--- /dev/null
+target
+qemu/machine
+*.o
+*.asm
+*.out
--- /dev/null
+# This file is automatically @generated by Cargo.
+# It is not intended for manual editing.
+version = 4
+
+[[package]]
+name = "anyhow"
+version = "1.0.100"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a23eb6b1614318a8071c9b2521f36b424b2c83db5eb3a0fead4a6c0809af6e61"
+
+[[package]]
+name = "equivalent"
+version = "1.0.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "877a4ace8713b0bcf2a4e7eec82529c029f1d0619886d18145fea96c3ffe5c0f"
+
+[[package]]
+name = "hashbrown"
+version = "0.16.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "841d1cc9bed7f9236f321df977030373f4a4163ae1a7dbfe1a51a2c1a51d9100"
+
+[[package]]
+name = "indexmap"
+version = "2.12.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0ad4bb2b565bca0645f4d68c5c9af97fba094e9791da685bf83cb5f3ce74acf2"
+dependencies = [
+ "equivalent",
+ "hashbrown",
+]
+
+[[package]]
+name = "itoa"
+version = "1.0.15"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4a5f13b858c8d314ee3e8f639011f7ccefe71f97f96e50151fb991f267928e2c"
+
+[[package]]
+name = "proc-macro2"
+version = "1.0.103"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5ee95bc4ef87b8d5ba32e8b7714ccc834865276eab0aed5c9958d00ec45f49e8"
+dependencies = [
+ "unicode-ident",
+]
+
+[[package]]
+name = "quote"
+version = "1.0.42"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a338cc41d27e6cc6dce6cefc13a0729dfbb81c262b1f519331575dd80ef3067f"
+dependencies = [
+ "proc-macro2",
+]
+
+[[package]]
+name = "ryu"
+version = "1.0.20"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "28d3b2b1366ec20994f1fd18c3c594f05c5dd4bc44d8bb0c1c632c8d6829481f"
+
+[[package]]
+name = "serde"
+version = "1.0.228"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9a8e94ea7f378bd32cbbd37198a4a91436180c5bb472411e48b5ec2e2124ae9e"
+dependencies = [
+ "serde_core",
+]
+
+[[package]]
+name = "serde_core"
+version = "1.0.228"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "41d385c7d4ca58e59fc732af25c3983b67ac852c1a25000afe1175de458b67ad"
+dependencies = [
+ "serde_derive",
+]
+
+[[package]]
+name = "serde_derive"
+version = "1.0.228"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d540f220d3187173da220f885ab66608367b6574e925011a9353e4badda91d79"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
+[[package]]
+name = "serde_yaml"
+version = "0.9.34+deprecated"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6a8b1a1a2ebf674015cc02edccce75287f1a0130d394307b36743c2f5d504b47"
+dependencies = [
+ "indexmap",
+ "itoa",
+ "ryu",
+ "serde",
+ "unsafe-libyaml",
+]
+
+[[package]]
+name = "sorel-ir"
+version = "0.1.0"
+dependencies = [
+ "serde",
+ "serde_derive",
+ "serde_yaml",
+]
+
+[[package]]
+name = "sorelc"
+version = "0.1.0"
+dependencies = [
+ "anyhow",
+ "sorel-ir",
+]
+
+[[package]]
+name = "syn"
+version = "2.0.111"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "390cc9a294ab71bdb1aa2e99d13be9c753cd2d7bd6560c77118597410c4d2e87"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "unicode-ident",
+]
+
+[[package]]
+name = "unicode-ident"
+version = "1.0.22"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9312f7c4f6ff9069b165498234ce8be658059c6728633667c526e27dc2cf1df5"
+
+[[package]]
+name = "unsafe-libyaml"
+version = "0.2.11"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "673aac59facbab8a9007c7f6108d11f63b603f7cabff99fabf650fea5c32b861"
--- /dev/null
+[workspace]
+
+resolver = "3"
+members = ["sorel-ir","sorelc"]
+
+
+[workspace.dependencies]
+sorel-ir = { path = "./sorel-ir", version = "0.1.0" }
--- /dev/null
+# sorel
+
+The name means "Stack-Oriented Rethought Language".
+
+
+## TODO
+
+* [x] Imports
+* [x] Syscalls
+* [x] Loops
+* [ ] Structs
+* [ ] many, many more things
+
+## Helpful External Resources
+
+* https://gpages.juszkiewicz.com.pl/syscalls-table/syscalls.html
+* https://godbolt.org/
+* https://projectf.io/posts/riscv-cheat-sheet/
+* https://github.com/dvoytik/riscv-cheats
--- /dev/null
+## Developer setup
+
+### Nix / NixOS
+
+If you're using `Nix` or `NixOS`, the provided `flake.nix` should give you a good start.
+You'll need a recent Rust toolchain, for which you can use [`rustup`](https://rustup.sh).
+Currently that's assumed to be included on the system.
+Also, the test scripts currently assume binfmt is set up for RISC-V 64 (or that you're running on RISC-V 64).
+Assuming you're not on RISC-V 64 (most people aren't), you can add the following to your Nix `configuration.nix`:
+
+```
+
+ boot.binfmt.emulatedSystems = [
+ "riscv64-linux"
+ # ....
+ ];
+```
+
+### Ubuntu
+
+```
+# On non RISC-V 64 systems:
+apt install -y build-essential gdb crossbuild-essential-riscv64 qemu-system qemu-user qemu-user-binfmt
+```
+
+## Building the Compiler
+
+The compiler is written in Rust, and can be compiled with
+
+```
+cargo build
+```
+
+This puts the compiler binary at `target/debug/sorelc`.
+Release builds can also be created in the usual Rust/`cargo` fasion.
+
+Note that if you change any stdlib files, you'll need to rebuild.
+
+## Compiling a Sorel Program
+
+Currently, `sorelc` takes only one argument, and that's an entrypoint for your program.
+Any words in that file outside word definitions will constitute the "main" of your program.
+
+This will create a file of the same name alongside it, with the extension changed to `asm`.
+This is a RISC-V 64 assembly file, suitable for use with GNU Assembler.
+You can assemble this as normal, and run your program.
+
+## Testing
+
+The `tests` directory contains a shell script you can invoke with `sh test.sh`.
+It should run just fine given the above setup, provided you've done a `cargo build` to build the compiler.
+This will compile the test binary using `sorelc` and your local `riscv64` assembler.
+It will run the binary using binfmt, which instructs the kernel to run your binary using `qemu-user`.
+
+### Debugging
+
+You can use `gdb`!
+Unfortunately, with `qemu-user` we can't just load the binary directly into `gdb`.
+Instead, we'll need to run `qemu-user` on a RISC-V 64 binary with a debugger port open:
+
+```
+qemu-riscv64 -g 4567 ./test1.out
+```
+
+Then, you can fire up `gdb`, and once inside it's prompt, connect to the "remote" port.
+
+```
+target remote :4567
+```
+
+Then you can use normal gdb commands!
+The `tui layout asm` is highly recommended.
+Note that if you'l linking `putstack.c`, that will use libc, which drastically inflates the binary and hijack's init.
+This can make things quite a pain in `gdb`, so try to build without it (i.e. uncomment the LD line and comment out the CC line in `test.sh`, and remove all references to `putstack` in code).
--- /dev/null
+# Language Overview
+
+**Sorel** is a concatenative stack language, similar in form to Forth.
+Unlike most Forth variants, it's strictly a compiled language, with no interpreter provided at run time.
+Today, it targets only 64-bit RISC-V architectures.
+It has a small standard library, that does not use libc.
+Instead it makes system calls directly, currently only supporting Linux.
+
+Like in Forth, code is written in Reverse Polish Notation.
+Every "word" corresponds to a function that manipulates the stack in some way.
+Unlike in Forth, strings count as a single word, and are written much like they are in C.
+
+## The Stack
+
+All Forths have a stack, and Sorel is no exception.
+Since Sorel's function calls follow the C ABI, there's no need for the traditional Forth return stack.
+There's only the data stack.
+Much like call stacks, it grows downward.
+Each stack cell is exactly 64 bits wide, regardless of type.
+For smaller types, the values are padded with 0s.
+Note that RISC-V systems are little-endian.
+
+## Type System
+
+All values are treated as 64-bit unsigned integers unles otherwise indicated.
+Number literals can provide type information after a colon.
+For example, `-37:i32` will add -37 as a signed 32-bit integer onto the stack.
+
+Values themselves don't have types.
+Operations have types.
+In other words, values are always just bits in memory, and only have a "type" when they are treated as such by a word manipulating the stack or heap memory.
+
+> This is, of course, very dangerous.
+> A planned improvement is to add static type-checking support via type comments.
+
+Complex types like arrays and structs don't exist (yet), but they can be maniplulated through direct memory operations.
+See `std:mem` for details.
+
+String literals, when pushed onto the stack, appear as pointers to null-terminated UTF-8 representations of them.
+Other than that, strings are just like any other complex type, in that there's no special support (yet).
+
+## Module System
+
+Sorel supports import statements of the form `import "./relative/path/to/file.sorel"`, with the exception of standard library imports (see below).
+Note that `import` is _not_ executed at run-time, and instead is evaluated as a pre-processing step.
+To make that clear, note that the import specifier comes _after_ the keyword.
+This indicates that `import` is not operating on a value in the stack.
+
+Similarly, words may be exported via `export word_name`.
+Only words deliberately exported will be made available to other files via `import`.
+
+Word invocations _outside_ word defintions will be executed as the program's entrypoint function (i.e. "main"), but _only_ if those invocations are in the entrypoint module.
+This means any module can have word invocations inside it that aren't executed when it's not the entrypoint.
+This is useful for writing tests.
+
+## Built-in Words
+
+<!-- TODO -->
+
+## Standard Library
+
+Built-in words are insufficient to create most programs.
+A standard library is provided with exported words to enable higher level operations.
+It's implemented primarily by wrapping system calls.
+
+The following standard library modules are provided, and can be imported by specificying their name as the import specifier:
+* **`std:mem`** : This module contains words to allocate and free memory.
+* **`std:out`** : This module contains words to print numbers and strings to stdout.
--- /dev/null
+\ vim: filetype=forth
+
+: mmap 9 sys6 ;
+
+: PROT_READ 1 ;
+: PROT_WRITE 2 ;
+: MAP_PRIVATE 2 ;
+: MAP_ANONYMOUS 32 ;
+
+: ALLOC_PROT PROT_READ PROT_WRITE | ;
+: ALLOC_MAP MAP_PRIVATE MAP_ANONYMOUS | ;
+
+: alloc 0 swap ALLOC_PROT ALLOC_MAP -1:i16 0 mmap ;
+
+1024 alloc
+putn
+swap
+putn
--- /dev/null
+\ vim: filetype=forth
+
+import "./put2.rel"
+
+: fib
+ dup 1 > if
+ dup 1 - fib
+ swap 2 - fib
+ +
+ endif
+;
+
+0 fib putn
+1 fib putn
+2 fib putn
+3 fib putn
+4 fib putn
+5 fib putn
+6 fib putn
+7 fib putn
+8 fib putn
+9 fib putn
+10 fib putn
+
+5 fib 6 fib put2
--- /dev/null
+../target/debug/sorelc fib.sorel
+riscv64-unknown-linux-gnu-as -o fib.o fib.asm
+riscv64-unknown-linux-gnu-cc -O1 -no-pie -o test.out fib.o putn.c -nostartfiles
+./test.out
--- /dev/null
+\ foo bar
+: fib
+ dup
+ 1
+ >
+ if
+ dup
+ 1 - fib
+ swap
+ 2
+ -
+ fib
+ +
+ endif
+;
+
+0 fib putn drop
+1 fib putn drop
+2 fib putn drop
+3 fib putn drop
+4 fib putn drop
+5 fib putn drop
+6 fib putn drop
+7 fib putn drop
--- /dev/null
+#include <stdio.h>
+
+extern unsigned long data_stack_end;
+register unsigned long * stack_pointer asm("s2");
+
+void putn() {
+ unsigned long * stack_index = &data_stack_end;
+ printf("stack: ");
+ while (stack_index != stack_pointer) {
+ printf("%ld ", *stack_index);
+ stack_index -= 1;
+ }
+ printf("%ld\n", *stack_pointer);
+}
+
+
--- /dev/null
+\ vim: filetype=forth
+
+: put2 putn putn ;
+
+: foobar dup dup ;
+
+export put2
--- /dev/null
+\ vim: filetype=forth
+
+: getpid
+ 39 sys0
+ drop
+;
+
+getpid putn
--- /dev/null
+{
+ "nodes": {
+ "nixpkgs": {
+ "locked": {
+ "lastModified": 1765425892,
+ "narHash": "sha256-jlQpSkg2sK6IJVzTQBDyRxQZgKADC2HKMRfGCSgNMHo=",
+ "owner": "nixos",
+ "repo": "nixpkgs",
+ "rev": "5d6bdbddb4695a62f0d00a3620b37a15275a5093",
+ "type": "github"
+ },
+ "original": {
+ "owner": "nixos",
+ "ref": "nixpkgs-unstable",
+ "repo": "nixpkgs",
+ "type": "github"
+ }
+ },
+ "root": {
+ "inputs": {
+ "nixpkgs": "nixpkgs"
+ }
+ }
+ },
+ "root": "root",
+ "version": 7
+}
--- /dev/null
+{
+ description = "uxn11";
+
+ inputs = {
+ nixpkgs.url = "github:nixos/nixpkgs/nixpkgs-unstable";
+ };
+
+ outputs = {nixpkgs, ...}: let
+ system = "x86_64-linux";
+ pkgs = import nixpkgs {
+ inherit system;
+ };
+ in {
+ devShells.${system}.default = pkgs.mkShell {
+ packages = [
+ pkgs.pkgsCross.riscv64.buildPackages.gcc
+ pkgs.pkgsCross.riscv64.buildPackages.gdb
+ pkgs.gdb
+ pkgs.qemu
+ ];
+ };
+ };
+}
--- /dev/null
+
+qemu-system-riscv64 \
+ -machine virt \
+ -cpu rv64 \
+ -m 1G \
+ -device virtio-blk-device,drive=hd \
+ -drive file=machine/overlay.qcow2,if=none,id=hd \
+ -device virtio-net-device,netdev=net \
+ -netdev user,id=net,hostfwd=tcp::2222-:22 \
+ -kernel machine/u-boot-qemu/uboot.elf \
+ -object rng-random,filename=/dev/urandom,id=rng \
+ -device virtio-rng-device,rng=rng \
+ -append "root=LABEL=rootfs console=ttyS0" \
+ -nographic \
+
+echo "now do:"
+echo " ssh debian@localhost -p 2222"
--- /dev/null
+
+mkdir machine
+cd machine
+wget "https://gitlab.com/api/v4/projects/giomasce%2Fdqib/jobs/artifacts/master/download?job=convert_riscv64-virt" -O debian-rv64.zip
+
+unzip ./debian-rv64.zip
+
+# Grab the URL from https://packages.debian.org/sid/u-boot-qemu
+wget "http://ftp.us.debian.org/debian/pool/main/u/u-boot/u-boot-qemu_2025.01-3.1_all.deb" -O u-boot-qemu.deb
+mkdir u-boot-qemu
+cd u-boot-qemu
+ar -x ../u-boot-qemu.deb
+tar xvf data.tar.xz
+cp ./usr/lib/u-boot/qemu-riscv64_smode/uboot.elf ./uboot.elf
+
+cd ..
+qemu-img create -o backing_file=./dqib_riscv64-virt/image.qcow2,backing_fmt=qcow2 -f qcow2 overlay.qcow2
+
+++ /dev/null
-target
-qemu/machine
-*.o
-*.asm
-*.out
+++ /dev/null
-# This file is automatically @generated by Cargo.
-# It is not intended for manual editing.
