) if ty1.len() == ty2.len() => {
-                let ty: Vec<_> = zip(ty1.iter(), ty2.iter())
-                    .map(|(a, b)| self.get_intersection(*a, *b))
-                    .try_collect()?;
-                if ty.iter().any(Option::is_some) {
-                    Ok(Some(self.add_ty(TTuple {
-                        ty: zip(ty, ty1.iter()).map(|(a, b)| a.unwrap_or(*b)).collect(),
-                    })))
+            (
+                TTuple { ty: ty1, is_vararg_ctx: is_vararg1 },
+                TTuple { ty: ty2, is_vararg_ctx: is_vararg2 },
+            ) => {
+                if *is_vararg1 && *is_vararg2 {
+                    let isect_ty = self.get_intersection(ty1[0], ty2[0])?;
+                    Ok(isect_ty.map(|ty| self.add_ty(TTuple { ty: vec![ty], is_vararg_ctx: true })))
                 } else {
-                    Ok(None)
+                    let zip_iter: Box<dyn Iterator<Item = (&Type, &Type)>> =
+                        match (*is_vararg1, *is_vararg2) {
+                            (true, _) => Box::new(repeat_n(&ty1[0], ty2.len()).zip(ty2.iter())),
+                            (_, false) => Box::new(ty1.iter().zip(repeat_n(&ty2[0], ty1.len()))),
+                            _ => {
+                                if ty1.len() != ty2.len() {
+                                    return Err(());
+                                }
+
+                                Box::new(ty1.iter().zip(ty2.iter()))
+                            }
+                        };
+
+                    let ty: Vec<_> =
+                        zip_iter.map(|(a, b)| self.get_intersection(*a, *b)).try_collect()?;
+                    Ok(if ty.iter().any(Option::is_some) {
+                        Some(self.add_ty(TTuple {
+                            ty: zip(ty, ty1.iter()).map(|(a, b)| a.unwrap_or(*b)).collect(),
+                            is_vararg_ctx: false,
+                        }))
+                    } else {
+                        None
+                    })
                 }
             }
             // TODO(Derppening): #444

nac3core/src/typecheck/typedef/test.rs

@@ -28,7 +28,10 @@ impl Unifier {
                 TypeEnum::TVar { fields: Some(map1), .. },
                 TypeEnum::TVar { fields: Some(map2), .. },
             ) => self.map_eq2(map1, map2),
-            (TypeEnum::TTuple { ty: ty1 }, TypeEnum::TTuple { ty: ty2 }) => {
+            (
+                TypeEnum::TTuple { ty: ty1, is_vararg_ctx: false },
+                TypeEnum::TTuple { ty: ty2, is_vararg_ctx: false },
+            ) => {
                 ty1.len() == ty2.len()
                     && ty1.iter().zip(ty2.iter()).all(|(t1, t2)| self.eq(*t1, *t2))
             }
@@ -178,7 +181,7 @@ impl TestEnvironment {
                     ty.push(result.0);
                     s = result.1;
                 }
-                (self.unifier.add_ty(TypeEnum::TTuple { ty }), &s[1..])
+                (self.unifier.add_ty(TypeEnum::TTuple { ty, is_vararg_ctx: false }), &s[1..])
             }
             "Record" => {
                 let mut s = &typ[end..];
@@ -608,7 +611,7 @@ fn test_instantiation() {
     let v1 = env.unifier.get_fresh_var_with_range(&[list_v, int], None, None).ty;
     let v2 = env.unifier.get_fresh_var_with_range(&[list_int, float], None, None).ty;
     let t = env.unifier.get_dummy_var().ty;
-    let tuple = env.unifier.add_ty(TypeEnum::TTuple { ty: vec![v, v1, v2] });
+    let tuple = env.unifier.add_ty(TypeEnum::TTuple { ty: vec![v, v1, v2], is_vararg_ctx: false });
     let v3 = env.unifier.get_fresh_var_with_range(&[tuple, t], None, None).ty;
     // t = TypeVar('t')
     // v = TypeVar('v', int, bool)

nac3standalone/demo/check_demo.sh

@@ -3,26 +3,49 @@
 set -e
 
 if [ -z "$1" ]; then
-  echo "Requires at least one argument"
-  exit 1
+    echo "No argument supplied"
+    exit 1
 fi
 
 declare -a nac3args
+while [ $# -ge 2 ]; do
+  case "$1" in
+    --help)
+      echo "Usage: check_demo.sh [-i686] -- demo [NAC3ARGS...]"
+      exit
+      ;;
+    -i686)
+      i686=1
+      ;;
+    --)
+      shift
+      break
+      ;;
+    *)
+      break
+      ;;
+  esac
+  shift
+done
+
+demo="$1"
+shift
 while [ $# -gt 1 ]; do
   nac3args+=("$1")
   shift
 done
-demo="$1"
+
 
 echo "### Checking $demo..."
 