-version = 4
-
-[[package]]
-name = "anyhow"
-version = "1.0.100"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a23eb6b1614318a8071c9b2521f36b424b2c83db5eb3a0fead4a6c0809af6e61"
-
-[[package]]
-name = "equivalent"
-version = "1.0.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "877a4ace8713b0bcf2a4e7eec82529c029f1d0619886d18145fea96c3ffe5c0f"
-
-[[package]]
-name = "hashbrown"
-version = "0.16.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "841d1cc9bed7f9236f321df977030373f4a4163ae1a7dbfe1a51a2c1a51d9100"
-
-[[package]]
-name = "indexmap"
-version = "2.12.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0ad4bb2b565bca0645f4d68c5c9af97fba094e9791da685bf83cb5f3ce74acf2"
-dependencies = [
- "equivalent",
- "hashbrown",
-]
-
-[[package]]
-name = "itoa"
-version = "1.0.15"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4a5f13b858c8d314ee3e8f639011f7ccefe71f97f96e50151fb991f267928e2c"
-
-[[package]]
-name = "proc-macro2"
-version = "1.0.103"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5ee95bc4ef87b8d5ba32e8b7714ccc834865276eab0aed5c9958d00ec45f49e8"
-dependencies = [
- "unicode-ident",
-]
-
-[[package]]
-name = "quote"
-version = "1.0.42"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a338cc41d27e6cc6dce6cefc13a0729dfbb81c262b1f519331575dd80ef3067f"
-dependencies = [
- "proc-macro2",
-]
-
-[[package]]
-name = "ryu"
-version = "1.0.20"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "28d3b2b1366ec20994f1fd18c3c594f05c5dd4bc44d8bb0c1c632c8d6829481f"
-
-[[package]]
-name = "serde"
-version = "1.0.228"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9a8e94ea7f378bd32cbbd37198a4a91436180c5bb472411e48b5ec2e2124ae9e"
-dependencies = [
- "serde_core",
-]
-
-[[package]]
-name = "serde_core"
-version = "1.0.228"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "41d385c7d4ca58e59fc732af25c3983b67ac852c1a25000afe1175de458b67ad"
-dependencies = [
- "serde_derive",
-]
-
-[[package]]
-name = "serde_derive"
-version = "1.0.228"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d540f220d3187173da220f885ab66608367b6574e925011a9353e4badda91d79"
-dependencies = [
- "proc-macro2",
- "quote",
- "syn",
-]
-
-[[package]]
-name = "serde_yaml"
-version = "0.9.34+deprecated"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6a8b1a1a2ebf674015cc02edccce75287f1a0130d394307b36743c2f5d504b47"
-dependencies = [
- "indexmap",
- "itoa",
- "ryu",
- "serde",
- "unsafe-libyaml",
-]
-
-[[package]]
-name = "sorel-ir"
-version = "0.1.0"
-dependencies = [
- "serde",
- "serde_derive",
- "serde_yaml",
-]
-
-[[package]]
-name = "sorelc"
-version = "0.1.0"
-dependencies = [
- "anyhow",
- "sorel-ir",
-]
-
-[[package]]
-name = "syn"
-version = "2.0.111"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "390cc9a294ab71bdb1aa2e99d13be9c753cd2d7bd6560c77118597410c4d2e87"
-dependencies = [
- "proc-macro2",
- "quote",
- "unicode-ident",
-]
-
-[[package]]
-name = "unicode-ident"
-version = "1.0.22"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9312f7c4f6ff9069b165498234ce8be658059c6728633667c526e27dc2cf1df5"
-
-[[package]]
-name = "unsafe-libyaml"
-version = "0.2.11"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "673aac59facbab8a9007c7f6108d11f63b603f7cabff99fabf650fea5c32b861"
+++ /dev/null
-[workspace]
-
-resolver = "3"
-members = ["sorel-ir","sorelc"]
-
-
-[workspace.dependencies]
-sorel-ir = { path = "./sorel-ir", version = "0.1.0" }
+++ /dev/null
-# sorel
-
-The name means "Stack-Oriented Rethought Language".
-
-
-## TODO
-
-* [x] Imports
-* [x] Syscalls
-* [x] Loops
-* [ ] Structs
-* [ ] many, many more things
-
-## Helpful External Resources
-
-* https://gpages.juszkiewicz.com.pl/syscalls-table/syscalls.html
-* https://godbolt.org/
-* https://projectf.io/posts/riscv-cheat-sheet/
-* https://github.com/dvoytik/riscv-cheats
+++ /dev/null
-## Developer setup
-
-### Nix / NixOS
-
-If you're using `Nix` or `NixOS`, the provided `flake.nix` should give you a good start.
-You'll need a recent Rust toolchain, for which you can use [`rustup`](https://rustup.sh).
-Currently that's assumed to be included on the system.
-Also, the test scripts currently assume binfmt is set up for RISC-V 64 (or that you're running on RISC-V 64).
-Assuming you're not on RISC-V 64 (most people aren't), you can add the following to your Nix `configuration.nix`:
-
-```
-
- boot.binfmt.emulatedSystems = [
- "riscv64-linux"
- # ....
- ];
-```
-
-### Ubuntu
-
-```
-# On non RISC-V 64 systems:
-apt install -y build-essential gdb crossbuild-essential-riscv64 qemu-system qemu-user qemu-user-binfmt
-```
-
-## Building the Compiler
-
-The compiler is written in Rust, and can be compiled with
-
-```
-cargo build
-```
-
-This puts the compiler binary at `target/debug/sorelc`.
-Release builds can also be created in the usual Rust/`cargo` fasion.
-
-Note that if you change any stdlib files, you'll need to rebuild.
-
-## Compiling a Sorel Program
-
-Currently, `sorelc` takes only one argument, and that's an entrypoint for your program.
-Any words in that file outside word definitions will constitute the "main" of your program.
-
-This will create a file of the same name alongside it, with the extension changed to `asm`.
-This is a RISC-V 64 assembly file, suitable for use with GNU Assembler.
-You can assemble this as normal, and run your program.
-
-## Testing
-
-The `tests` directory contains a shell script you can invoke with `sh test.sh`.
-It should run just fine given the above setup, provided you've done a `cargo build` to build the compiler.
-This will compile the test binary using `sorelc` and your local `riscv64` assembler.
-It will run the binary using binfmt, which instructs the kernel to run your binary using `qemu-user`.
-
-### Debugging
-
-You can use `gdb`!
-Unfortunately, with `qemu-user` we can't just load the binary directly into `gdb`.
-Instead, we'll need to run `qemu-user` on a RISC-V 64 binary with a debugger port open:
-
-```
-qemu-riscv64 -g 4567 ./test1.out
-```
-
-Then, you can fire up `gdb`, and once inside it's prompt, connect to the "remote" port.
-
-```
-target remote :4567
-```
-
-Then you can use normal gdb commands!
-The `tui layout asm` is highly recommended.
-Note that if you'l linking `putstack.c`, that will use libc, which drastically inflates the binary and hijack's init.
-This can make things quite a pain in `gdb`, so try to build without it (i.e. uncomment the LD line and comment out the CC line in `test.sh`, and remove all references to `putstack` in code).
+++ /dev/null
-# Language Overview
-
-**Sorel** is a concatenative stack language, similar in form to Forth.
-Unlike most Forth variants, it's strictly a compiled language, with no interpreter provided at run time.
-Today, it targets only 64-bit RISC-V architectures.
-It has a small standard library, that does not use libc.
-Instead it makes system calls directly, currently only supporting Linux.
-
-Like in Forth, code is written in Reverse Polish Notation.
-Every "word" corresponds to a function that manipulates the stack in some way.
-Unlike in Forth, strings count as a single word, and are written much like they are in C.
-
-## The Stack
-
-All Forths have a stack, and Sorel is no exception.
-Since Sorel's function calls follow the C ABI, there's no need for the traditional Forth return stack.
-There's only the data stack.
-Much like call stacks, it grows downward.
-Each stack cell is exactly 64 bits wide, regardless of type.
-For smaller types, the values are padded with 0s.
-Note that RISC-V systems are little-endian.
-
-## Type System
-
-All values are treated as 64-bit unsigned integers unles otherwise indicated.
-Number literals can provide type information after a colon.
-For example, `-37:i32` will add -37 as a signed 32-bit integer onto the stack.
-
-Values themselves don't have types.
-Operations have types.
-In other words, values are always just bits in memory, and only have a "type" when they are treated as such by a word manipulating the stack or heap memory.
-
-> This is, of course, very dangerous.
-> A planned improvement is to add static type-checking support via type comments.
-
-Complex types like arrays and structs don't exist (yet), but they can be maniplulated through direct memory operations.
-See `std:mem` for details.
-
-String literals, when pushed onto the stack, appear as pointers to null-terminated UTF-8 representations of them.
-Other than that, strings are just like any other complex type, in that there's no special support (yet).
-
-## Module System
-
-Sorel supports import statements of the form `import "./relative/path/to/file.sorel"`, with the exception of standard library imports (see below).
-Note that `import` is _not_ executed at run-time, and instead is evaluated as a pre-processing step.
-To make that clear, note that the import specifier comes _after_ the keyword.
-This indicates that `import` is not operating on a value in the stack.
-
-Similarly, words may be exported via `export word_name`.
-Only words deliberately exported will be made available to other files via `import`.
-
-Word invocations _outside_ word defintions will be executed as the program's entrypoint function (i.e. "main"), but _only_ if those invocations are in the entrypoint module.
-This means any module can have word invocations inside it that aren't executed when it's not the entrypoint.
-This is useful for writing tests.
-
-## Built-in Words
-
-<!-- TODO -->
-
-## Standard Library
-
-Built-in words are insufficient to create most programs.
-A standard library is provided with exported words to enable higher level operations.
-It's implemented primarily by wrapping system calls.
-
-The following standard library modules are provided, and can be imported by specificying their name as the import specifier:
-* **`std:mem`** : This module contains words to allocate and free memory.
-* **`std:out`** : This module contains words to print numbers and strings to stdout.
+++ /dev/null
-\ vim: filetype=forth
-
-: mmap 9 sys6 ;
-
-: PROT_READ 1 ;
-: PROT_WRITE 2 ;
-: MAP_PRIVATE 2 ;
-: MAP_ANONYMOUS 32 ;
-
-: ALLOC_PROT PROT_READ PROT_WRITE | ;
-: ALLOC_MAP MAP_PRIVATE MAP_ANONYMOUS | ;
-
-: alloc 0 swap ALLOC_PROT ALLOC_MAP -1:i16 0 mmap ;
-
-1024 alloc
-putn
-swap
-putn
+++ /dev/null
-\ vim: filetype=forth
-
-import "./put2.rel"
-
-: fib
- dup 1 > if
- dup 1 - fib
- swap 2 - fib
- +
- endif
-;
-
-0 fib putn
-1 fib putn
-2 fib putn
-3 fib putn
-4 fib putn
-5 fib putn
-6 fib putn
-7 fib putn
-8 fib putn
-9 fib putn
-10 fib putn
-
-5 fib 6 fib put2
+++ /dev/null
-../target/debug/sorelc fib.sorel
-riscv64-unknown-linux-gnu-as -o fib.o fib.asm
-riscv64-unknown-linux-gnu-cc -O1 -no-pie -o test.out fib.o putn.c -nostartfiles
-./test.out
+++ /dev/null
-\ foo bar
-: fib
- dup
- 1
- >
- if
- dup
- 1 - fib
- swap
- 2
- -
- fib
- +
- endif
-;
-
-0 fib putn drop
-1 fib putn drop
-2 fib putn drop
-3 fib putn drop
-4 fib putn drop
-5 fib putn drop
-6 fib putn drop
-7 fib putn drop
+++ /dev/null
-#include <stdio.h>
-
-extern unsigned long data_stack_end;
-register unsigned long * stack_pointer asm("s2");
-
-void putn() {
- unsigned long * stack_index = &data_stack_end;
- printf("stack: ");
- while (stack_index != stack_pointer) {
- printf("%ld ", *stack_index);
- stack_index -= 1;
- }
- printf("%ld\n", *stack_pointer);
-}
-
-
+++ /dev/null
-\ vim: filetype=forth
-
-: put2 putn putn ;
-
-: foobar dup dup ;
-
-export put2
+++ /dev/null
-\ vim: filetype=forth
-
-: getpid
- 39 sys0
- drop
-;
-
-getpid putn
+++ /dev/null
-{
- "nodes": {
- "nixpkgs": {
- "locked": {
- "lastModified": 1765425892,
- "narHash": "sha256-jlQpSkg2sK6IJVzTQBDyRxQZgKADC2HKMRfGCSgNMHo=",
- "owner": "nixos",
- "repo": "nixpkgs",
- "rev": "5d6bdbddb4695a62f0d00a3620b37a15275a5093",
- "type": "github"
- },
- "original": {
- "owner": "nixos",
- "ref": "nixpkgs-unstable",
- "repo": "nixpkgs",
- "type": "github"
- }
- },
- "root": {
- "inputs": {
- "nixpkgs": "nixpkgs"
- }
- }
- },
- "root": "root",
- "version": 7
-}
+++ /dev/null
-{
- description = "uxn11";
-
- inputs = {
- nixpkgs.url = "github:nixos/nixpkgs/nixpkgs-unstable";
- };
-
- outputs = {nixpkgs, ...}: let
- system = "x86_64-linux";
- pkgs = import nixpkgs {
- inherit system;
- };
- in {
- devShells.${system}.default = pkgs.mkShell {
- packages = [
- pkgs.pkgsCross.riscv64.buildPackages.gcc
- pkgs.pkgsCross.riscv64.buildPackages.gdb
- pkgs.gdb
- pkgs.qemu
- ];
- };
- };
-}
+++ /dev/null
-
-qemu-system-riscv64 \
- -machine virt \
- -cpu rv64 \
- -m 1G \
- -device virtio-blk-device,drive=hd \
- -drive file=machine/overlay.qcow2,if=none,id=hd \
- -device virtio-net-device,netdev=net \
- -netdev user,id=net,hostfwd=tcp::2222-:22 \
- -kernel machine/u-boot-qemu/uboot.elf \
- -object rng-random,filename=/dev/urandom,id=rng \
- -device virtio-rng-device,rng=rng \
- -append "root=LABEL=rootfs console=ttyS0" \
- -nographic \
-
-echo "now do:"
-echo " ssh debian@localhost -p 2222"
+++ /dev/null
-
-mkdir machine
-cd machine
-wget "https://gitlab.com/api/v4/projects/giomasce%2Fdqib/jobs/artifacts/master/download?job=convert_riscv64-virt" -O debian-rv64.zip
-
-unzip ./debian-rv64.zip
-
-# Grab the URL from https://packages.debian.org/sid/u-boot-qemu
-wget "http://ftp.us.debian.org/debian/pool/main/u/u-boot/u-boot-qemu_2025.01-3.1_all.deb" -O u-boot-qemu.deb
-mkdir u-boot-qemu
-cd u-boot-qemu
-ar -x ../u-boot-qemu.deb
-tar xvf data.tar.xz
-cp ./usr/lib/u-boot/qemu-riscv64_smode/uboot.elf ./uboot.elf
-
-cd ..
-qemu-img create -o backing_file=./dqib_riscv64-virt/image.qcow2,backing_fmt=qcow2 -f qcow2 overlay.qcow2
-
+++ /dev/null
-[package]
-name = "sorel-ir"
-version = "0.1.0"
-edition = "2024"
-
-[dependencies]
-serde = "1.0.228"
-serde_derive = "1.0.228"
-serde_yaml = "0.9.34"
+++ /dev/null
-use serde_yaml::{from_str, to_string, Error};
-use serde_derive::{Serialize, Deserialize};
-
-#[derive(Serialize, Deserialize, Debug, Clone)]
-pub enum IR {
- Label(String),
- Call(String),
- Ret,
- StackPush(u64),
- StackPushString(String), // refers to string label, not the string itself
- StringDef(String, String), // first is string label, second is string value
-
- // These next ones should always be inlined, so they're in IR.
- Load, // @ ( addr -- x ) -- Fetch memory contents at addr
- Load8,
- Load16,
- Load32,
- Store, // ! ( x addr -- ) -- Store x at addr
- Store8,
- Store16,
- Store32,
-
- // These ones might not be inlined, but should be built-in, so a compiler might
- // turn this into `Call(String)` before translating to assembly/machine-code, but
- // an IR interpreter may just execute them.
- AddU64,
- SubtractU64,
- MultiplyU64,
- DivideU64,
- ModU64,
- Equals,
- GreaterThan,
- LessThan,
- BitwiseOr,
- Dup,
- Swap,
- Drop,
- Over,
- Rot,
- StackPointer,
- If,
- Else,
- EndIf,
- Loop,
- EndLoop,
-
- // System calls
- Sys0,
- Sys1,
- Sys2,
- Sys3,
- Sys4,
- Sys5,
- Sys6,
-}
-
-// This is like an .o file.
-#[derive(Serialize, Deserialize, Debug)]
-pub struct IRObject {
- pub text: Vec<IR>,
- pub data: Vec<IR>,
-}
-
-impl IRObject {
- pub fn to_s(&self) -> Result<String, Error> {
- to_string(self)
- }
-
- pub fn from_s(source: &str) -> Result<Self, Error> {
- from_str(source)
- }
-}
+++ /dev/null
-[package]
-name = "sorelc"
-version = "0.1.0"
-edition = "2024"
-
-[dependencies]
-sorel-ir = { workspace = true }
-anyhow = "1.0.100"
+++ /dev/null
-use crate::parser::Module;
-use crate::tokenizer::{Token, tokenize};
-use sorel_ir::*;
-
-use std::collections::{HashSet, HashMap};
-use std::path::PathBuf;
-use std::rc::Rc;
-use std::include_str;
-
-use anyhow::{Result, bail};
-
-macro_rules! push_num {
- ($num:ident) => { IR::StackPush(*$num as u64) };
- ($num:ident, $num_typ:ty) => { IR::StackPush(*$num as $num_typ as u64) };
-}
-
-#[derive(Debug, Default)]
-struct IRModule {
- data: Vec<IR>,
- text: Vec<IR>,
- imports: Vec<Rc<IRModule>>,
- exports: Vec<String>,
- externs: Vec<String>,
- // TODO these next two should form an enum, not two options
- source_file: Option<PathBuf>,
- std_specifier: Option<String>,
- number: usize,
-}
-
-impl IRModule {
- fn get_label_for_call(&self, name: &String) -> String {
- if self.externs.contains(name) {
- return name.clone();
- }
- let mut found: Option<usize> = None;
- for imported in &self.imports {
- if imported.exports.contains(name) {
- found = Some(imported.number);
- // Don't break here, since the last one should win.
- }
- }
- if let Some(found) = found {
- format!("_m{}_{}", found, name)
- } else {
- // TODO check if it's even a word locally. If not, bail.
- format!("_m{}_{}", self.number, name)
- }
- }
-
- fn get_label(&self, name: &String) -> String {
- format!("_m{}_{}", self.number, name)
- }
-}
-
-#[derive(Default)]
-struct ImportTree {
- data: Vec<IR>,
- text: Vec<IR>,
- all_modules: HashMap<String, Rc<IRModule>>,
- all_exports: HashSet<String>,
- entrypoint: Rc<IRModule>,
- module_count: usize,
- collapse_seen: HashSet<String>,
-}
-
-fn std_import(specifier: &str) -> Result<&str> {
- match specifier {
- "std:mem" => Ok(include_str!("../../stdlib/mem.sorel")),
- "std:out" => Ok(include_str!("../../stdlib/out.sorel")),
- "std:string" => Ok(include_str!("../../stdlib/string.sorel")),
- "std:process" => Ok(include_str!("../../stdlib/process.sorel")),
- _ => bail!("{} is not a standard library module", specifier),
- }
-}
-
-impl ImportTree {
- fn import(&mut self, importer_dir: &PathBuf, specifier: &str, is_entrypoint: bool) -> Result<Rc<IRModule>> {
- Ok(if specifier.starts_with("std:") {
- if self.all_modules.contains_key(specifier) {
- let module = self.all_modules.get(specifier).unwrap().clone();
- return Ok(module);
- }
- let contents = std_import(specifier)?;
- let parsed = &Module::parse(tokenize(&contents)?, is_entrypoint)?;
- let module = self.generate_internal(None, Some(specifier.to_string()), parsed);
- let module = Rc::new(module);
- self.all_modules.insert(specifier.to_string(), module.clone());
- module
- } else {
- let mut path = PathBuf::from(specifier);
- if path.is_relative() {
- let mut new_path = importer_dir.clone();
- new_path.push(path);
- path = new_path.canonicalize()?;
- }
- let path_key = path.to_string_lossy().to_string();
- if self.all_modules.contains_key(&path_key) {
- let module = self.all_modules.get(&path_key).unwrap().clone();
- return Ok(module);
- }
-
- let contents = &std::fs::read_to_string(&path)?;
- let tokens = tokenize(&contents)?;
- let parsed = &Module::parse(tokens, is_entrypoint)?;
-
- let module = self.generate_internal(Some(path), None, parsed);
- let module = Rc::new(module);
- self.all_modules.insert(path_key, module.clone());
- if is_entrypoint {
- self.entrypoint = module.clone();
- }
- module
- })
-
- }
-
- fn generate_internal(&mut self, path: Option<PathBuf>, std_specifier: Option<String>, module: &Module) -> IRModule {
- // Eventually these will end up being sections in assembly
- let mut text = vec![];
- let mut data = vec![];
-
- let mut imports = vec![];
- if let Some(ref path) = path {
- module.imports.iter().for_each(|imported| {
- if let Some(parent_path) = path.parent() {
- match self.import(&parent_path.to_path_buf(), imported, false) {
- Ok(module) => {
- imports.push(module);
- },
- Err(msg) => {
- eprintln!("{}", msg);
- }
- }
- } else {
- }
- });
- } else {
- // We're in a stdlib module, which can only import other stdlib
- // modules.
- module.imports.iter().for_each(|imported| {
- match self.import(&PathBuf::new(), imported, false) {
- Ok(module) => {
- imports.push(module);
- },
- Err(msg) => {
- eprintln!("{}", msg);
- }
- }
- });
- }
-
- let exports: Vec<_> = module.exports.iter().map(|s| {
- self.all_exports.insert(s.to_string());
- s.to_string()
- }).collect();
-
- let externs = module.externs.iter().map(|s| s.to_string()).collect();
-
- text.push(module.words.iter().map(|def| {
- let mut body = def.instructions.iter().map(|inst| {
- let mapped_ir = match inst {
- Token::Word(word) => {
- match *word {
- "@" => IR::Load,
- "@:8" => IR::Load8,
- "@:16" => IR::Load16,
- "@:32" => IR::Load32,
- "!" => IR::Store,
- "!:8" => IR::Store8,
- "!:16" => IR::Store16,
- "!:32" => IR::Store32,
- "dup" => IR::Dup,
- "swap" => IR::Swap,
- "drop" => IR::Drop,
- "over" => IR::Over,
- "rot" => IR::Rot,
- "sp" => IR::StackPointer,
- "if" => IR::If,
- "else" => IR::Else,
- "endif" => IR::EndIf,
- "loop" => IR::Loop,
- "endloop" => IR::EndLoop,
- "=" => IR::Equals,
- ">" => IR::GreaterThan,
- "<" => IR::LessThan,
- "+" => IR::AddU64,
- "-" => IR::SubtractU64,
- "*" => IR::MultiplyU64,
- "/" => IR::DivideU64,
- "%" => IR::ModU64,
- "|" => IR::BitwiseOr,
- "sys0" => IR::Sys0,
- "sys1" => IR::Sys1,
- "sys2" => IR::Sys2,
- "sys3" => IR::Sys3,
- "sys4" => IR::Sys4,
- "sys5" => IR::Sys5,
- "sys6" => IR::Sys6,
- // TODO num type specfic math like `+:i32`, etc.
- _ => IR::Call(String::from(*word))
- }
- },
- Token::String(text) => {
- let string_label = format!("string_{}", data.len());
- data.push(IR::StringDef(string_label.clone(), String::from(*text)));
- IR::StackPushString(string_label)
- },
- Token::NumU8(num) => push_num!(num),
- Token::NumI8(num) => push_num!(num, u8),
- Token::NumU16(num) => push_num!(num),
- Token::NumI16(num) => push_num!(num, u16),
- Token::NumU32(num) => push_num!(num),
- Token::NumI32(num) => push_num!(num, u32),
- Token::NumU64(num) => push_num!(num),
- Token::NumI64(num) => push_num!(num),
- Token::NumF32(num) => push_num!(num),
- Token::NumF64(num) => push_num!(num),
- };
- mapped_ir
- }).collect::<Vec<_>>();
-
- let mut result = vec![IR::Label(def.name.to_string())];
- result.append(&mut body);
- result.push(IR::Ret);
- result
- }).flatten().collect::<Vec<_>>());
-
- let number = self.module_count;
- self.module_count += 1;
-
- IRModule {
- text: text.into_iter().flatten().collect::<Vec<_>>(),
- data,
- imports,
- exports,
- externs,
- source_file: path,
- std_specifier,
- number,
- }
- }
-
- fn collapse(&mut self, module: Rc<IRModule>) -> Result<()> {
- let seen_key = if let Some(source_file) = &module.source_file {
- source_file.to_string_lossy().to_string()
- } else {
- module.std_specifier.clone().unwrap()
- };
- if self.collapse_seen.contains(&seen_key) {
- return Ok(())
- }
-
- for imported in module.imports.clone() {
- self.collapse(imported)?;
- }
-
- let is_entrypoint = module.source_file == self.entrypoint.source_file;
-
- let module_number = module.number;
-
- for string in &module.data {
- if let IR::StringDef(name, val) = string {
- let new_name = format!("{}_{}", name, module_number);
- self.data.push(IR::StringDef(new_name, val.clone()));
- } else {
- bail!("non-string data");
- }
- }
-
- for instruction in &module.text {
- let new_instruction = match instruction {
- IR::StackPushString(name) => {
- let new_name = format!("{}_{}", name, module_number);
- IR::StackPushString(new_name)
- },
- IR::Label(name) => {
- if is_entrypoint && name == "main" {
- instruction.clone()
- } else {
- IR::Label(module.get_label(name))
- }
- },
- IR::Call(name) => {
- IR::Call(module.get_label_for_call(name))
- },
- _ => instruction.clone()
- };
- self.text.push(new_instruction);
- }
-
- self.collapse_seen.insert(seen_key);
-
- Ok(())
- }
-}
-
-pub fn compile(path: &str) -> Result<IRObject> {
- let dir = std::env::current_dir()?;
- let mut tree: ImportTree = Default::default();
- let module = tree.import(&dir, path, true)?;
- tree.collapse(module)?;
- // TODO remove unused words
- Ok(IRObject {
- data: tree.data,
- text: tree.text,
- })
-}
+++ /dev/null
-mod tokenizer;
-mod parser;
-mod ir;
-mod riscv_asm_codegen;
-
-use anyhow::Result;
-
-use std::fs::File;
-use std::io::Write;
-use std::path::PathBuf;
-
-fn main() -> Result<()> {
- let filename = std::env::args().nth(1).expect("must provide a file to compile");
- let module = ir::compile(&filename)?;
- let mut generator = riscv_asm_codegen::CodeGen::new(&module, 4096);
- let mut asm_path = PathBuf::from(filename);
- asm_path.set_extension("asm");
- let mut output = File::create(asm_path)?;
- write!(output, "{}\n", generator.assembly()?)?;
- Ok(())
-}
+++ /dev/null
-use crate::tokenizer::Token;
-use anyhow::{Result, bail};
-
-#[derive(Debug)]
-pub struct WordDefinition<'a> {
- pub name: &'a str,
- pub instructions: Vec<Token<'a>>,
-}
-
-#[derive(Debug)]
-pub struct Module<'a> {
- pub words: Vec<WordDefinition<'a>>,
- pub imports: Vec<&'a str>,
- pub exports: Vec<&'a str>,
- pub externs: Vec<&'a str>,
-}
-
-impl<'a> Module<'a> {
- pub fn parse(input: Vec<Token<'a>>, is_entrypoint: bool) -> Result<Self> {
- let mut result = vec![];
- let mut main = vec![];
- let mut exports = vec![];
- let mut imports = vec![];
- let mut externs = vec![];
- let mut current_word: Option<WordDefinition> = None;
- let mut about_to_start_word_def = false;
- let mut last_was_import = false;
- let mut last_was_export = false;
- let mut last_was_extern = false;
-
- for token in input {
- if about_to_start_word_def {
- if let Token::Word(name) = token {
- current_word = Some(WordDefinition {
- name,
- instructions: vec![],
- });
- about_to_start_word_def = false;
- continue;
- } else {
- bail!("{:?} is not a valid word name!", token);
- }
- } else if let Token::Word(word) = token {
- if word == ":" {
- if current_word.is_some() {
- bail!("can't define words inside word definitions!");
- }
- about_to_start_word_def = true;
- continue;
- }
- if word == ";" {
- let word = current_word.take();
- if let Some(word) = word {
- result.push(word);
- continue;
- } else {
- bail!("`;` must be at the end of a word definition");
- }
- }
- }
- if let Some(ref mut current_word) = current_word {
- current_word.instructions.push(token);
- } else {
- match token {
- Token::Word(word) => {
- if word == "import" {
- last_was_import = true;
- } else if word == "export" {
- last_was_export = true;
- } else if word == "extern" {
- last_was_extern = true;
- } else {
- if last_was_export {
- exports.push(word);
- last_was_export = false;
- } else if last_was_extern {
- externs.push(word);
- last_was_extern = false;
- } else {
- main.push(token.clone());
- }
- }
- },
- Token::String(string) => {
- if last_was_import {
- imports.push(string);
- last_was_import = false;
- } else {
- main.push(token.clone());
- }
- },
- _ => {
- main.push(token.clone());
- }
- };
- }
- }
-
- if about_to_start_word_def || current_word.is_some() {
- bail!("unfinished word definition!");
- }
-
- if is_entrypoint {
- result.push(WordDefinition {
- name: "main",
- instructions: main,
- });
- }
-
- Ok(Module { words: result, imports, exports, externs })
- }
-
- #[cfg(test)]
- pub fn debug_print(&self) {
- for word in &self.words {
- println!("{}", word.name);
- for instruction in &word.instructions {
- println!(" {:?}", instruction);
- }
- }
- }
-}
-
-
-
-#[cfg(test)]
-mod tests {
- use super::*;
-
- #[test]
- fn try_some_parsing() {
- let result = Module::parse(crate::tokenizer::tokenize("
-: hello world 16 \"planet\" ;
-: soup chicken 4.5 hello ;
-
-hello soup
-").unwrap(), true).unwrap();
- result.debug_print();
- }
-}
+++ /dev/null
-use sorel_ir::*;
-
-use anyhow::*;
-
-use std::collections::{HashMap, HashSet};
-use std::fmt::Display;
-
-pub struct CodeGen<'a> {
- module: &'a IRObject,
- data_stack_size: usize,
- lines: Vec<String>,
-}
-
-
-// Some inspiration
-// ================
-//
-// * https://github.com/aw/fiveforths/blob/master/docs/REFERENCE.md#registers-list
-// * Except using sp as a more C ABI style stack pointer, and s2 for the data stack
-//
-
-// Implementation Choices
-// ======================
-//
-// Data Stack pointer: s2
-// No return stack pointer (using C ABI, so sp, sorta)
-// Use t0, t1, t2 for temporary values in words
-// Data stack grows down
-
-
-macro_rules! asm_macro {
- ($name:ident, $src:expr) => {
- fn $name(&mut self) {
- self.line($src);
- }
- };
- ($name:ident, $src:expr, $arg0:ty) => {
- fn $name(&mut self, val0: $arg0) {
- self.line(format!($src, val0));
- }
- };
- ($name:ident, $src:expr, $arg0:ty, $arg1:ty) => {
- fn $name(&mut self, val0: $arg0, val1: $arg1) {
- self.line(format!($src, val0, val1));
- }
- };
-}
-
-fn mangle(input: &str) -> String {
- input
- .replace("<", "_LT_")
- .replace(">", "_GT_")
- .replace("-", "___")
-}
-
-impl<'a> CodeGen<'a> {
- pub fn new(ir_mod: &'a IRObject, data_stack_size: usize) -> Self {
- Self {
- module: ir_mod,
- data_stack_size,
- lines: vec![],
- }
- }
-
- fn line<S: Display>(&mut self, line: S) {
- self.lines.push(format!(" {}", line));
-
- }
-
- fn label<S: Display>(&mut self, line: S) {
- self.lines.push(line.to_string());
- }
-
- asm_macro!(copy_top_stack_value_to, "ld {}, 0(s2)", &str);
- asm_macro!(copy_offset_stack_value_to, "ld {}, {}*8(s2)", &str, isize);
- asm_macro!(copy_to_top_of_stack, "sd {}, 0(s2)", &str);
- asm_macro!(move_stack_ptr_by_cells, "addi s2, s2, {}*8", isize);
-
- fn pop_to(&mut self, reg: &str) {
- self.copy_top_stack_value_to(reg);
- self.move_stack_ptr_by_cells(1);
- }
-
- fn pop_some_to(&mut self, regs: &str) {
- let mut regs = regs.trim().split(" ").collect::<Vec<_>>();
- regs.reverse();
- let count = regs.len();
- let mut index = 0;
- for reg in regs {
- self.copy_offset_stack_value_to(reg, index);
- index += 1;
- }
- self.move_stack_ptr_by_cells(count as isize);
- }
-
- fn push_from(&mut self, reg: &str) {
- self.move_stack_ptr_by_cells(-1);
- self.copy_to_top_of_stack(reg);
- }
-
- fn pop_call_push(&mut self, regs: &str, call: &str, reg: &str) {
- self.pop_some_to(regs);
- self.line(call);
- self.push_from(reg);
- }
-
- pub fn assembly(&mut self) -> Result<String>{
- let mut string_table = HashMap::new();
-
- // Static strings
- self.label(".section .rodata\n");
- self.label(".align 3\n");
- for ir in &self.module.data {
- match ir {
- IR::StringDef(string_label, some_string) => {
- string_table.insert(some_string.clone(), string_label);
- self.label(format!("{}:", string_label));
- self.line(format!(".asciz \"{}\"", some_string)); // should this be .asciz?