-# Get reference output
 echo ">>>>>> Running $demo with the Python interpreter"
 ./interpret_demo.py "$demo" > interpreted.log
 
-echo "...... Trying NAC3's 32-bit code generator output"
-./run_demo.sh -i386 --out run_32.log "${nac3args[@]}" "$demo"
-diff -Nau interpreted.log run_32.log
+if [ -n "$i686" ]; then
+  echo "...... Trying NAC3's 32-bit code generator output"
+  ./run_demo.sh -i686 --out run_32.log "${nac3args[@]}" "$demo"
+  diff -Nau interpreted.log run_32.log
+fi
 
 echo "...... Trying NAC3's 64-bit code generator output"
 ./run_demo.sh --out run_64.log "${nac3args[@]}" "$demo"

nac3standalone/demo/interpret_demo.py

@@ -6,6 +6,7 @@ import importlib.machinery
 import math
 import numpy as np
 import numpy.typing as npt
+import scipy as sp
 import pathlib
 
 from numpy import int32, int64, uint32, uint64
@@ -217,6 +218,8 @@ def patch(module):
     module.np_ldexp = np.ldexp
     module.np_hypot = np.hypot
     module.np_nextafter = np.nextafter
+    module.np_transpose = np.transpose
+    module.np_reshape = np.reshape
 
     # SciPy Math functions
     module.sp_spec_erf = special.erf
@@ -226,6 +229,20 @@ def patch(module):
     module.sp_spec_j0 = special.j0
     module.sp_spec_j1 = special.j1
 
+    # Linalg functions
+    module.np_dot = np.dot
+    module.np_linalg_cholesky = np.linalg.cholesky
+    module.np_linalg_qr = np.linalg.qr
+    module.np_linalg_svd = np.linalg.svd
+    module.np_linalg_inv = np.linalg.inv
+    module.np_linalg_pinv = np.linalg.pinv
+    module.np_linalg_matrix_power = np.linalg.matrix_power
+    module.np_linalg_det = np.linalg.det
+    
+    module.sp_linalg_lu = lambda x: sp.linalg.lu(x, True)
+    module.sp_linalg_schur = sp.linalg.schur
+    module.sp_linalg_hessenberg = lambda x: sp.linalg.hessenberg(x, True)
+
 def file_import(filename, prefix="file_import_"):
     filename = pathlib.Path(filename)
     modname = prefix + filename.stem

nac3standalone/demo/linalg/Cargo.lock

@@ -0,0 +1,114 @@
+# This file is automatically @generated by Cargo.
+# It is not intended for manual editing.
+version = 3
+
+[[package]]
+name = "approx"
+version = "0.5.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "cab112f0a86d568ea0e627cc1d6be74a1e9cd55214684db5561995f6dad897c6"
+dependencies = [
+ "num-traits",
+]
+
+[[package]]
+name = "autocfg"
+version = "1.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0c4b4d0bd25bd0b74681c0ad21497610ce1b7c91b1022cd21c80c6fbdd9476b0"
+
+[[package]]
+name = "cslice"
+version = "0.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0f8cb7306107e4b10e64994de6d3274bd08996a7c1322a27b86482392f96be0a"
+
+[[package]]
+name = "libm"
+version = "0.2.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4ec2a862134d2a7d32d7983ddcdd1c4923530833c9f2ea1a44fc5fa473989058"
+
+[[package]]
+name = "linalg"
+version = "0.1.0"
+dependencies = [
+ "cslice",
+ "nalgebra",
+]
+
+[[package]]
+name = "nalgebra"
+version = "0.32.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7b5c17de023a86f59ed79891b2e5d5a94c705dbe904a5b5c9c952ea6221b03e4"
+dependencies = [
+ "approx",
+ "num-complex",
+ "num-rational",
+ "num-traits",
+ "simba",
+ "typenum",
+]
+
+[[package]]
+name = "num-complex"
+version = "0.4.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "73f88a1307638156682bada9d7604135552957b7818057dcef22705b4d509495"
+dependencies = [
+ "num-traits",
+]
+
+[[package]]
+name = "num-integer"
+version = "0.1.46"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7969661fd2958a5cb096e56c8e1ad0444ac2bbcd0061bd28660485a44879858f"
+dependencies = [
+ "num-traits",
+]
+
+[[package]]
+name = "num-rational"
+version = "0.4.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f83d14da390562dca69fc84082e73e548e1ad308d24accdedd2720017cb37824"
+dependencies = [
+ "num-integer",
+ "num-traits",
+]
+
+[[package]]
+name = "num-traits"
+version = "0.2.19"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "071dfc062690e90b734c0b2273ce72ad0ffa95f0c74596bc250dcfd960262841"
+dependencies = [
+ "autocfg",
+ "libm",
+]
+
+[[package]]
+name = "paste"
+version = "1.0.15"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "57c0d7b74b563b49d38dae00a0c37d4d6de9b432382b2892f0574ddcae73fd0a"
+
+[[package]]
+name = "simba"
+version = "0.8.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "061507c94fc6ab4ba1c9a0305018408e312e17c041eb63bef8aa726fa33aceae"
+dependencies = [
+ "approx",
+ "num-complex",
+ "num-traits",
+ "paste",
+]
+
+[[package]]
+name = "typenum"
+version = "1.17.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "42ff0bf0c66b8238c6f3b578df37d0b7848e55df8577b3f74f92a69acceeb825"