- self.label("");
- },
- _ => bail!("Currently only string definitions are supported in the data section.")
- }
- }
-
- // Data stack
- self.label(".data\n");
- self.label(".align 3\n");
- self.label("data_stack:");
- self.line(format!(".space {}", self.data_stack_size));
- self.label(".globl data_stack_end\ndata_stack_end:\n");
-
- // Code
- self.label(".text\n");
- self.label(".align 3\n");
-
- let mut if_block_count = 0;
- let mut if_stack = vec![];
- let mut loop_count = 0;
- let mut loop_stack = vec![];
- let mut seen_else = HashSet::new();
- let mut last_label = "";
-
- for ir in &self.module.text {
- match ir {
- IR::Label(name) => {
- last_label = name;
- if name == "main" {
- self.label(".globl _start"); // TODO is globl necessary?
- self.label("_start:");
- self.line("la s2, data_stack_end # set initial data stack pointer");
- } else {
- let mangled = mangle(name);
- self.label(format!(".globl {}", mangled));
- self.label(format!("{}:", mangled));
- }
- self.line("addi sp, sp, -16 # allocate 16 bytes on stack"); // allocate 16 bytes on stack
- self.line("sd ra, 8(sp) # store return address on stack"); // store return address on stack
- },
- IR::Call(name) => {
- let mangled = mangle(name);
- self.label(format!("# call {}", mangled));
- self.line(format!("call {}", mangled));
- },
- IR::Ret => {
- if last_label == "main" {
- self.label("# exit 0 syscall");
- self.line("li a7, 93");
- self.line("mv a0, x0");
- self.line("ecall");
- } else {
- self.line("ld ra, 8(sp)"); // load return address from stack
- self.line("addi sp, sp, 16"); // restore stack pointer
- self.line("ret");
- }
- },
- IR::Load8 => {
- self.label("# load 8");
- self.copy_top_stack_value_to("t0");
- self.line("lbu t0, 0(t0)"); // deref pointer in t0 to t0
- self.copy_to_top_of_stack("t0");
- },
- IR::Load16 => {
- self.label("# load 16");
- self.copy_top_stack_value_to("t0");
- self.line("lhu t0, 0(t0)"); // deref pointer in t0 to t0
- self.copy_to_top_of_stack("t0");
- },
- IR::Load32 => {
- self.label("# load 32");
- self.copy_top_stack_value_to("t0");
- self.line("lwu t0, 0(t0)"); // deref pointer in t0 to t0
- self.copy_to_top_of_stack("t0");
- },
- IR::Load => {
- self.label("# load 64");
- self.copy_top_stack_value_to("t0");
- self.line("ld t0, 0(t0)"); // deref pointer in t0 to t0
- self.copy_to_top_of_stack("t0");
- },
- IR::Store8 => { // ( x addr -- )
- self.pop_some_to("t0 t1");
- self.line("sb t0, 0(t1)"); // store x at addr
- },
- IR::Store16 => { // ( x addr -- )
- self.pop_some_to("t0 t1");
- self.line("sh t0, 0(t1)"); // store x at addr
- },
- IR::Store32 => { // ( x addr -- )
- self.pop_some_to("t0 t1");
- self.line("sw t0, 0(t1)"); // store x at addr
- },
- IR::Store => { // ( x addr -- )
- self.pop_some_to("t0 t1");
- self.line("sd t0, 0(t1)"); // store x at addr
- },
- IR::StackPush(num) => {
- self.label(format!("# stackpush {}", num));
- self.line(format!("li t0, {}", num));
- self.push_from("t0");
- },
- IR::StackPushString(name) => {
- self.label(format!("# stackpushstring {}", name));
- self.line(format!("la t0, {}", name));
- self.push_from("t0");
- },
- IR::AddU64 => {
- self.label("# add");
- self.pop_call_push("t0 t1", "add t0, t0, t1", "t0");
- },
- IR::SubtractU64 => {
- self.label("# sub");
- self.pop_call_push("t0 t1", "sub t0, t0, t1", "t0");
- },
- IR::MultiplyU64 => {
- self.pop_call_push("t0 t1", "mul t0, t0, t1", "t0");
- },
- IR::DivideU64 => {
- self.pop_call_push("t0 t1", "div t0, t0, t1", "t0");
- },
- IR::ModU64 => {
- self.pop_call_push("t0 t1", "rem t0, t0, t1", "t0");
- },
- IR::Dup => {
- self.label("# dup");
- self.copy_top_stack_value_to("t0");
- self.push_from("t0");
- },
- IR::Swap => {
- self.label("# swap");
- self.pop_some_to("t1 t0");
- self.push_from("t0");
- self.push_from("t1");
- },
- IR::Rot => {
- self.label("# rot");
- self.pop_some_to("t0 t1 t2");
- self.push_from("t1");
- self.push_from("t2");
- self.push_from("t0");
- },
- IR::StackPointer => {
- self.label("# sp");
- self.line("addi t0, s2, 0");
- self.push_from("t0");
- },
- IR::Drop => {
- self.label("# drop");
- self.move_stack_ptr_by_cells(1);
- },
- IR::Equals => {
- // Yes, this is the same as subtract, since we're treating 0 as true, and
- // others as false.
- self.pop_call_push("t0 t1", "sub t0, t0, t1", "t0");
- },
- IR::GreaterThan => {
- self.label("# >");
- self.pop_some_to("t0 t1");
- self.line("sgt t0, t0, t1");
- self.line("seqz t0, t0"); // remember, 0 is true, others are false
- self.push_from("t0");
- },
- IR::LessThan => {
- self.label("# <");
- self.pop_some_to("t0 t1");
- self.line("slt t0, t0, t1");
- self.line("seqz t0, t0"); // remember, 0 is true, others are false
- self.push_from("t0");
- },
- IR::BitwiseOr => {
- self.pop_call_push("t0 t1", "or t0, t0, t1", "t0");
- },
- IR::Sys0 => {
- self.pop_call_push("a7", "ecall", "a0");
- },
- IR::Sys1 => {
- self.pop_call_push("a0 a7", "ecall", "a0");
- },
- IR::Sys2 => {
- self.pop_call_push("a0 a1 a7", "ecall", "a0");
- },
- IR::Sys3 => {
- self.pop_call_push("a0 a1 a2 a7", "ecall", "a0");
- },
- IR::Sys4 => {
- self.pop_call_push("a0 a1 a2 a3 a7", "ecall", "a0");
- },
- IR::Sys5 => {
- self.pop_call_push("a0 a1 a2 a3 a4 a7", "ecall", "a0");
- },
- IR::Sys6 => {
- self.pop_call_push("a0 a1 a2 a3 a4 a5 a7", "ecall", "a0");
- },
- // https://cmput229.github.io/229-labs-RISCV/RISC-V-Examples_Public/03-Conditionals/03b-If_Else.html
- IR::If => {
- self.label("# if");
- self.pop_to("t0");
- self.line(format!("bnez t0, _else_{}", if_block_count));
- if_stack.push(if_block_count);
- if_block_count += 1;
- },
- IR::Else => {
- self.label("# else");
- let if_counter = if_stack.last().unwrap().clone();
- self.line(format!("j _endif_{}", if_counter));
- self.label(format!("_else_{}:", if_counter));
- seen_else.insert(if_counter);
- },
- IR::EndIf => {
- self.label("# endif");
- let stack = &mut if_stack;
- let if_counter = stack.last().unwrap().clone();
- if !seen_else.contains(&if_counter) {
- self.label(format!("_else_{}:", if_counter));
- } else {
- self.label(format!("_endif_{}:", if_counter));
- seen_else.remove(&if_counter);
- }
- stack.pop();
- },
- IR::Loop => { // keep looping until is true/0
- self.label(format!("_loop_{}:", loop_count));
- self.pop_to("t0");
- self.line(format!("beqz t0, _endloop_{}", loop_count));
- loop_stack.push(loop_count);
- loop_count += 1;
- },
- IR::EndLoop => {
- let stack = &mut loop_stack;
- let loop_counter = stack.last().unwrap().clone();
- self.line(format!("j _loop_{}", loop_counter));
- self.label(format!("_endloop_{}:", loop_counter));
- stack.pop();
- },
- _ => bail!("not implemented yet: {:?}", ir),
- }
- }
-
- Ok(self.lines.join("\n"))
- }
-}
-
+++ /dev/null
-use anyhow::{Result, anyhow};
-
-#[derive(Debug, Clone)]
-pub enum Token<'a> {
- Word(&'a str),
- String(&'a str),
- NumU8(u8),
- NumI8(i8),
- NumU16(u16),
- NumI16(i16),
- NumU32(u32),
- NumI32(i32),
- NumU64(u64),
- NumI64(i64),
- NumF32(f32),
- NumF64(f64),
-}
-
-impl<'a> Token<'a>{
- fn parse_word_or_num(input: &'a str) -> Result<Token<'a>> {
- if input == "-" {
- return Ok(Token::Word(input))
- }
-
- // we're assuming any token starting with `-` with length greater than one
- // is a negative number
- if input.starts_with('-') || input.chars().nth(0).map(|x| x.is_numeric()).unwrap_or(false) {
- if input.contains(':') {
- let mut splat = input.split(':');
- let num = splat.next().ok_or(anyhow!("no number found"))?;
- let typ = splat.next().ok_or(anyhow!("no number type found"))?;
- match typ {
- "u8" => Ok(Token::NumU8(num.parse()?)),
- "i8" => Ok(Token::NumI8(num.parse()?)),
- "u16" => Ok(Token::NumU16(num.parse()?)),
- "i16" => Ok(Token::NumI16(num.parse()?)),
- "u32" => Ok(Token::NumU32(num.parse()?)),
- "i32" => Ok(Token::NumI32(num.parse()?)),
- "u64" => Ok(Token::NumU64(num.parse()?)),
- "i64" => Ok(Token::NumI64(num.parse()?)),
- "f32" => Ok(Token::NumF32(num.parse()?)),
- "f64" => Ok(Token::NumF64(num.parse()?)),
- _ => panic!("unknown number type")
- }
- } else {
- if input.contains('.') {
- Ok(Token::NumF64(input.parse()?))
- } else if input.starts_with('-') {
- Ok(Token::NumI64(input.parse()?))
- } else {
- Ok(Token::NumU64(input.parse()?))
- }
- }
- } else {
- Ok(Token::Word(input))
- }
- }
-}
-
-// TODO really want an iterator, not a vector
-pub fn tokenize<'a>(input: &'a str) -> Result<Vec<Token<'a>>> {
- let mut result = vec![];
- let mut string_start: Option<usize> = None;
- let mut word_or_num_start: Option<usize> = None;
- let mut last_is_backslash = false;
- let mut last_is_whitespace = true;
- let mut in_doc_comment = false;
- let mut in_line_comment = false;
- let mut index = 0;
- let mut first_char = true;
-
-
- for char in input.chars() {
- if first_char {
- first_char = false;
- } else {
- index += 1;
- }
-
- if in_doc_comment {
- if char == ')' {
- in_doc_comment = false;
- last_is_whitespace = true; // not really true, but means don't need space after
- }
- continue;
- }
-
- if in_line_comment {
- word_or_num_start = None;
- if char == '\n' {
- in_line_comment = false;
- last_is_whitespace = true; // not really true, but means don't need space after
- }
- continue;
- }
-
- if char == '"' {
- if let Some(start) = string_start {
- if !last_is_backslash {
- result.push(Token::String(&input[start..index]));
- string_start = None;
- }
- } else {
- string_start = Some(index + 1)
- }
- last_is_backslash = false;
- last_is_whitespace = false;
- continue;
- }
-
-
- if string_start.is_some() {
- last_is_backslash = char == '\\';
- continue;
- }
-
- if char.is_whitespace() {
- if last_is_backslash {
- in_line_comment = true;
- } else if !last_is_whitespace && let Some(start) = word_or_num_start {
- let token = &input[start..index];
- if token == "(" {
- in_doc_comment = true;
- } else {
- result.push(Token::parse_word_or_num(&input[start..index])?);
- }
- word_or_num_start = None;
- }
- last_is_whitespace = true;
- last_is_backslash = false;
- continue;
- }
-
- last_is_backslash = char == '\\';
-
- if index == input.len() - 1 {
- if !last_is_whitespace && let Some(start) = word_or_num_start {
- result.push(Token::parse_word_or_num(&input[start..])?);
- }
- continue;
- }
-
- if last_is_whitespace { // start of word or num (we already handled strings)
- word_or_num_start = Some(index);
- last_is_whitespace = false;
- }
- }
- Ok(result)
-}
-
-#[cfg(test)]
-mod tests {
- use super::*;
-
- #[test]
- fn try_some_tokenizing() {
- let result = tokenize("
-
- \\ soup
- 2 3.4 - -88 bacon \"hello\" 43:f32 2345:u32 -57:i8 soup
-");
- println!("result: {:?}", result);
- }
-
- #[test]
- fn comments() {
- let result = tokenize("
- (
- foo
- bar
- )
- : baz ( x y -- z )
- chicken
- soup
- ;
- ");
- println!("result: {:?}", result);
- }
-
- #[test]
- fn strings() {
- let result = tokenize("
-dup \\ ( stuff )
-\"hello!\"
-");
- println!("result: {:?}", result);
- }
-}
+++ /dev/null
-\ vim: filetype=forth
-
-: mmap 222 sys6 ;
-: munmap 215 sys2 ;
-
-: PROT_READ 1 ;
-: PROT_WRITE 2 ;
-: MAP_PRIVATE 2 ;
-: MAP_ANONYMOUS 32 ;
-
-: ALLOC_PROT PROT_READ PROT_WRITE | ;
-: ALLOC_MAP MAP_PRIVATE MAP_ANONYMOUS | ;
-
-\ This is a raw way to allocate memory. Callers to this will need to know
-\ how much memory was allocated, to call dealloc later.
-: alloc_raw ( byte-count -- addr )
- 0 swap ALLOC_PROT ALLOC_MAP -1:i32 0 mmap
-;
-
-: dealloc ( addr byte-count -- )
- munmap
-;
-
-export alloc
-: alloc ( byte-count -- addr )
- dup \ we'll need the length again later
- 8 + \ add 64 bits to the front where we'll store the length
- alloc_raw \ ( byte-count addr )
- dup \ ( byte-count addr addr)
- rot \ ( addr addr byte-count )
- rot \ ( addr byte-count addr )
- ! \ write the length to the addr, now ( addr )
- 8 + \ here's the address we'll return, 8 bytes forward to account for the length
-;
-
-export free
-: free ( addr -- )
- 8 - \ the real memory start is 8 bytes before
- dup \ ( addr addr )
- @ \ ( addr byte-length )
- 8 + \ add 8 to the byte length to refer to the whole region from alloc.