nac3standalone/demo/linalg/Cargo.toml

@@ -0,0 +1,13 @@
+[package]
+name = "linalg"
+version = "0.1.0"
+edition = "2021"
+
+[lib]
+crate-type = ["staticlib"]
+
+[dependencies]
+nalgebra = {version = "0.32.6", default-features = false, features = ["libm", "alloc"]}
+cslice = "0.3.0"
+
+[workspace]

nac3standalone/demo/linalg/src/lib.rs

@@ -0,0 +1,406 @@
+// Uses `nalgebra` crate to invoke `np_linalg` and `sp_linalg` functions
+// When converting between `nalgebra::Matrix` and `NDArray` following considerations are necessary
+//
+// * Both `nalgebra::Matrix` and `NDArray` require their content to be stored in row-major order
+// * `NDArray` data pointer can be directly read and converted to `nalgebra::Matrix` (row and column number must be known)
+// * `nalgebra::Matrix::as_slice` returns the content of matrix in column-major order and initial data needs to be transposed before storing it in `NDArray` data pointer
+
+use core::slice;
+use nalgebra::DMatrix;
+
+fn report_error(
+    error_name: &str,
+    fn_name: &str,
+    file_name: &str,
+    line_num: u32,
+    col_num: u32,
+    err_msg: &str,
+) -> ! {
+    panic!(
+        "Exception {} from {} in {}:{}:{}, message: {}",
+        error_name, fn_name, file_name, line_num, col_num, err_msg
+    );
+}
+
+pub struct InputMatrix {
+    pub ndims: usize,
+    pub dims: *const usize,
+    pub data: *mut f64,
+}
+impl InputMatrix {
+    fn get_dims(&mut self) -> Vec<usize> {
+        let dims = unsafe { slice::from_raw_parts(self.dims, self.ndims) };
+        dims.to_vec()
+    }
+}
+
+/// # Safety
+///
+/// `mat1` should point to a valid 2DArray of `f64` floats in row-major order
+#[no_mangle]
+pub unsafe extern "C" fn np_linalg_cholesky(mat1: *mut InputMatrix, out: *mut InputMatrix) {
+    let mat1 = mat1.as_mut().unwrap();
+    let out = out.as_mut().unwrap();
+
+    if mat1.ndims != 2 {
+        let err_msg = format!("expected 2D Vector Input, but received {}D input", mat1.ndims);
+        report_error("ValueError", "np_linalg_cholesky", file!(), line!(), column!(), &err_msg);
+    }
+
+    let dim1 = (*mat1).get_dims();
+    if dim1[0] != dim1[1] {
+        let err_msg =
+            format!("last 2 dimensions of the array must be square: {0} != {1}", dim1[0], dim1[1]);
+        report_error("LinAlgError", "np_linalg_cholesky", file!(), line!(), column!(), &err_msg);
+    }
+
+    let outdim = out.get_dims();
+    let out_slice = unsafe { slice::from_raw_parts_mut(out.data, outdim[0] * outdim[1]) };
+    let data_slice1 = unsafe { slice::from_raw_parts_mut(mat1.data, dim1[0] * dim1[1]) };
+
+    let matrix1 = DMatrix::from_row_slice(dim1[0], dim1[1], data_slice1);
+    let result = matrix1.cholesky();
+    match result {
+        Some(res) => {
+            out_slice.copy_from_slice(res.unpack().transpose().as_slice());
+        }
+        None => {
+            report_error(
+                "LinAlgError",
+                "np_linalg_cholesky",
+                file!(),
+                line!(),
+                column!(),
+                "Matrix is not positive definite",
+            );
+        }
+    };
+}
+
+/// # Safety
+///
+/// `mat1` should point to a valid 2DArray of `f64` floats in row-major order
+#[no_mangle]
+pub unsafe extern "C" fn np_linalg_qr(
+    mat1: *mut InputMatrix,
+    out_q: *mut InputMatrix,
+    out_r: *mut InputMatrix,
+) {
+    let mat1 = mat1.as_mut().unwrap();
+    let out_q = out_q.as_mut().unwrap();
+    let out_r = out_r.as_mut().unwrap();
+
+    if mat1.ndims != 2 {
+        let err_msg = format!("expected 2D Vector Input, but received {}D input", mat1.ndims);
+        report_error("ValueError", "np_linalg_cholesky", file!(), line!(), column!(), &err_msg);
+    }
+
+    let dim1 = (*mat1).get_dims();
+    let outq_dim = (*out_q).get_dims();
+    let outr_dim = (*out_r).get_dims();
+
+    let data_slice1 = unsafe { slice::from_raw_parts_mut(mat1.data, dim1[0] * dim1[1]) };