- dealloc
-;
-
-import "std:out"
-
-: copy_byte ( addr_from addr_to -- addr_from addr_to )
- swap \ ( addr_to addr_from )
- dup \ ( addr_to addr_from addr_from )
- @:8 \ ( addr_to addr_from byte )
- rot \ ( addr_from byte addr_to )
- dup \ ( addr_from byte addr_to addr_to)
- rot \ ( addr_from addr_to addr_to byte )
- swap \ ( addr_from addr_to byte addr_to )
- !:8 \ ( addr_from addr_to )
-;
-
-export memcopy
-: memcopy ( addr_from addr_to count -- )
- dup \ ( addr_from addr_to count count )
- loop \ ( addr_from addr_to count )
- rot \ ( addr_to count addr_from )
- rot \ ( count addr_from addr_to )
- copy_byte \ ( count addr_from addr_to )
- 1 + \ ( count addr_from addr_to+1 )
- rot \ ( addr_from addr_to+1 count )
- 1 - \ ( addr_from addr_to+1 count-1 )
- rot \ ( addr_to+1 count-1 addr_from )
- 1 + \ ( addr_to+1 count-1 addr_from+1 )
- rot rot \ ( addr_from+1 addr_to+1 count-1 )
- dup \ ( addr_from+1 addr_to+1 count-1 count-1)
- endloop \ ( addr_from+count addr_to+count 0 )
- drop drop drop \ ( )
-;
+++ /dev/null
-\ vim: filetype=forth
-
-: write ( fd ptr len -- bytes-written-or-err )
- 64
- sys3
-;
-
-import "std:string"
-
-export puts
-
-: puts ( addr -- addr)
- dup dup \ ( addr addr addr )
- strlen \ ( addr addr len )
- 1 \ ( addr addr len 1 )
- rot \ ( addr len 1 addr )
- rot \ ( addr 1 addr len )
- write \ ( addr bytes-written-or-err )
- drop \ ( addr )
-;
-
-: ZERO_CHAR 48 ;
-: NEWLINE_CHAR 10 ;
-
-export putn
-
-: putn ( num -- num )
- dup dup \ ( num num num )
- 10 rot rot \ ( num 10 num num ) // Mutltidigit stop point
- 10 < \ ( num 10 num is<10 )
- loop \ ( num 10 num' )
- dup \ ( num 10 num' num' )
- 10 % \ ( num 10 num' digit )
- swap \ ( num 10 digit num' )
- 10 / \ ( num 10 digit num'/10 )
- dup 10 < \ ( num 10 digit num'/10 is<10 )
- endloop \ ( num 10 digitn ... digit1 )
- dup 9 > \ ( num 10 digitn ... digit1 digit1>9 )
- loop \ ( num 10 digitn ... digit1 )
- ZERO_CHAR + \ \ ( num 10 digitn ... digit1 )
- sp \ ( num 10 digitn ... digit1 ptr )
- 1 \ ( num 10 digitn ... digit1 ptr 1 )
- swap \ ( num 10 digitn ... digit1 1 ptr )
- 1 \ ( num 10 digitn ... digit1 1 ptr 1 )
- write \ ( num 10 digitn ... digit1 result )
- drop drop \ ( num 10 digitn ... digit2 )
- dup 9 > \ ( num 10 digitn ... digit2 digit2>9 )
- endloop \ ( num 10 )
- drop \ ( num )
- NEWLINE_CHAR \ ( num 10 )
- sp \ ( num 10 ptr )
- 1 \ ( num 10 ptr 1 )
- swap \ (num 10 1 ptr )
- 1 \ ( num 10 1 ptr 1 )
- write \ ( num 10 result )
- drop \ ( num 10 )
- drop \ ( num )
-;
-
+++ /dev/null
-\ vim: filetype=forth
-
-export exit
-
-: exit ( code -- )
- 93
- sys1
-;
+++ /dev/null
-\ vim: filetype=forth
-
-: strlen ( addr -- len )
- dup dup \ ( addr addr addr )
- @:8 \ ( addr addr byte )
- loop \ ( addr addr )
- 1 + \ ( addr addr+1 )
- dup \ ( addr addr+1 addr+1)
- @:8 \ ( addr addr+1 byte )
- endloop \ ( addr addr+len )
- swap \ ( addr+len addr )
- - \ ( len )
-;
-
-export strlen
+++ /dev/null
-\ vim: filetype=forth
-
-import "std:process"
-
-export assert
-
-: assert ( x -- )
- if
- drop
- else
- 1 exit
- endif
-;
+++ /dev/null
-#include <stdio.h>
-
-extern unsigned long data_stack_end;
-register unsigned long * stack_pointer asm("s2");
-
-void putstack() {
- unsigned long * stack_index = &data_stack_end;
- printf("stack: ");
- while (stack_index != stack_pointer) {
- printf("%ld ", *stack_index);
- stack_index -= 1;
- }
- printf("%ld\n", *stack_pointer);
-}
-
-
+++ /dev/null
-UNAME=$(uname -m)
-CMD_PREFIX=$([ "$UNAME" = "riscv64" ] && echo "" || echo "riscv64-unknown-linux-gnu-")
-AS="${CMD_PREFIX}as"
-LD="${CMD_PREFIX}ld"
-CC="${CMD_PREFIX}cc"
-
-../target/debug/sorelc test1.sorel
-$AS -g -o test1.o test1.asm
-$CC -O1 -no-pie -o test1.out test1.o putstack.c -nostartfiles
-# $LD -o test1.out test1.o
-./test1.out
+++ /dev/null
-\ vim: filetype=forth
-
-import "./assert.sorel"
-import "std:out"
-
-"Hello, World! \n" puts
-drop
-
-42 putn
-drop
-
-import "std:mem"
-
-64 alloc
-free
-
-0 assert
-
-import "std:string"
-import "std:mem"
-import "std:out"
-
-extern putstack
-
-16 alloc \ ( ptr )
-dup \ ( ptr ptr )
-"To be copied!" \ ( ptr ptr strptr )
-dup \ ( ptr ptr strptr strptr )
-strlen 1 + \ ( ptr ptr strptr len )
-rot \ ( ptr strptr len ptr )
-swap \ ( ptr strptr ptr len )
-memcopy \ ( ptr )
-puts \ ( ptr )
-free \ ( )
-
-
--- /dev/null
+[package]
+name = "sorel-ir"
+version = "0.1.0"
+edition = "2024"
+
+[dependencies]
+serde = "1.0.228"
+serde_derive = "1.0.228"
+serde_yaml = "0.9.34"
--- /dev/null
+use serde_yaml::{from_str, to_string, Error};
+use serde_derive::{Serialize, Deserialize};
+
+#[derive(Serialize, Deserialize, Debug, Clone)]
+pub enum IR {
+ Label(String),
+ Call(String),
+ Ret,
+ StackPush(u64),
+ StackPushString(String), // refers to string label, not the string itself
+ StringDef(String, String), // first is string label, second is string value
+
+ // These next ones should always be inlined, so they're in IR.
+ Load, // @ ( addr -- x ) -- Fetch memory contents at addr
+ Load8,
+ Load16,
+ Load32,
+ Store, // ! ( x addr -- ) -- Store x at addr
+ Store8,
+ Store16,
+ Store32,
+
+ // These ones might not be inlined, but should be built-in, so a compiler might
+ // turn this into `Call(String)` before translating to assembly/machine-code, but
+ // an IR interpreter may just execute them.
+ AddU64,
+ SubtractU64,
+ MultiplyU64,
+ DivideU64,
+ ModU64,
+ Equals,
+ GreaterThan,
+ LessThan,
+ BitwiseOr,
+ Dup,
+ Swap,
+ Drop,
+ Over,
+ Rot,
+ StackPointer,
+ If,
+ Else,
+ EndIf,
+ Loop,
+ EndLoop,
+
+ // System calls
+ Sys0,
+ Sys1,
+ Sys2,
+ Sys3,
+ Sys4,
+ Sys5,
+ Sys6,
+}
+
+// This is like an .o file.
+#[derive(Serialize, Deserialize, Debug)]
+pub struct IRObject {
+ pub text: Vec<IR>,
+ pub data: Vec<IR>,
+}
+
+impl IRObject {
+ pub fn to_s(&self) -> Result<String, Error> {
+ to_string(self)
+ }
+
+ pub fn from_s(source: &str) -> Result<Self, Error> {
+ from_str(source)
+ }
+}
--- /dev/null
+[package]
+name = "sorelc"
+version = "0.1.0"
+edition = "2024"
+
+[dependencies]
+sorel-ir = { workspace = true }
+anyhow = "1.0.100"
--- /dev/null
+use crate::parser::Module;
+use crate::tokenizer::{Token, tokenize};
+use sorel_ir::*;
+
+use std::collections::{HashSet, HashMap};
+use std::path::PathBuf;
+use std::rc::Rc;
+use std::include_str;
+
+use anyhow::{Result, bail};
+
+macro_rules! push_num {
+ ($num:ident) => { IR::StackPush(*$num as u64) };
+ ($num:ident, $num_typ:ty) => { IR::StackPush(*$num as $num_typ as u64) };
+}
+
+#[derive(Debug, Default)]
+struct IRModule {
+ data: Vec<IR>,
+ text: Vec<IR>,
+ imports: Vec<Rc<IRModule>>,
+ exports: Vec<String>,
+ externs: Vec<String>,
+ // TODO these next two should form an enum, not two options
+ source_file: Option<PathBuf>,
+ std_specifier: Option<String>,
+ number: usize,
+}
+
+impl IRModule {
+ fn get_label_for_call(&self, name: &String) -> String {
+ if self.externs.contains(name) {
+ return name.clone();
+ }
+ let mut found: Option<usize> = None;
+ for imported in &self.imports {
+ if imported.exports.contains(name) {
+ found = Some(imported.number);
+ // Don't break here, since the last one should win.
+ }
+ }
+ if let Some(found) = found {
+ format!("_m{}_{}", found, name)
+ } else {
+ // TODO check if it's even a word locally. If not, bail.
+ format!("_m{}_{}", self.number, name)
+ }
+ }
+
+ fn get_label(&self, name: &String) -> String {
+ format!("_m{}_{}", self.number, name)
+ }
+}
+
+#[derive(Default)]
+struct ImportTree {
+ data: Vec<IR>,
+ text: Vec<IR>,
+ all_modules: HashMap<String, Rc<IRModule>>,
+ all_exports: HashSet<String>,
+ entrypoint: Rc<IRModule>,
+ module_count: usize,
+ collapse_seen: HashSet<String>,
+}
+
+fn std_import(specifier: &str) -> Result<&str> {
+ match specifier {
+ "std:mem" => Ok(include_str!("../../stdlib/mem.sorel")),
+ "std:out" => Ok(include_str!("../../stdlib/out.sorel")),
+ "std:string" => Ok(include_str!("../../stdlib/string.sorel")),
+ "std:process" => Ok(include_str!("../../stdlib/process.sorel")),
+ _ => bail!("{} is not a standard library module", specifier),
+ }
+}
+
+impl ImportTree {
+ fn import(&mut self, importer_dir: &PathBuf, specifier: &str, is_entrypoint: bool) -> Result<Rc<IRModule>> {
+ Ok(if specifier.starts_with("std:") {
+ if self.all_modules.contains_key(specifier) {
+ let module = self.all_modules.get(specifier).unwrap().clone();
+ return Ok(module);
+ }
+ let contents = std_import(specifier)?;
+ let parsed = &Module::parse(tokenize(&contents)?, is_entrypoint)?;
+ let module = self.generate_internal(None, Some(specifier.to_string()), parsed);
+ let module = Rc::new(module);
+ self.all_modules.insert(specifier.to_string(), module.clone());
+ module
+ } else {
+ let mut path = PathBuf::from(specifier);
+ if path.is_relative() {
+ let mut new_path = importer_dir.clone();
+ new_path.push(path);
+ path = new_path.canonicalize()?;
+ }
+ let path_key = path.to_string_lossy().to_string();
+ if self.all_modules.contains_key(&path_key) {
+ let module = self.all_modules.get(&path_key).unwrap().clone();
+ return Ok(module);
+ }
+
+ let contents = &std::fs::read_to_string(&path)?;
+ let tokens = tokenize(&contents)?;
+ let parsed = &Module::parse(tokens, is_entrypoint)?;
+
+ let module = self.generate_internal(Some(path), None, parsed);
+ let module = Rc::new(module);
+ self.all_modules.insert(path_key, module.clone());
+ if is_entrypoint {
+ self.entrypoint = module.clone();
+ }
+ module
+ })
+
+ }
+
+ fn generate_internal(&mut self, path: Option<PathBuf>, std_specifier: Option<String>, module: &Module) -> IRModule {
+ // Eventually these will end up being sections in assembly
+ let mut text = vec![];
+ let mut data = vec![];
+
+ let mut imports = vec![];
+ if let Some(ref path) = path {
+ module.imports.iter().for_each(|imported| {
+ if let Some(parent_path) = path.parent() {
+ match self.import(&parent_path.to_path_buf(), imported, false) {
+ Ok(module) => {
+ imports.push(module);
+ },
+ Err(msg) => {
+ eprintln!("{}", msg);
+ }
+ }
+ } else {
+ }
+ });
+ } else {
+ // We're in a stdlib module, which can only import other stdlib
+ // modules.
+ module.imports.iter().for_each(|imported| {
+ match self.import(&PathBuf::new(), imported, false) {
+ Ok(module) => {
+ imports.push(module);
+ },
+ Err(msg) => {
+ eprintln!("{}", msg);
+ }
+ }
+ });
+ }
+
+ let exports: Vec<_> = module.exports.iter().map(|s| {
+ self.all_exports.insert(s.to_string());
+ s.to_string()
+ }).collect();
+
+ let externs = module.externs.iter().map(|s| s.to_string()).collect();
+
+ text.push(module.words.iter().map(|def| {
+ let mut body = def.instructions.iter().map(|inst| {
+ let mapped_ir = match inst {
+ Token::Word(word) => {
+ match *word {
+ "@" => IR::Load,
+ "@:8" => IR::Load8,
+ "@:16" => IR::Load16,
+ "@:32" => IR::Load32,
+ "!" => IR::Store,
+ "!:8" => IR::Store8,
+ "!:16" => IR::Store16,
+ "!:32" => IR::Store32,
+ "dup" => IR::Dup,
+ "swap" => IR::Swap,
+ "drop" => IR::Drop,
+ "over" => IR::Over,
+ "rot" => IR::Rot,
+ "sp" => IR::StackPointer,
+ "if" => IR::If,
+ "else" => IR::Else,
+ "endif" => IR::EndIf,
+ "loop" => IR::Loop,
+ "endloop" => IR::EndLoop,
+ "=" => IR::Equals,
+ ">" => IR::GreaterThan,
+ "<" => IR::LessThan,
+ "+" => IR::AddU64,
+ "-" => IR::SubtractU64,
+ "*" => IR::MultiplyU64,
+ "/" => IR::DivideU64,
+ "%" => IR::ModU64,
+ "|" => IR::BitwiseOr,
+ "sys0" => IR::Sys0,
+ "sys1" => IR::Sys1,
+ "sys2" => IR::Sys2,
+ "sys3" => IR::Sys3,
+ "sys4" => IR::Sys4,
+ "sys5" => IR::Sys5,
+ "sys6" => IR::Sys6,
+ // TODO num type specfic math like `+:i32`, etc.