+    let out_q_slice = unsafe { slice::from_raw_parts_mut(out_q.data, outq_dim[0] * outq_dim[1]) };
+    let out_r_slice = unsafe { slice::from_raw_parts_mut(out_r.data, outr_dim[0] * outr_dim[1]) };
+
+    // Refer to https://github.com/dimforge/nalgebra/issues/735
+    let matrix1 = DMatrix::from_row_slice(dim1[0], dim1[1], data_slice1);
+
+    let res = matrix1.qr();
+    let (q, r) = res.unpack();
+
+    // Uses different algo need to match numpy
+    out_q_slice.copy_from_slice(q.transpose().as_slice());
+    out_r_slice.copy_from_slice(r.transpose().as_slice());
+}
+
+/// # Safety
+///
+/// `mat1` should point to a valid 2DArray of `f64` floats in row-major order
+#[no_mangle]
+pub unsafe extern "C" fn np_linalg_svd(
+    mat1: *mut InputMatrix,
+    outu: *mut InputMatrix,
+    outs: *mut InputMatrix,
+    outvh: *mut InputMatrix,
+) {
+    let mat1 = mat1.as_mut().unwrap();
+    let outu = outu.as_mut().unwrap();
+    let outs = outs.as_mut().unwrap();
+    let outvh = outvh.as_mut().unwrap();
+
+    if mat1.ndims != 2 {
+        let err_msg = format!("expected 2D Vector Input, but received {}D input", mat1.ndims);
+        report_error("ValueError", "np_linalg_svd", file!(), line!(), column!(), &err_msg);
+    }
+
+    let dim1 = (*mat1).get_dims();
+    let outu_dim = (*outu).get_dims();
+    let outs_dim = (*outs).get_dims();
+    let outvh_dim = (*outvh).get_dims();
+
+    let data_slice1 = unsafe { slice::from_raw_parts_mut(mat1.data, dim1[0] * dim1[1]) };
+    let out_u_slice = unsafe { slice::from_raw_parts_mut(outu.data, outu_dim[0] * outu_dim[1]) };
+    let out_s_slice = unsafe { slice::from_raw_parts_mut(outs.data, outs_dim[0]) };
+    let out_vh_slice =
+        unsafe { slice::from_raw_parts_mut(outvh.data, outvh_dim[0] * outvh_dim[1]) };
+
+    let matrix = DMatrix::from_row_slice(dim1[0], dim1[1], data_slice1);
+    let result = matrix.svd(true, true);
+    out_u_slice.copy_from_slice(result.u.unwrap().transpose().as_slice());
+    out_s_slice.copy_from_slice(result.singular_values.as_slice());
+    out_vh_slice.copy_from_slice(result.v_t.unwrap().transpose().as_slice());
+}
+
+/// # Safety
+///
+/// `mat1` should point to a valid 2DArray of `f64` floats in row-major order
+#[no_mangle]
+pub unsafe extern "C" fn np_linalg_inv(mat1: *mut InputMatrix, out: *mut InputMatrix) {
+    let mat1 = mat1.as_mut().unwrap();
+    let out = out.as_mut().unwrap();
+
+    if mat1.ndims != 2 {
+        let err_msg = format!("expected 2D Vector Input, but received {}D input", mat1.ndims);
+        report_error("ValueError", "np_linalg_inv", file!(), line!(), column!(), &err_msg);
+    }
+    let dim1 = (*mat1).get_dims();
+
+    if dim1[0] != dim1[1] {
+        let err_msg =
+            format!("last 2 dimensions of the array must be square: {0} != {1}", dim1[0], dim1[1]);
+        report_error("LinAlgError", "np_linalg_inv", file!(), line!(), column!(), &err_msg);
+    }
+
+    let outdim = out.get_dims();
+    let out_slice = unsafe { slice::from_raw_parts_mut(out.data, outdim[0] * outdim[1]) };
+    let data_slice1 = unsafe { slice::from_raw_parts_mut(mat1.data, dim1[0] * dim1[1]) };
+
+    let matrix = DMatrix::from_row_slice(dim1[0], dim1[1], data_slice1);
+    if !matrix.is_invertible() {
+        report_error(
+            "LinAlgError",
+            "np_linalg_inv",
+            file!(),
+            line!(),
+            column!(),
+            "no inverse for Singular Matrix",
+        );
+    }
+    let inv = matrix.try_inverse().unwrap();
+    out_slice.copy_from_slice(inv.transpose().as_slice());
+}
+
+/// # Safety
+///
+/// `mat1` should point to a valid 2DArray of `f64` floats in row-major order
+#[no_mangle]
+pub unsafe extern "C" fn np_linalg_pinv(mat1: *mut InputMatrix, out: *mut InputMatrix) {
+    let mat1 = mat1.as_mut().unwrap();
+    let out = out.as_mut().unwrap();
+
+    if mat1.ndims != 2 {