+ _ => IR::Call(String::from(*word))
+ }
+ },
+ Token::String(text) => {
+ let string_label = format!("string_{}", data.len());
+ data.push(IR::StringDef(string_label.clone(), String::from(*text)));
+ IR::StackPushString(string_label)
+ },
+ Token::NumU8(num) => push_num!(num),
+ Token::NumI8(num) => push_num!(num, u8),
+ Token::NumU16(num) => push_num!(num),
+ Token::NumI16(num) => push_num!(num, u16),
+ Token::NumU32(num) => push_num!(num),
+ Token::NumI32(num) => push_num!(num, u32),
+ Token::NumU64(num) => push_num!(num),
+ Token::NumI64(num) => push_num!(num),
+ Token::NumF32(num) => push_num!(num),
+ Token::NumF64(num) => push_num!(num),
+ };
+ mapped_ir
+ }).collect::<Vec<_>>();
+
+ let mut result = vec![IR::Label(def.name.to_string())];
+ result.append(&mut body);
+ result.push(IR::Ret);
+ result
+ }).flatten().collect::<Vec<_>>());
+
+ let number = self.module_count;
+ self.module_count += 1;
+
+ IRModule {
+ text: text.into_iter().flatten().collect::<Vec<_>>(),
+ data,
+ imports,
+ exports,
+ externs,
+ source_file: path,
+ std_specifier,
+ number,
+ }
+ }
+
+ fn collapse(&mut self, module: Rc<IRModule>) -> Result<()> {
+ let seen_key = if let Some(source_file) = &module.source_file {
+ source_file.to_string_lossy().to_string()
+ } else {
+ module.std_specifier.clone().unwrap()
+ };
+ if self.collapse_seen.contains(&seen_key) {
+ return Ok(())
+ }
+
+ for imported in module.imports.clone() {
+ self.collapse(imported)?;
+ }
+
+ let is_entrypoint = module.source_file == self.entrypoint.source_file;
+
+ let module_number = module.number;
+
+ for string in &module.data {
+ if let IR::StringDef(name, val) = string {
+ let new_name = format!("{}_{}", name, module_number);
+ self.data.push(IR::StringDef(new_name, val.clone()));
+ } else {
+ bail!("non-string data");
+ }
+ }
+
+ for instruction in &module.text {
+ let new_instruction = match instruction {
+ IR::StackPushString(name) => {
+ let new_name = format!("{}_{}", name, module_number);
+ IR::StackPushString(new_name)
+ },
+ IR::Label(name) => {
+ if is_entrypoint && name == "main" {
+ instruction.clone()
+ } else {
+ IR::Label(module.get_label(name))
+ }
+ },
+ IR::Call(name) => {
+ IR::Call(module.get_label_for_call(name))
+ },
+ _ => instruction.clone()
+ };
+ self.text.push(new_instruction);
+ }
+
+ self.collapse_seen.insert(seen_key);
+
+ Ok(())
+ }
+}
+
+pub fn compile(path: &str) -> Result<IRObject> {
+ let dir = std::env::current_dir()?;
+ let mut tree: ImportTree = Default::default();
+ let module = tree.import(&dir, path, true)?;
+ tree.collapse(module)?;
+ // TODO remove unused words
+ Ok(IRObject {
+ data: tree.data,
+ text: tree.text,
+ })
+}
--- /dev/null
+mod tokenizer;
+mod parser;
+mod ir;
+mod riscv_asm_codegen;
+
+use anyhow::Result;
+
+use std::fs::File;
+use std::io::Write;
+use std::path::PathBuf;
+
+fn main() -> Result<()> {
+ let filename = std::env::args().nth(1).expect("must provide a file to compile");
+ let module = ir::compile(&filename)?;
+ let mut generator = riscv_asm_codegen::CodeGen::new(&module, 4096);
+ let mut asm_path = PathBuf::from(filename);
+ asm_path.set_extension("asm");
+ let mut output = File::create(asm_path)?;
+ write!(output, "{}\n", generator.assembly()?)?;
+ Ok(())
+}
--- /dev/null
+use crate::tokenizer::Token;
+use anyhow::{Result, bail};
+
+#[derive(Debug)]
+pub struct WordDefinition<'a> {
+ pub name: &'a str,
+ pub instructions: Vec<Token<'a>>,
+}
+
+#[derive(Debug)]
+pub struct Module<'a> {
+ pub words: Vec<WordDefinition<'a>>,
+ pub imports: Vec<&'a str>,
+ pub exports: Vec<&'a str>,
+ pub externs: Vec<&'a str>,
+}
+
+impl<'a> Module<'a> {
+ pub fn parse(input: Vec<Token<'a>>, is_entrypoint: bool) -> Result<Self> {
+ let mut result = vec![];
+ let mut main = vec![];
+ let mut exports = vec![];
+ let mut imports = vec![];
+ let mut externs = vec![];
+ let mut current_word: Option<WordDefinition> = None;
+ let mut about_to_start_word_def = false;
+ let mut last_was_import = false;
+ let mut last_was_export = false;
+ let mut last_was_extern = false;
+
+ for token in input {
+ if about_to_start_word_def {
+ if let Token::Word(name) = token {
+ current_word = Some(WordDefinition {
+ name,
+ instructions: vec![],
+ });
+ about_to_start_word_def = false;
+ continue;
+ } else {
+ bail!("{:?} is not a valid word name!", token);
+ }
+ } else if let Token::Word(word) = token {
+ if word == ":" {
+ if current_word.is_some() {
+ bail!("can't define words inside word definitions!");
+ }
+ about_to_start_word_def = true;
+ continue;
+ }
+ if word == ";" {
+ let word = current_word.take();
+ if let Some(word) = word {
+ result.push(word);
+ continue;
+ } else {
+ bail!("`;` must be at the end of a word definition");
+ }
+ }
+ }
+ if let Some(ref mut current_word) = current_word {
+ current_word.instructions.push(token);
+ } else {
+ match token {
+ Token::Word(word) => {
+ if word == "import" {
+ last_was_import = true;
+ } else if word == "export" {
+ last_was_export = true;
+ } else if word == "extern" {
+ last_was_extern = true;
+ } else {
+ if last_was_export {
+ exports.push(word);
+ last_was_export = false;
+ } else if last_was_extern {
+ externs.push(word);
+ last_was_extern = false;
+ } else {
+ main.push(token.clone());
+ }
+ }
+ },
+ Token::String(string) => {
+ if last_was_import {
+ imports.push(string);
+ last_was_import = false;
+ } else {
+ main.push(token.clone());
+ }
+ },
+ _ => {
+ main.push(token.clone());
+ }
+ };
+ }
+ }
+
+ if about_to_start_word_def || current_word.is_some() {
+ bail!("unfinished word definition!");
+ }
+
+ if is_entrypoint {
+ result.push(WordDefinition {
+ name: "main",
+ instructions: main,
+ });
+ }
+
+ Ok(Module { words: result, imports, exports, externs })
+ }
+
+ #[cfg(test)]
+ pub fn debug_print(&self) {
+ for word in &self.words {
+ println!("{}", word.name);
+ for instruction in &word.instructions {
+ println!(" {:?}", instruction);
+ }
+ }
+ }
+}
+
+
+
+#[cfg(test)]
+mod tests {
+ use super::*;
+
+ #[test]
+ fn try_some_parsing() {
+ let result = Module::parse(crate::tokenizer::tokenize("
+: hello world 16 \"planet\" ;
+: soup chicken 4.5 hello ;
+
+hello soup
+").unwrap(), true).unwrap();
+ result.debug_print();
+ }
+}
--- /dev/null
+use sorel_ir::*;
+
+use anyhow::*;
+
+use std::collections::{HashMap, HashSet};
+use std::fmt::Display;
+
+pub struct CodeGen<'a> {
+ module: &'a IRObject,
+ data_stack_size: usize,
+ lines: Vec<String>,
+}
+
+
+// Some inspiration
+// ================
+//
+// * https://github.com/aw/fiveforths/blob/master/docs/REFERENCE.md#registers-list
+// * Except using sp as a more C ABI style stack pointer, and s2 for the data stack
+//
+
+// Implementation Choices
+// ======================
+//
+// Data Stack pointer: s2
+// No return stack pointer (using C ABI, so sp, sorta)
+// Use t0, t1, t2 for temporary values in words
+// Data stack grows down
+
+
+macro_rules! asm_macro {
+ ($name:ident, $src:expr) => {
+ fn $name(&mut self) {
+ self.line($src);
+ }
+ };
+ ($name:ident, $src:expr, $arg0:ty) => {
+ fn $name(&mut self, val0: $arg0) {
+ self.line(format!($src, val0));
+ }
+ };
+ ($name:ident, $src:expr, $arg0:ty, $arg1:ty) => {
+ fn $name(&mut self, val0: $arg0, val1: $arg1) {
+ self.line(format!($src, val0, val1));
+ }
+ };
+}
+
+fn mangle(input: &str) -> String {
+ input
+ .replace("<", "_LT_")
+ .replace(">", "_GT_")
+ .replace("-", "___")
+}
+
+impl<'a> CodeGen<'a> {
+ pub fn new(ir_mod: &'a IRObject, data_stack_size: usize) -> Self {
+ Self {
+ module: ir_mod,
+ data_stack_size,
+ lines: vec![],
+ }
+ }
+
+ fn line<S: Display>(&mut self, line: S) {
+ self.lines.push(format!(" {}", line));
+
+ }
+
+ fn label<S: Display>(&mut self, line: S) {
+ self.lines.push(line.to_string());
+ }
+
+ asm_macro!(copy_top_stack_value_to, "ld {}, 0(s2)", &str);
+ asm_macro!(copy_offset_stack_value_to, "ld {}, {}*8(s2)", &str, isize);
+ asm_macro!(copy_to_top_of_stack, "sd {}, 0(s2)", &str);
+ asm_macro!(move_stack_ptr_by_cells, "addi s2, s2, {}*8", isize);
+
+ fn pop_to(&mut self, reg: &str) {
+ self.copy_top_stack_value_to(reg);
+ self.move_stack_ptr_by_cells(1);
+ }
+
+ fn pop_some_to(&mut self, regs: &str) {
+ let mut regs = regs.trim().split(" ").collect::<Vec<_>>();
+ regs.reverse();
+ let count = regs.len();
+ let mut index = 0;
+ for reg in regs {
+ self.copy_offset_stack_value_to(reg, index);
+ index += 1;
+ }
+ self.move_stack_ptr_by_cells(count as isize);
+ }
+
+ fn push_from(&mut self, reg: &str) {
+ self.move_stack_ptr_by_cells(-1);
+ self.copy_to_top_of_stack(reg);
+ }
+
+ fn pop_call_push(&mut self, regs: &str, call: &str, reg: &str) {
+ self.pop_some_to(regs);
+ self.line(call);
+ self.push_from(reg);
+ }
+
+ pub fn assembly(&mut self) -> Result<String>{
+ let mut string_table = HashMap::new();
+
+ // Static strings
+ self.label(".section .rodata\n");
+ self.label(".align 3\n");
+ for ir in &self.module.data {
+ match ir {
+ IR::StringDef(string_label, some_string) => {
+ string_table.insert(some_string.clone(), string_label);
+ self.label(format!("{}:", string_label));
+ self.line(format!(".asciz \"{}\"", some_string)); // should this be .asciz?
+ self.label("");
+ },
+ _ => bail!("Currently only string definitions are supported in the data section.")
+ }
+ }
+
+ // Data stack
+ self.label(".data\n");
+ self.label(".align 3\n");
+ self.label("data_stack:");
+ self.line(format!(".space {}", self.data_stack_size));
+ self.label(".globl data_stack_end\ndata_stack_end:\n");
+
+ // Code
+ self.label(".text\n");
+ self.label(".align 3\n");
+
+ let mut if_block_count = 0;
+ let mut if_stack = vec![];
+ let mut loop_count = 0;
+ let mut loop_stack = vec![];
+ let mut seen_else = HashSet::new();
+ let mut last_label = "";
+
+ for ir in &self.module.text {
+ match ir {
+ IR::Label(name) => {
+ last_label = name;
+ if name == "main" {
+ self.label(".globl _start"); // TODO is globl necessary?
+ self.label("_start:");
+ self.line("la s2, data_stack_end # set initial data stack pointer");
+ } else {
+ let mangled = mangle(name);
+ self.label(format!(".globl {}", mangled));
+ self.label(format!("{}:", mangled));
+ }
+ self.line("addi sp, sp, -16 # allocate 16 bytes on stack"); // allocate 16 bytes on stack
+ self.line("sd ra, 8(sp) # store return address on stack"); // store return address on stack
+ },
+ IR::Call(name) => {
+ let mangled = mangle(name);
+ self.label(format!("# call {}", mangled));
+ self.line(format!("call {}", mangled));
+ },
+ IR::Ret => {
+ if last_label == "main" {
+ self.label("# exit 0 syscall");
+ self.line("li a7, 93");
+ self.line("mv a0, x0");
+ self.line("ecall");
+ } else {
+ self.line("ld ra, 8(sp)"); // load return address from stack
+ self.line("addi sp, sp, 16"); // restore stack pointer
+ self.line("ret");
+ }
+ },
+ IR::Load8 => {
+ self.label("# load 8");
+ self.copy_top_stack_value_to("t0");
+ self.line("lbu t0, 0(t0)"); // deref pointer in t0 to t0
+ self.copy_to_top_of_stack("t0");
+ },
+ IR::Load16 => {
+ self.label("# load 16");
+ self.copy_top_stack_value_to("t0");
+ self.line("lhu t0, 0(t0)"); // deref pointer in t0 to t0
+ self.copy_to_top_of_stack("t0");
+ },
+ IR::Load32 => {
+ self.label("# load 32");
+ self.copy_top_stack_value_to("t0");
+ self.line("lwu t0, 0(t0)"); // deref pointer in t0 to t0
+ self.copy_to_top_of_stack("t0");
+ },
+ IR::Load => {
+ self.label("# load 64");
+ self.copy_top_stack_value_to("t0");
+ self.line("ld t0, 0(t0)"); // deref pointer in t0 to t0
+ self.copy_to_top_of_stack("t0");
+ },
+ IR::Store8 => { // ( x addr -- )
+ self.pop_some_to("t0 t1");
+ self.line("sb t0, 0(t1)"); // store x at addr
+ },
+ IR::Store16 => { // ( x addr -- )
+ self.pop_some_to("t0 t1");
+ self.line("sh t0, 0(t1)"); // store x at addr
+ },
+ IR::Store32 => { // ( x addr -- )
+ self.pop_some_to("t0 t1");
+ self.line("sw t0, 0(t1)"); // store x at addr
+ },
+ IR::Store => { // ( x addr -- )
+ self.pop_some_to("t0 t1");
+ self.line("sd t0, 0(t1)"); // store x at addr
+ },
+ IR::StackPush(num) => {
+ self.label(format!("# stackpush {}", num));
+ self.line(format!("li t0, {}", num));
+ self.push_from("t0");
+ },
+ IR::StackPushString(name) => {
+ self.label(format!("# stackpushstring {}", name));
+ self.line(format!("la t0, {}", name));
+ self.push_from("t0");
+ },
+ IR::AddU64 => {
+ self.label("# add");
+ self.pop_call_push("t0 t1", "add t0, t0, t1", "t0");
+ },
+ IR::SubtractU64 => {
+ self.label("# sub");
+ self.pop_call_push("t0 t1", "sub t0, t0, t1", "t0");
+ },
+ IR::MultiplyU64 => {
+ self.pop_call_push("t0 t1", "mul t0, t0, t1", "t0");
+ },
+ IR::DivideU64 => {
+ self.pop_call_push("t0 t1", "div t0, t0, t1", "t0");
+ },
+ IR::ModU64 => {
+ self.pop_call_push("t0 t1", "rem t0, t0, t1", "t0");
+ },
+ IR::Dup => {
+ self.label("# dup");
+ self.copy_top_stack_value_to("t0");
+ self.push_from("t0");
+ },
+ IR::Swap => {
+ self.label("# swap");
+ self.pop_some_to("t1 t0");
+ self.push_from("t0");
+ self.push_from("t1");
+ },
+ IR::Rot => {
+ self.label("# rot");
+ self.pop_some_to("t0 t1 t2");
+ self.push_from("t1");
+ self.push_from("t2");
+ self.push_from("t0");
+ },
+ IR::StackPointer => {
+ self.label("# sp");
+ self.line("addi t0, s2, 0");
+ self.push_from("t0");
+ },
+ IR::Drop => {
+ self.label("# drop");
+ self.move_stack_ptr_by_cells(1);
+ },
+ IR::Equals => {
+ // Yes, this is the same as subtract, since we're treating 0 as true, and
+ // others as false.