+        let err_msg = format!("expected 2D Vector Input, but received {}D input", mat1.ndims);
+        report_error("ValueError", "np_linalg_pinv", file!(), line!(), column!(), &err_msg);
+    }
+    let dim1 = (*mat1).get_dims();
+    let outdim = out.get_dims();
+    let out_slice = unsafe { slice::from_raw_parts_mut(out.data, outdim[0] * outdim[1]) };
+    let data_slice1 = unsafe { slice::from_raw_parts_mut(mat1.data, dim1[0] * dim1[1]) };
+
+    let matrix = DMatrix::from_row_slice(dim1[0], dim1[1], data_slice1);
+    let svd = matrix.svd(true, true);
+    let inv = svd.pseudo_inverse(1e-15);
+
+    match inv {
+        Ok(m) => {
+            out_slice.copy_from_slice(m.transpose().as_slice());
+        }
+        Err(err_msg) => {
+            report_error("LinAlgError", "np_linalg_pinv", file!(), line!(), column!(), err_msg);
+        }
+    }
+}
+
+/// # Safety
+///
+/// `mat1` should point to a valid 2DArray of `f64` floats in row-major order
+#[no_mangle]
+pub unsafe extern "C" fn np_linalg_matrix_power(
+    mat1: *mut InputMatrix,
+    mat2: *mut InputMatrix,
+    out: *mut InputMatrix,
+) {
+    let mat1 = mat1.as_mut().unwrap();
+    let mat2 = mat2.as_mut().unwrap();
+    let out = out.as_mut().unwrap();
+
+    if mat1.ndims != 2 {
+        let err_msg = format!("expected 2D Vector Input, but received {}D", mat1.ndims);
+        report_error("ValueError", "np_linalg_matrix_power", file!(), line!(), column!(), &err_msg);
+    }
+
+    let dim1 = (*mat1).get_dims();
+    let power = unsafe { slice::from_raw_parts_mut(mat2.data, 1) };
+    let power = power[0];
+    let outdim = out.get_dims();
+    let out_slice = unsafe { slice::from_raw_parts_mut(out.data, outdim[0] * outdim[1]) };
+    let data_slice1 = unsafe { slice::from_raw_parts_mut(mat1.data, dim1[0] * dim1[1]) };
+
+    let abs_pow = power.abs();
+    let matrix1 = DMatrix::from_row_slice(dim1[0], dim1[1], data_slice1);
+    let mut result = matrix1.pow(abs_pow as u32);
+
+    if power < 0.0 {
+        if !result.is_invertible() {
+            report_error(
+                "LinAlgError",
+                "np_linalg_inv",
+                file!(),
+                line!(),
+                column!(),
+                "no inverse for Singular Matrix",
+            );
+        }
+        result = result.try_inverse().unwrap();
+    }
+    out_slice.copy_from_slice(result.transpose().as_slice());
+}
+
+/// # Safety
+///
+/// `mat1` should point to a valid 2DArray of `f64` floats in row-major order
+#[no_mangle]
+pub unsafe extern "C" fn np_linalg_det(mat1: *mut InputMatrix, out: *mut InputMatrix) {
+    let mat1 = mat1.as_mut().unwrap();
+    let out = out.as_mut().unwrap();
+
+    if mat1.ndims != 2 {
+        let err_msg = format!("expected 2D Vector Input, but received {}D input", mat1.ndims);
+        report_error("ValueError", "np_linalg_det", file!(), line!(), column!(), &err_msg);
+    }
+    let dim1 = (*mat1).get_dims();
+    let out_slice = unsafe { slice::from_raw_parts_mut(out.data, 1) };
+    let data_slice1 = unsafe { slice::from_raw_parts_mut(mat1.data, dim1[0] * dim1[1]) };
+
+    let matrix = DMatrix::from_row_slice(dim1[0], dim1[1], data_slice1);
+    if !matrix.is_square() {
+        let err_msg =
+            format!("last 2 dimensions of the array must be square: {0} != {1}", dim1[0], dim1[1]);
+        report_error("LinAlgError", "np_linalg_inv", file!(), line!(), column!(), &err_msg);
+    }
+    out_slice[0] = matrix.determinant();
+}
+
+/// # Safety
+///
+/// `mat1` should point to a valid 2DArray of `f64` floats in row-major order
+#[no_mangle]
+pub unsafe extern "C" fn sp_linalg_lu(
+    mat1: *mut InputMatrix,
+    out_l: *mut InputMatrix,
+    out_u: *mut InputMatrix,
+) {
+    let mat1 = mat1.as_mut().unwrap();
+    let out_l = out_l.as_mut().unwrap();
+    let out_u = out_u.as_mut().unwrap();
+
+    if mat1.ndims != 2 {