+ self.pop_call_push("t0 t1", "sub t0, t0, t1", "t0");
+ },
+ IR::GreaterThan => {
+ self.label("# >");
+ self.pop_some_to("t0 t1");
+ self.line("sgt t0, t0, t1");
+ self.line("seqz t0, t0"); // remember, 0 is true, others are false
+ self.push_from("t0");
+ },
+ IR::LessThan => {
+ self.label("# <");
+ self.pop_some_to("t0 t1");
+ self.line("slt t0, t0, t1");
+ self.line("seqz t0, t0"); // remember, 0 is true, others are false
+ self.push_from("t0");
+ },
+ IR::BitwiseOr => {
+ self.pop_call_push("t0 t1", "or t0, t0, t1", "t0");
+ },
+ IR::Sys0 => {
+ self.pop_call_push("a7", "ecall", "a0");
+ },
+ IR::Sys1 => {
+ self.pop_call_push("a0 a7", "ecall", "a0");
+ },
+ IR::Sys2 => {
+ self.pop_call_push("a0 a1 a7", "ecall", "a0");
+ },
+ IR::Sys3 => {
+ self.pop_call_push("a0 a1 a2 a7", "ecall", "a0");
+ },
+ IR::Sys4 => {
+ self.pop_call_push("a0 a1 a2 a3 a7", "ecall", "a0");
+ },
+ IR::Sys5 => {
+ self.pop_call_push("a0 a1 a2 a3 a4 a7", "ecall", "a0");
+ },
+ IR::Sys6 => {
+ self.pop_call_push("a0 a1 a2 a3 a4 a5 a7", "ecall", "a0");
+ },
+ // https://cmput229.github.io/229-labs-RISCV/RISC-V-Examples_Public/03-Conditionals/03b-If_Else.html
+ IR::If => {
+ self.label("# if");
+ self.pop_to("t0");
+ self.line(format!("bnez t0, _else_{}", if_block_count));
+ if_stack.push(if_block_count);
+ if_block_count += 1;
+ },
+ IR::Else => {
+ self.label("# else");
+ let if_counter = if_stack.last().unwrap().clone();
+ self.line(format!("j _endif_{}", if_counter));
+ self.label(format!("_else_{}:", if_counter));
+ seen_else.insert(if_counter);
+ },
+ IR::EndIf => {
+ self.label("# endif");
+ let stack = &mut if_stack;
+ let if_counter = stack.last().unwrap().clone();
+ if !seen_else.contains(&if_counter) {
+ self.label(format!("_else_{}:", if_counter));
+ } else {
+ self.label(format!("_endif_{}:", if_counter));
+ seen_else.remove(&if_counter);
+ }
+ stack.pop();
+ },
+ IR::Loop => { // keep looping until is true/0
+ self.label(format!("_loop_{}:", loop_count));
+ self.pop_to("t0");
+ self.line(format!("beqz t0, _endloop_{}", loop_count));
+ loop_stack.push(loop_count);
+ loop_count += 1;
+ },
+ IR::EndLoop => {
+ let stack = &mut loop_stack;
+ let loop_counter = stack.last().unwrap().clone();
+ self.line(format!("j _loop_{}", loop_counter));
+ self.label(format!("_endloop_{}:", loop_counter));
+ stack.pop();
+ },
+ _ => bail!("not implemented yet: {:?}", ir),
+ }
+ }
+
+ Ok(self.lines.join("\n"))
+ }
+}
+
--- /dev/null
+use anyhow::{Result, anyhow};
+
+#[derive(Debug, Clone)]
+pub enum Token<'a> {
+ Word(&'a str),
+ String(&'a str),
+ NumU8(u8),
+ NumI8(i8),
+ NumU16(u16),
+ NumI16(i16),
+ NumU32(u32),
+ NumI32(i32),
+ NumU64(u64),
+ NumI64(i64),
+ NumF32(f32),
+ NumF64(f64),
+}
+
+impl<'a> Token<'a>{
+ fn parse_word_or_num(input: &'a str) -> Result<Token<'a>> {
+ if input == "-" {
+ return Ok(Token::Word(input))
+ }
+
+ // we're assuming any token starting with `-` with length greater than one
+ // is a negative number
+ if input.starts_with('-') || input.chars().nth(0).map(|x| x.is_numeric()).unwrap_or(false) {
+ if input.contains(':') {
+ let mut splat = input.split(':');
+ let num = splat.next().ok_or(anyhow!("no number found"))?;
+ let typ = splat.next().ok_or(anyhow!("no number type found"))?;
+ match typ {
+ "u8" => Ok(Token::NumU8(num.parse()?)),
+ "i8" => Ok(Token::NumI8(num.parse()?)),
+ "u16" => Ok(Token::NumU16(num.parse()?)),
+ "i16" => Ok(Token::NumI16(num.parse()?)),
+ "u32" => Ok(Token::NumU32(num.parse()?)),
+ "i32" => Ok(Token::NumI32(num.parse()?)),
+ "u64" => Ok(Token::NumU64(num.parse()?)),
+ "i64" => Ok(Token::NumI64(num.parse()?)),
+ "f32" => Ok(Token::NumF32(num.parse()?)),
+ "f64" => Ok(Token::NumF64(num.parse()?)),
+ _ => panic!("unknown number type")
+ }
+ } else {
+ if input.contains('.') {
+ Ok(Token::NumF64(input.parse()?))
+ } else if input.starts_with('-') {
+ Ok(Token::NumI64(input.parse()?))
+ } else {
+ Ok(Token::NumU64(input.parse()?))
+ }
+ }
+ } else {
+ Ok(Token::Word(input))
+ }
+ }
+}
+
+// TODO really want an iterator, not a vector
+pub fn tokenize<'a>(input: &'a str) -> Result<Vec<Token<'a>>> {
+ let mut result = vec![];
+ let mut string_start: Option<usize> = None;
+ let mut word_or_num_start: Option<usize> = None;
+ let mut last_is_backslash = false;
+ let mut last_is_whitespace = true;
+ let mut in_doc_comment = false;
+ let mut in_line_comment = false;
+ let mut index = 0;
+ let mut first_char = true;
+
+
+ for char in input.chars() {
+ if first_char {
+ first_char = false;
+ } else {
+ index += 1;
+ }
+
+ if in_doc_comment {
+ if char == ')' {
+ in_doc_comment = false;
+ last_is_whitespace = true; // not really true, but means don't need space after
+ }
+ continue;
+ }
+
+ if in_line_comment {
+ word_or_num_start = None;
+ if char == '\n' {
+ in_line_comment = false;
+ last_is_whitespace = true; // not really true, but means don't need space after
+ }
+ continue;
+ }
+
+ if char == '"' {
+ if let Some(start) = string_start {
+ if !last_is_backslash {
+ result.push(Token::String(&input[start..index]));
+ string_start = None;
+ }
+ } else {
+ string_start = Some(index + 1)
+ }
+ last_is_backslash = false;
+ last_is_whitespace = false;
+ continue;
+ }
+
+
+ if string_start.is_some() {
+ last_is_backslash = char == '\\';
+ continue;
+ }
+
+ if char.is_whitespace() {
+ if last_is_backslash {
+ in_line_comment = true;
+ } else if !last_is_whitespace && let Some(start) = word_or_num_start {
+ let token = &input[start..index];
+ if token == "(" {
+ in_doc_comment = true;
+ } else {
+ result.push(Token::parse_word_or_num(&input[start..index])?);
+ }
+ word_or_num_start = None;
+ }
+ last_is_whitespace = true;
+ last_is_backslash = false;
+ continue;
+ }
+
+ last_is_backslash = char == '\\';
+
+ if index == input.len() - 1 {
+ if !last_is_whitespace && let Some(start) = word_or_num_start {
+ result.push(Token::parse_word_or_num(&input[start..])?);
+ }
+ continue;
+ }
+
+ if last_is_whitespace { // start of word or num (we already handled strings)
+ word_or_num_start = Some(index);
+ last_is_whitespace = false;
+ }
+ }
+ Ok(result)
+}
+
+#[cfg(test)]
+mod tests {
+ use super::*;
+
+ #[test]
+ fn try_some_tokenizing() {
+ let result = tokenize("
+
+ \\ soup
+ 2 3.4 - -88 bacon \"hello\" 43:f32 2345:u32 -57:i8 soup
+");
+ println!("result: {:?}", result);
+ }
+
+ #[test]
+ fn comments() {
+ let result = tokenize("
+ (
+ foo
+ bar
+ )
+ : baz ( x y -- z )
+ chicken
+ soup
+ ;
+ ");
+ println!("result: {:?}", result);
+ }
+
+ #[test]
+ fn strings() {
+ let result = tokenize("
+dup \\ ( stuff )
+\"hello!\"
+");
+ println!("result: {:?}", result);
+ }
+}
--- /dev/null
+\ vim: filetype=forth
+
+: mmap 222 sys6 ;
+: munmap 215 sys2 ;
+
+: PROT_READ 1 ;
+: PROT_WRITE 2 ;
+: MAP_PRIVATE 2 ;
+: MAP_ANONYMOUS 32 ;
+
+: ALLOC_PROT PROT_READ PROT_WRITE | ;
+: ALLOC_MAP MAP_PRIVATE MAP_ANONYMOUS | ;
+
+\ This is a raw way to allocate memory. Callers to this will need to know
+\ how much memory was allocated, to call dealloc later.
+: alloc_raw ( byte-count -- addr )
+ 0 swap ALLOC_PROT ALLOC_MAP -1:i32 0 mmap
+;
+
+: dealloc ( addr byte-count -- )
+ munmap
+;
+
+export alloc
+: alloc ( byte-count -- addr )
+ dup \ we'll need the length again later
+ 8 + \ add 64 bits to the front where we'll store the length
+ alloc_raw \ ( byte-count addr )
+ dup \ ( byte-count addr addr)
+ rot \ ( addr addr byte-count )
+ rot \ ( addr byte-count addr )
+ ! \ write the length to the addr, now ( addr )
+ 8 + \ here's the address we'll return, 8 bytes forward to account for the length
+;
+
+export free
+: free ( addr -- )
+ 8 - \ the real memory start is 8 bytes before
+ dup \ ( addr addr )
+ @ \ ( addr byte-length )
+ 8 + \ add 8 to the byte length to refer to the whole region from alloc.
+ dealloc
+;
+
+import "std:out"
+
+: copy_byte ( addr_from addr_to -- addr_from addr_to )
+ swap \ ( addr_to addr_from )
+ dup \ ( addr_to addr_from addr_from )
+ @:8 \ ( addr_to addr_from byte )
+ rot \ ( addr_from byte addr_to )
+ dup \ ( addr_from byte addr_to addr_to)
+ rot \ ( addr_from addr_to addr_to byte )
+ swap \ ( addr_from addr_to byte addr_to )
+ !:8 \ ( addr_from addr_to )
+;
+
+export memcopy
+: memcopy ( addr_from addr_to count -- )
+ dup \ ( addr_from addr_to count count )
+ loop \ ( addr_from addr_to count )
+ rot \ ( addr_to count addr_from )
+ rot \ ( count addr_from addr_to )
+ copy_byte \ ( count addr_from addr_to )
+ 1 + \ ( count addr_from addr_to+1 )
+ rot \ ( addr_from addr_to+1 count )
+ 1 - \ ( addr_from addr_to+1 count-1 )
+ rot \ ( addr_to+1 count-1 addr_from )
+ 1 + \ ( addr_to+1 count-1 addr_from+1 )
+ rot rot \ ( addr_from+1 addr_to+1 count-1 )
+ dup \ ( addr_from+1 addr_to+1 count-1 count-1)
+ endloop \ ( addr_from+count addr_to+count 0 )
+ drop drop drop \ ( )
+;
--- /dev/null
+\ vim: filetype=forth
+
+: write ( fd ptr len -- bytes-written-or-err )
+ 64
+ sys3
+;
+
+import "std:string"
+
+export puts
+
+: puts ( addr -- addr)
+ dup dup \ ( addr addr addr )
+ strlen \ ( addr addr len )
+ 1 \ ( addr addr len 1 )
+ rot \ ( addr len 1 addr )
+ rot \ ( addr 1 addr len )
+ write \ ( addr bytes-written-or-err )
+ drop \ ( addr )
+;
+
+: ZERO_CHAR 48 ;
+: NEWLINE_CHAR 10 ;
+
+export putn
+
+: putn ( num -- num )
+ dup dup \ ( num num num )
+ 10 rot rot \ ( num 10 num num ) // Mutltidigit stop point
+ 10 < \ ( num 10 num is<10 )
+ loop \ ( num 10 num' )
+ dup \ ( num 10 num' num' )
+ 10 % \ ( num 10 num' digit )
+ swap \ ( num 10 digit num' )
+ 10 / \ ( num 10 digit num'/10 )
+ dup 10 < \ ( num 10 digit num'/10 is<10 )
+ endloop \ ( num 10 digitn ... digit1 )
+ dup 9 > \ ( num 10 digitn ... digit1 digit1>9 )
+ loop \ ( num 10 digitn ... digit1 )
+ ZERO_CHAR + \ \ ( num 10 digitn ... digit1 )
+ sp \ ( num 10 digitn ... digit1 ptr )
+ 1 \ ( num 10 digitn ... digit1 ptr 1 )
+ swap \ ( num 10 digitn ... digit1 1 ptr )
+ 1 \ ( num 10 digitn ... digit1 1 ptr 1 )
+ write \ ( num 10 digitn ... digit1 result )
+ drop drop \ ( num 10 digitn ... digit2 )
+ dup 9 > \ ( num 10 digitn ... digit2 digit2>9 )
+ endloop \ ( num 10 )
+ drop \ ( num )
+ NEWLINE_CHAR \ ( num 10 )
+ sp \ ( num 10 ptr )
+ 1 \ ( num 10 ptr 1 )
+ swap \ (num 10 1 ptr )
+ 1 \ ( num 10 1 ptr 1 )
+ write \ ( num 10 result )
+ drop \ ( num 10 )
+ drop \ ( num )
+;
+
--- /dev/null
+\ vim: filetype=forth
+
+export exit
+
+: exit ( code -- )
+ 93
+ sys1
+;
--- /dev/null
+\ vim: filetype=forth
+
+: strlen ( addr -- len )
+ dup dup \ ( addr addr addr )
+ @:8 \ ( addr addr byte )
+ loop \ ( addr addr )
+ 1 + \ ( addr addr+1 )
+ dup \ ( addr addr+1 addr+1)
+ @:8 \ ( addr addr+1 byte )
+ endloop \ ( addr addr+len )
+ swap \ ( addr+len addr )
+ - \ ( len )
+;
+
+export strlen
+++ /dev/null
- Finished `dev` profile [unoptimized + debuginfo] target(s) in 0.03s
+++ /dev/null
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#proc-macro2@1.0.103","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/proc-macro2-1.0.103/Cargo.toml","target":{"kind":["custom-build"],"crate_types":["bin"],"name":"build-script-build","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/proc-macro2-1.0.103/build.rs","edition":"2021","doc":false,"doctest":false,"test":false},"profile":{"opt_level":"0","debuginfo":0,"debug_assertions":true,"overflow_checks":true,"test":false},"features":["default","proc-macro"],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/build/proc-macro2-bdf9ed00dad30299/build-script-build"],"executable":null,"fresh":true}
-{"reason":"build-script-executed","package_id":"registry+https://github.com/rust-lang/crates.io-index#proc-macro2@1.0.103","linked_libs":[],"linked_paths":[],"cfgs":["wrap_proc_macro","proc_macro_span_location","proc_macro_span_file"],"env":[],"out_dir":"/home/bengl/rethought/hylo-lang/target/debug/build/proc-macro2-01b6d2c5fdefd348/out"}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#unicode-ident@1.0.22","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/unicode-ident-1.0.22/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"unicode_ident","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/unicode-ident-1.0.22/src/lib.rs","edition":"2018","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":0,"debug_assertions":true,"overflow_checks":true,"test":false},"features":[],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libunicode_ident-4fec18d14b15363e.rlib","/home/bengl/rethought/hylo-lang/target/debug/deps/libunicode_ident-4fec18d14b15363e.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#quote@1.0.42","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/quote-1.0.42/Cargo.toml","target":{"kind":["custom-build"],"crate_types":["bin"],"name":"build-script-build","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/quote-1.0.42/build.rs","edition":"2018","doc":false,"doctest":false,"test":false},"profile":{"opt_level":"0","debuginfo":0,"debug_assertions":true,"overflow_checks":true,"test":false},"features":["default","proc-macro"],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/build/quote-03d3df1af8178ab4/build-script-build"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#proc-macro2@1.0.103","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/proc-macro2-1.0.103/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"proc_macro2","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/proc-macro2-1.0.103/src/lib.rs","edition":"2021","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":0,"debug_assertions":true,"overflow_checks":true,"test":false},"features":["default","proc-macro"],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libproc_macro2-696ff2d61f065b42.rlib","/home/bengl/rethought/hylo-lang/target/debug/deps/libproc_macro2-696ff2d61f065b42.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#serde_core@1.0.228","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/serde_core-1.0.228/Cargo.toml","target":{"kind":["custom-build"],"crate_types":["bin"],"name":"build-script-build","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/serde_core-1.0.228/build.rs","edition":"2021","doc":false,"doctest":false,"test":false},"profile":{"opt_level":"0","debuginfo":0,"debug_assertions":true,"overflow_checks":true,"test":false},"features":["result","std"],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/build/serde_core-31b971cf163f0f6c/build-script-build"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#serde@1.0.228","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/serde-1.0.228/Cargo.toml","target":{"kind":["custom-build"],"crate_types":["bin"],"name":"build-script-build","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/serde-1.0.228/build.rs","edition":"2021","doc":false,"doctest":false,"test":false},"profile":{"opt_level":"0","debuginfo":0,"debug_assertions":true,"overflow_checks":true,"test":false},"features":["default","derive","serde_derive","std"],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/build/serde-49a10a9683562367/build-script-build"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#hashbrown@0.16.1","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/hashbrown-0.16.1/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"hashbrown","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/hashbrown-0.16.1/src/lib.rs","edition":"2021","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":false},"features":[],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libhashbrown-3d3c587f4038bc13.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#equivalent@1.0.2","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/equivalent-1.0.2/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"equivalent","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/equivalent-1.0.2/src/lib.rs","edition":"2015","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":false},"features":[],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libequivalent-9b88d636aa69c8ed.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#unsafe-libyaml@0.2.11","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/unsafe-libyaml-0.2.11/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"unsafe_libyaml","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/unsafe-libyaml-0.2.11/src/lib.rs","edition":"2021","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":false},"features":[],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libunsafe_libyaml-5d3cabcbf60870d4.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#itoa@1.0.15","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/itoa-1.0.15/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"itoa","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/itoa-1.0.15/src/lib.rs","edition":"2018","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":false},"features":[],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libitoa-bc1ca429a31fb99c.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#ryu@1.0.20","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/ryu-1.0.20/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"ryu","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/ryu-1.0.20/src/lib.rs","edition":"2018","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":false},"features":[],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libryu-e3abc3dffcbd16d9.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#syscalls@0.7.0","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/syscalls-0.7.0/Cargo.toml","target":{"kind":["custom-build"],"crate_types":["bin"],"name":"build-script-build","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/syscalls-0.7.0/build.rs","edition":"2021","doc":false,"doctest":false,"test":false},"profile":{"opt_level":"0","debuginfo":0,"debug_assertions":true,"overflow_checks":true,"test":false},"features":["default","serde","serde_repr","std"],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/build/syscalls-fd98ea5acf3ec59b/build-script-build"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#anyhow@1.0.100","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/anyhow-1.0.100/Cargo.toml","target":{"kind":["custom-build"],"crate_types":["bin"],"name":"build-script-build","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/anyhow-1.0.100/build.rs","edition":"2018","doc":false,"doctest":false,"test":false},"profile":{"opt_level":"0","debuginfo":0,"debug_assertions":true,"overflow_checks":true,"test":false},"features":["default","std"],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/build/anyhow-f0f8ac34947eb6de/build-script-build"],"executable":null,"fresh":true}
-{"reason":"build-script-executed","package_id":"registry+https://github.com/rust-lang/crates.io-index#quote@1.0.42","linked_libs":[],"linked_paths":[],"cfgs":[],"env":[],"out_dir":"/home/bengl/rethought/hylo-lang/target/debug/build/quote-4fccd938ab7a0173/out"}
-{"reason":"build-script-executed","package_id":"registry+https://github.com/rust-lang/crates.io-index#serde_core@1.0.228","linked_libs":[],"linked_paths":[],"cfgs":[],"env":[],"out_dir":"/home/bengl/rethought/hylo-lang/target/debug/build/serde_core-7b7cb0cfdf46fc20/out"}
-{"reason":"build-script-executed","package_id":"registry+https://github.com/rust-lang/crates.io-index#serde@1.0.228","linked_libs":[],"linked_paths":[],"cfgs":["if_docsrs_then_no_serde_core"],"env":[],"out_dir":"/home/bengl/rethought/hylo-lang/target/debug/build/serde-0c79bc1bb5bf9eba/out"}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#indexmap@2.12.1","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/indexmap-2.12.1/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"indexmap","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/indexmap-2.12.1/src/lib.rs","edition":"2021","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":false},"features":["default","std"],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libindexmap-274ebdc57a13914e.rmeta"],"executable":null,"fresh":true}
-{"reason":"build-script-executed","package_id":"registry+https://github.com/rust-lang/crates.io-index#anyhow@1.0.100","linked_libs":[],"linked_paths":[],"cfgs":["std_backtrace"],"env":[],"out_dir":"/home/bengl/rethought/hylo-lang/target/debug/build/anyhow-7dffd08ca73f1c01/out"}
-{"reason":"build-script-executed","package_id":"registry+https://github.com/rust-lang/crates.io-index#syscalls@0.7.0","linked_libs":[],"linked_paths":[],"cfgs":[],"env":[],"out_dir":"/home/bengl/rethought/hylo-lang/target/debug/build/syscalls-45c42af557e79235/out"}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#quote@1.0.42","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/quote-1.0.42/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"quote","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/quote-1.0.42/src/lib.rs","edition":"2018","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":0,"debug_assertions":true,"overflow_checks":true,"test":false},"features":["default","proc-macro"],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libquote-49a93d0cff3d7298.rlib","/home/bengl/rethought/hylo-lang/target/debug/deps/libquote-49a93d0cff3d7298.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#serde_core@1.0.228","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/serde_core-1.0.228/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"serde_core","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/serde_core-1.0.228/src/lib.rs","edition":"2021","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":false},"features":["result","std"],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libserde_core-a2045ddd2bbb2331.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#anyhow@1.0.100","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/anyhow-1.0.100/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"anyhow","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/anyhow-1.0.100/src/lib.rs","edition":"2018","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":false},"features":["default","std"],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libanyhow-a8e29080dfa88f6a.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#syn@2.0.111","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/syn-2.0.111/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"syn","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/syn-2.0.111/src/lib.rs","edition":"2021","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":0,"debug_assertions":true,"overflow_checks":true,"test":false},"features":["clone-impls","default","derive","parsing","printing","proc-macro"],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libsyn-d395eb58cbb9ffa6.rlib","/home/bengl/rethought/hylo-lang/target/debug/deps/libsyn-d395eb58cbb9ffa6.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#serde_derive@1.0.228","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/serde_derive-1.0.228/Cargo.toml","target":{"kind":["proc-macro"],"crate_types":["proc-macro"],"name":"serde_derive","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/serde_derive-1.0.228/src/lib.rs","edition":"2021","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":0,"debug_assertions":true,"overflow_checks":true,"test":false},"features":["default"],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libserde_derive-dca8d717b71cd876.so"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#serde_repr@0.1.20","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/serde_repr-0.1.20/Cargo.toml","target":{"kind":["proc-macro"],"crate_types":["proc-macro"],"name":"serde_repr","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/serde_repr-0.1.20/src/lib.rs","edition":"2021","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":0,"debug_assertions":true,"overflow_checks":true,"test":false},"features":[],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libserde_repr-1a41abf411086f19.so"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#serde@1.0.228","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/serde-1.0.228/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"serde","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/serde-1.0.228/src/lib.rs","edition":"2021","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":false},"features":["default","derive","serde_derive","std"],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libserde-a90341a4e2031dae.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#serde_yaml@0.9.34+deprecated","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/serde_yaml-0.9.34+deprecated/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"serde_yaml","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/serde_yaml-0.9.34+deprecated/src/lib.rs","edition":"2021","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":false},"features":[],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libserde_yaml-22ed54ad12da1e59.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"registry+https://github.com/rust-lang/crates.io-index#syscalls@0.7.0","manifest_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/syscalls-0.7.0/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"syscalls","src_path":"/home/bengl/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/syscalls-0.7.0/src/lib.rs","edition":"2021","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":false},"features":["default","serde","serde_repr","std"],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libsyscalls-7a60ab2d0d4883d8.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"path+file:///home/bengl/rethought/hylo-lang/hylo-ir#0.1.0","manifest_path":"/home/bengl/rethought/hylo-lang/hylo-ir/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"hylo_ir","src_path":"/home/bengl/rethought/hylo-lang/hylo-ir/src/lib.rs","edition":"2024","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":false},"features":[],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libhylo_ir-4b38e4623b5ebbc5.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"path+file:///home/bengl/rethought/hylo-lang/hylo-ir#0.1.0","manifest_path":"/home/bengl/rethought/hylo-lang/hylo-ir/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"hylo_ir","src_path":"/home/bengl/rethought/hylo-lang/hylo-ir/src/lib.rs","edition":"2024","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":true},"features":[],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libhylo_ir-6637db13cb880a28.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-message","package_id":"path+file:///home/bengl/rethought/hylo-lang/hylo-interpret#0.1.0","manifest_path":"/home/bengl/rethought/hylo-lang/hylo-interpret/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"hylo_interpret","src_path":"/home/bengl/rethought/hylo-lang/hylo-interpret/src/lib.rs","edition":"2024","doc":true,"doctest":true,"test":true},"message":{"rendered":"warning: field `strings` is never read\n --> hylo-interpret/src/lib.rs:14:5\n |\n 8 | pub struct Interpreter<'a> {\n | ----------- field in this struct\n...\n14 | strings: Vec<String>,\n | ^^^^^^^\n |\n = note: `#[warn(dead_code)]` (part of `#[warn(unused)]`) on by default\n\n","$message_type":"diagnostic","children":[{"children":[],"code":null,"level":"note","message":"`#[warn(dead_code)]` (part of `#[warn(unused)]`) on by default","rendered":null,"spans":[]}],"level":"warning","message":"field `strings` is never read","spans":[{"byte_end":119,"byte_start":108,"column_end":23,"column_start":12,"expansion":null,"file_name":"hylo-interpret/src/lib.rs","is_primary":false,"label":"field in this struct","line_end":8,"line_start":8,"suggested_replacement":null,"suggestion_applicability":null,"text":[{"highlight_end":23,"highlight_start":12,"text":"pub struct Interpreter<'a> {"}]},{"byte_end":287,"byte_start":280,"column_end":12,"column_start":5,"expansion":null,"file_name":"hylo-interpret/src/lib.rs","is_primary":true,"label":null,"line_end":14,"line_start":14,"suggested_replacement":null,"suggestion_applicability":null,"text":[{"highlight_end":12,"highlight_start":5,"text":" strings: Vec<String>,"}]}],"code":{"code":"dead_code","explanation":null}}}
-{"reason":"compiler-artifact","package_id":"path+file:///home/bengl/rethought/hylo-lang/hylo-interpret#0.1.0","manifest_path":"/home/bengl/rethought/hylo-lang/hylo-interpret/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"hylo_interpret","src_path":"/home/bengl/rethought/hylo-lang/hylo-interpret/src/lib.rs","edition":"2024","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":false},"features":[],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libhylo_interpret-3e4441b288f50835.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-message","package_id":"path+file:///home/bengl/rethought/hylo-lang/hylo-interpret#0.1.0","manifest_path":"/home/bengl/rethought/hylo-lang/hylo-interpret/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"hylo_interpret","src_path":"/home/bengl/rethought/hylo-lang/hylo-interpret/src/lib.rs","edition":"2024","doc":true,"doctest":true,"test":true},"message":{"rendered":"warning: field `strings` is never read\n --> hylo-interpret/src/lib.rs:14:5\n |\n 8 | pub struct Interpreter<'a> {\n | ----------- field in this struct\n...\n14 | strings: Vec<String>,\n | ^^^^^^^\n |\n = note: `#[warn(dead_code)]` (part of `#[warn(unused)]`) on by default\n\n","$message_type":"diagnostic","children":[{"children":[],"code":null,"level":"note","message":"`#[warn(dead_code)]` (part of `#[warn(unused)]`) on by default","rendered":null,"spans":[]}],"level":"warning","message":"field `strings` is never read","spans":[{"byte_end":119,"byte_start":108,"column_end":23,"column_start":12,"expansion":null,"file_name":"hylo-interpret/src/lib.rs","is_primary":false,"label":"field in this struct","line_end":8,"line_start":8,"suggested_replacement":null,"suggestion_applicability":null,"text":[{"highlight_end":23,"highlight_start":12,"text":"pub struct Interpreter<'a> {"}]},{"byte_end":287,"byte_start":280,"column_end":12,"column_start":5,"expansion":null,"file_name":"hylo-interpret/src/lib.rs","is_primary":true,"label":null,"line_end":14,"line_start":14,"suggested_replacement":null,"suggestion_applicability":null,"text":[{"highlight_end":12,"highlight_start":5,"text":" strings: Vec<String>,"}]}],"code":{"code":"dead_code","explanation":null}}}
-{"reason":"compiler-artifact","package_id":"path+file:///home/bengl/rethought/hylo-lang/hylo-interpret#0.1.0","manifest_path":"/home/bengl/rethought/hylo-lang/hylo-interpret/Cargo.toml","target":{"kind":["lib"],"crate_types":["lib"],"name":"hylo_interpret","src_path":"/home/bengl/rethought/hylo-lang/hylo-interpret/src/lib.rs","edition":"2024","doc":true,"doctest":true,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":true},"features":[],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libhylo_interpret-8c8a8fc209c55986.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"path+file:///home/bengl/rethought/hylo-lang/hyloc#0.1.0","manifest_path":"/home/bengl/rethought/hylo-lang/hyloc/Cargo.toml","target":{"kind":["bin"],"crate_types":["bin"],"name":"hyloc","src_path":"/home/bengl/rethought/hylo-lang/hyloc/src/main.rs","edition":"2024","doc":true,"doctest":false,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":false},"features":[],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libhyloc-58f075f82bafc9dc.rmeta"],"executable":null,"fresh":true}
-{"reason":"compiler-artifact","package_id":"path+file:///home/bengl/rethought/hylo-lang/hyloc#0.1.0","manifest_path":"/home/bengl/rethought/hylo-lang/hyloc/Cargo.toml","target":{"kind":["bin"],"crate_types":["bin"],"name":"hyloc","src_path":"/home/bengl/rethought/hylo-lang/hyloc/src/main.rs","edition":"2024","doc":true,"doctest":false,"test":true},"profile":{"opt_level":"0","debuginfo":2,"debug_assertions":true,"overflow_checks":true,"test":true},"features":[],"filenames":["/home/bengl/rethought/hylo-lang/target/debug/deps/libhyloc-4df3a47409d99128.rmeta"],"executable":null,"fresh":true}
-{"reason":"build-finished","success":true}
--- /dev/null
+\ vim: filetype=forth
+
+import "std:process"
+
+export assert
+
+: assert ( x -- )
+ if
+ drop
+ else
+ 1 exit
+ endif
+;
--- /dev/null
+#include <stdio.h>
+
+extern unsigned long data_stack_end;
+register unsigned long * stack_pointer asm("s2");
+
+void putstack() {
+ unsigned long * stack_index = &data_stack_end;
+ printf("stack: ");
+ while (stack_index != stack_pointer) {
+ printf("%ld ", *stack_index);
+ stack_index -= 1;
+ }
+ printf("%ld\n", *stack_pointer);
+}
+
+
--- /dev/null
+UNAME=$(uname -m)
+CMD_PREFIX=$([ "$UNAME" = "riscv64" ] && echo "" || echo "riscv64-unknown-linux-gnu-")
+AS="${CMD_PREFIX}as"
+LD="${CMD_PREFIX}ld"
+CC="${CMD_PREFIX}cc"
+
+../target/debug/sorelc test1.sorel
+$AS -g -o test1.o test1.asm
+$CC -O1 -no-pie -o test1.out test1.o putstack.c -nostartfiles
+# $LD -o test1.out test1.o
+./test1.out
--- /dev/null
+\ vim: filetype=forth
+
+import "./assert.sorel"
+import "std:out"
+
+"Hello, World! \n" puts
+drop
+
+42 putn
+drop
+
+import "std:mem"
+
+64 alloc
+free
+
+0 assert
+
+import "std:string"
+import "std:mem"
+import "std:out"
+
+extern putstack
+
+16 alloc \ ( ptr )
+dup \ ( ptr ptr )
+"To be copied!" \ ( ptr ptr strptr )
+dup \ ( ptr ptr strptr strptr )
+strlen 1 + \ ( ptr ptr strptr len )
+rot \ ( ptr strptr len ptr )
+swap \ ( ptr strptr ptr len )
+memcopy \ ( ptr )
+puts \ ( ptr )
+free \ ( )
+
+
projects / sorel-lang.git / commitdiff