+        let err_msg = format!("expected 2D Vector Input, but received {}D input", mat1.ndims);
+        report_error("ValueError", "sp_linalg_lu", file!(), line!(), column!(), &err_msg);
+    }
+
+    let dim1 = (*mat1).get_dims();
+    let outl_dim = (*out_l).get_dims();
+    let outu_dim = (*out_u).get_dims();
+
+    let data_slice1 = unsafe { slice::from_raw_parts_mut(mat1.data, dim1[0] * dim1[1]) };
+    let out_l_slice = unsafe { slice::from_raw_parts_mut(out_l.data, outl_dim[0] * outl_dim[1]) };
+    let out_u_slice = unsafe { slice::from_raw_parts_mut(out_u.data, outu_dim[0] * outu_dim[1]) };
+
+    let matrix = DMatrix::from_row_slice(dim1[0], dim1[1], data_slice1);
+    let (_, l, u) = matrix.lu().unpack();
+
+    out_l_slice.copy_from_slice(l.transpose().as_slice());
+    out_u_slice.copy_from_slice(u.transpose().as_slice());
+}
+
+/// # Safety
+///
+/// `mat1` should point to a valid 2DArray of `f64` floats in row-major order
+#[no_mangle]
+pub unsafe extern "C" fn sp_linalg_schur(
+    mat1: *mut InputMatrix,
+    out_t: *mut InputMatrix,
+    out_z: *mut InputMatrix,
+) {
+    let mat1 = mat1.as_mut().unwrap();
+    let out_t = out_t.as_mut().unwrap();
+    let out_z = out_z.as_mut().unwrap();
+
+    if mat1.ndims != 2 {
+        let err_msg = format!("expected 2D Vector Input, but received {}D input", mat1.ndims);
+        report_error("ValueError", "sp_linalg_schur", file!(), line!(), column!(), &err_msg);
+    }
+
+    let dim1 = (*mat1).get_dims();
+
+    if dim1[0] != dim1[1] {
+        let err_msg =
+            format!("last 2 dimensions of the array must be square: {0} != {1}", dim1[0], dim1[1]);
+        report_error("LinAlgError", "np_linalg_schur", file!(), line!(), column!(), &err_msg);
+    }
+
+    let out_t_dim = (*out_t).get_dims();
+    let out_z_dim = (*out_z).get_dims();
+
+    let data_slice1 = unsafe { slice::from_raw_parts_mut(mat1.data, dim1[0] * dim1[1]) };
+    let out_t_slice = unsafe { slice::from_raw_parts_mut(out_t.data, out_t_dim[0] * out_t_dim[1]) };
+    let out_z_slice = unsafe { slice::from_raw_parts_mut(out_z.data, out_z_dim[0] * out_z_dim[1]) };
+
+    let matrix = DMatrix::from_row_slice(dim1[0], dim1[1], data_slice1);
+    let (z, t) = matrix.schur().unpack();
+
+    out_t_slice.copy_from_slice(t.transpose().as_slice());
+    out_z_slice.copy_from_slice(z.transpose().as_slice());
+}
+
+/// # Safety
+///
+/// `mat1` should point to a valid 2DArray of `f64` floats in row-major order
+#[no_mangle]
+pub unsafe extern "C" fn sp_linalg_hessenberg(
+    mat1: *mut InputMatrix,
+    out_h: *mut InputMatrix,
+    out_q: *mut InputMatrix,
+) {
+    let mat1 = mat1.as_mut().unwrap();
+    let out_h = out_h.as_mut().unwrap();
+    let out_q = out_q.as_mut().unwrap();
+
+    if mat1.ndims != 2 {
+        let err_msg = format!("expected 2D Vector Input, but received {}D input", mat1.ndims);
+        report_error("ValueError", "sp_linalg_hessenberg", file!(), line!(), column!(), &err_msg);
+    }
+
+    let dim1 = (*mat1).get_dims();
+
+    if dim1[0] != dim1[1] {
+        let err_msg =
+            format!("last 2 dimensions of the array must be square: {} != {}", dim1[0], dim1[1]);
+        report_error("LinAlgError", "sp_linalg_hessenberg", file!(), line!(), column!(), &err_msg);
+    }
+
+    let out_h_dim = (*out_h).get_dims();
+    let out_q_dim = (*out_q).get_dims();
+
+    let data_slice1 = unsafe { slice::from_raw_parts_mut(mat1.data, dim1[0] * dim1[1]) };
+    let out_h_slice = unsafe { slice::from_raw_parts_mut(out_h.data, out_h_dim[0] * out_h_dim[1]) };
+    let out_q_slice = unsafe { slice::from_raw_parts_mut(out_q.data, out_q_dim[0] * out_q_dim[1]) };
+
+    let matrix = DMatrix::from_row_slice(dim1[0], dim1[1], data_slice1);
+    let (q, h) = matrix.hessenberg().unpack();
+
+    out_h_slice.copy_from_slice(h.transpose().as_slice());
+    out_q_slice.copy_from_slice(q.transpose().as_slice());
+}

nac3standalone/demo/run_demo.sh

@@ -2,6 +2,9 @@
 
 set -e
 
+: "${DEMO_LINALG_STUB:=linalg/target/release/liblinalg.a}"
+: "${DEMO_LINALG_STUB32:=linalg/target/i686-unknown-linux-gnu/release/liblinalg.a}"
+
 if [ -z "$1" ]; then
     echo "No argument supplied"
     exit 1
@@ -11,7 +14,7 @@ declare -a nac3args
 while [ $# -ge 1 ]; do
   case "$1" in
     --help)
-      echo "Usage: run_demo.sh [--help] [--out OUTFILE] [--debug] [-i386] -- [NAC3ARGS...]"
+      echo "Usage: run_demo.sh [--help] [--out OUTFILE] [--debug] [-i686] -- [NAC3ARGS...]"
       exit
       ;;
     --out)
@@ -21,8 +24,8 @@ while [ $# -ge 1 ]; do
     --debug)
       debug=1
       ;;
-    -i386)
-      i386=1
+    -i686)
+      i686=1
       ;;
     --)
       shift
@@ -51,18 +54,14 @@ fi
 
 rm -f ./*.o ./*.bc demo
 
-if [ -z "$i386" ]; then
+if [ -z "$i686" ]; then
   $nac3standalone "${nac3args[@]}"
-
   clang -c -std=gnu11 -Wall -Wextra -O3 -o demo.o demo.c
-  clang -lm -Wl,--no-warn-search-mismatch -o demo module.o demo.o
+  clang -o demo module.o demo.o $DEMO_LINALG_STUB -lm -Wl,--no-warn-search-mismatch
 else
-   # Enable SSE2 to avoid rounding errors with X87's 80-bit fp precision computations
-
-   $nac3standalone --triple i386-pc-linux-gnu --target-features +sse2 "${nac3args[@]}"
-
-   clang -m32 -c -std=gnu11 -Wall -Wextra -O3 -msse2 -o demo.o demo.c
-   clang -m32 -lm -Wl,--no-warn-search-mismatch -o demo module.o demo.o
+  $nac3standalone --triple i686-unknown-linux-gnu "${nac3args[@]}"
+  clang -m32 -c -std=gnu11 -Wall -Wextra -O3 -msse2 -o demo.o demo.c
+  clang -m32 -o demo module.o demo.o $DEMO_LINALG_STUB32 -lm -Wl,--no-warn-search-mismatch
 fi
 
 if [ -z "$outfile" ]; then

nac3standalone/demo/src/assignment.py

@@ -0,0 +1,66 @@
+@extern
+def output_int32(x: int32):
+    ...
+
+@extern
+def output_bool(x: bool):
+    ...
+
+def example1():
+    x, *ys, z = (1, 2, 3, 4, 5)
+    output_int32(x)
+    output_int32(ys[0])
+    output_int32(ys[1])
+    output_int32(ys[2])
+    output_int32(z)
+
+def example2():
+    x, y, *zs = (1, 2, 3, 4, 5)
+    output_int32(x)
+    output_int32(y)
+    output_int32(zs[0])
+    output_int32(zs[1])
+    output_int32(zs[2])
+
+def example3():
+    *xs, y, z = (1, 2, 3, 4, 5)
+    output_int32(xs[0])
+    output_int32(xs[1])
+    output_int32(xs[2])
+    output_int32(y)
+    output_int32(z)
+
+def example4():
+    # Example from: https://docs.python.org/3/reference/simple_stmts.html#assignment-statements
+    x = [0, 1]
+    i = 0
+    i, x[i] = 1, 2 # i is updated, then x[i] is updated
+    output_int32(i)
+    output_int32(x[0])
+    output_int32(x[1])
+
+class A:
+    value: int32
+    def __init__(self):
+        self.value = 1000
+
+def example5():
+    ws = [88, 7, 8]
+    a = A()
+    x, [y, *ys, a.value], ws[0], (ws[0],) = 1, (2, False, 4, 5), 99, (6,)
+    output_int32(x)
+    output_int32(y)
+    output_bool(ys[0])
+    output_int32(ys[1])
+    output_int32(a.value)
+    output_int32(ws[0])
+    output_int32(ws[1])
+    output_int32(ws[2])
+
+def run() -> int32:
+    example1()
+    example2()
+    example3()
+    example4()
+    example5()
+    return 0

nac3standalone/demo/src/ndarray.py

@@ -1429,6 +1429,142 @@ def test_ndarray_nextafter_broadcast_rhs_scalar():
     output_ndarray_float_2(nextafter_x_zeros)
     output_ndarray_float_2(nextafter_x_ones)
 
+def test_ndarray_transpose():
+    x: ndarray[float, 2] = np_array([[1., 2., 3.], [4., 5., 6.]])
+    y = np_transpose(x)
+    z = np_transpose(y)
+
+    output_ndarray_float_2(x)
+    output_ndarray_float_2(y)
+
+def test_ndarray_reshape():
+    w: ndarray[float, 1] = np_array([1., 2., 3., 4., 5., 6., 7., 8., 9., 10.])
+    x = np_reshape(w, (1, 2, 1, -1))
+    y = np_reshape(x, [2, -1])
+    z = np_reshape(y, 10)
+
+    x1: ndarray[int32, 1] = np_array([1, 2, 3, 4])
+    x2: ndarray[int32, 2] = np_reshape(x1, (2, 2))
+
+    output_ndarray_float_1(w)
+    output_ndarray_float_2(y)
+    output_ndarray_float_1(z)
+
+def test_ndarray_dot():
+    x1: ndarray[float, 1] = np_array([5.0, 1.0, 4.0, 2.0])
+    y1: ndarray[float, 1] = np_array([5.0, 1.0, 6.0, 6.0])
+    z1 = np_dot(x1, y1)
+
+    x2: ndarray[int32, 1] = np_array([5, 1, 4, 2])
+    y2: ndarray[int32, 1] = np_array([5, 1, 6, 6])
+    z2 = np_dot(x2, y2)
+
+    x3: ndarray[bool, 1] = np_array([True, True, True, True])
+    y3: ndarray[bool, 1] = np_array([True, True, True, True])
+    z3 = np_dot(x3, y3)
+
+    z4 = np_dot(2, 3)
+    z5 = np_dot(2., 3.)
+    z6 = np_dot(True, False)
+
+    output_float64(z1)
+    output_int32(z2)
+    output_bool(z3)
+    output_int32(z4)
+    output_float64(z5)
+    output_bool(z6)
+
+def test_ndarray_cholesky():
+    x: ndarray[float, 2] = np_array([[5.0, 1.0], [1.0, 4.0]])
+    y = np_linalg_cholesky(x)
+
+    output_ndarray_float_2(x)
+    output_ndarray_float_2(y)
+
+def test_ndarray_qr():
+    x: ndarray[float, 2] = np_array([[-5.0, -1.0, 2.0], [-1.0, 4.0, 7.5], [-1.0, 8.0, -8.5]])
+    y, z  = np_linalg_qr(x)
+
+    output_ndarray_float_2(x)
+
+    # QR Factorization is not unique and gives different results in numpy and nalgebra
+    # Reverting the decomposition to compare the initial arrays
+    a = y @ z
+    output_ndarray_float_2(a)
+
+def test_ndarray_linalg_inv():
+    x: ndarray[float, 2] = np_array([[-5.0, -1.0, 2.0], [-1.0, 4.0, 7.5], [-1.0, 8.0, -8.5]])
+    y = np_linalg_inv(x)
+
+    output_ndarray_float_2(x)
+    output_ndarray_float_2(y)
+
+def test_ndarray_pinv():
+    x: ndarray[float, 2] = np_array([[-5.0, -1.0, 2.0], [-1.0, 4.0, 7.5]])
+    y = np_linalg_pinv(x)
+
+    output_ndarray_float_2(x)
+    output_ndarray_float_2(y)
+
+def test_ndarray_matrix_power():
+    x: ndarray[float, 2] = np_array([[-5.0, -1.0, 2.0], [-1.0, 4.0, 7.5], [-1.0, 8.0, -8.5]])
+    y = np_linalg_matrix_power(x, -9)
+
+    output_ndarray_float_2(x)
+    output_ndarray_float_2(y)
+
+def test_ndarray_det():
+    x: ndarray[float, 2] = np_array([[-5.0, -1.0, 2.0], [-1.0, 4.0, 7.5], [-1.0, 8.0, -8.5]])
+    y = np_linalg_det(x)
+
+    output_ndarray_float_2(x)
+    output_float64(y)
+
+def test_ndarray_schur():
+    x: ndarray[float, 2] = np_array([[-5.0, -1.0, 2.0], [-1.0, 4.0, 7.5], [-1.0, 8.0, -8.5]])
+    t, z = sp_linalg_schur(x)
+
+    output_ndarray_float_2(x)
+
+    # Schur Factorization is not unique and gives different results in scipy and nalgebra
+    # Reverting the decomposition to compare the initial arrays
+    a = (z @ t) @ np_linalg_inv(z)
+    output_ndarray_float_2(a)
+
+def test_ndarray_hessenberg():
+    x: ndarray[float, 2] = np_array([[-5.0, -1.0, 2.0], [-1.0, 4.0, 7.5], [-1.0, 5.0, 8.5]])
+    h, q = sp_linalg_hessenberg(x)
+
+    output_ndarray_float_2(x)
+
+    # Hessenberg Factorization is not unique and gives different results in scipy and nalgebra
+    # Reverting the decomposition to compare the initial arrays
+    a = (q @ h) @ np_linalg_inv(q)
+    output_ndarray_float_2(a)
+
+
+def test_ndarray_lu():
+    x: ndarray[float, 2] = np_array([[-5.0, -1.0, 2.0], [-1.0, 4.0, 7.5]])
+    l, u = sp_linalg_lu(x)
+
+    output_ndarray_float_2(x)
+    output_ndarray_float_2(l)
+    output_ndarray_float_2(u)
+
+
+def test_ndarray_svd():
+    w: ndarray[float, 2] = np_array([[-5.0, -1.0, 2.0], [-1.0, 4.0, 7.5], [-1.0, 8.0, -8.5]])
+    x, y, z = np_linalg_svd(w)
+
+    output_ndarray_float_2(w)
+
+    # SVD Factorization is not unique and gives different results in numpy and nalgebra
+    # Reverting the decomposition to compare the initial arrays
+    a = x @ z
+    output_ndarray_float_2(a)
+    output_ndarray_float_1(y)
+
+
 def run() -> int32:
     test_ndarray_ctor()
     test_ndarray_empty()
@@ -1607,5 +1743,18 @@ def run() -> int32:
     test_ndarray_nextafter_broadcast()
     test_ndarray_nextafter_broadcast_lhs_scalar()
     test_ndarray_nextafter_broadcast_rhs_scalar()
+    test_ndarray_transpose()
+    test_ndarray_reshape()
     
+    test_ndarray_dot()
+    test_ndarray_cholesky()
+    test_ndarray_qr()
+    test_ndarray_svd()
+    test_ndarray_linalg_inv()
+    test_ndarray_pinv()
+    test_ndarray_matrix_power()
+    test_ndarray_det()
+    test_ndarray_lu()
+    test_ndarray_schur()
+    test_ndarray_hessenberg()
     return 0

nac3standalone/demo/src/vararg.py

@@ -0,0 +1,11 @@
+def f(*args: int32):
+    pass
+
+
+def run() -> int32:
+    f()
+    f(1)
+    f(1, 2)
+    f(1, 2, 3)
+
+    return 0