Documented hbasm

Reviewed-on: #17

.cargo/config.toml

@@ -0,0 +1,2 @@
+[alias]
+xtask = "r -p xtask --"

Cargo.lock

@@ -3,14 +3,105 @@
 version = 3
 
 [[package]]
-name = "ahash"
-version = "0.8.3"
+name = "addr2line"
+version = "0.21.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2c99f64d1e06488f620f932677e24bc6e2897582980441ae90a671415bd7ec2f"
+checksum = "8a30b2e23b9e17a9f90641c7ab1549cd9b44f296d3ccbf309d2863cfe398a0cb"
+dependencies = [
+ "gimli",
+]
+
+[[package]]
+name = "adler"
+version = "1.0.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f26201604c87b1e01bd3d98f8d5d9a8fcbb815e8cedb41ffccbeb4bf593a35fe"
+
+[[package]]
+name = "ahash"
+version = "0.8.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "42cd52102d3df161c77a887b608d7a4897d7cc112886a9537b738a887a03aaff"
 dependencies = [
  "cfg-if",
+ "const-random",
+ "getrandom",
  "once_cell",
  "version_check",
+ "zerocopy",
+]
+
+[[package]]
+name = "argh"
+version = "0.1.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7af5ba06967ff7214ce4c7419c7d185be7ecd6cc4965a8f6e1d8ce0398aad219"
+dependencies = [
+ "argh_derive",
+ "argh_shared",
+]
+
+[[package]]
+name = "argh_derive"
+version = "0.1.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "56df0aeedf6b7a2fc67d06db35b09684c3e8da0c95f8f27685cb17e08413d87a"
+dependencies = [
+ "argh_shared",
+ "proc-macro2",
+ "quote",
+ "syn 2.0.48",
+]
+
+[[package]]
+name = "argh_shared"
+version = "0.1.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5693f39141bda5760ecc4111ab08da40565d1771038c4a0250f03457ec707531"
+dependencies = [
+ "serde",
+]
+
+[[package]]
+name = "autocfg"
+version = "1.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d468802bab17cbc0cc575e9b053f41e72aa36bfa6b7f55e3529ffa43161b97fa"
+
+[[package]]
+name = "backtrace"
+version = "0.3.69"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "2089b7e3f35b9dd2d0ed921ead4f6d318c27680d4a5bd167b3ee120edb105837"
+dependencies = [
+ "addr2line",
+ "cc",
+ "cfg-if",
+ "libc",
+ "miniz_oxide",
+ "object",
+ "rustc-demangle",
+]
+
+[[package]]
+name = "beef"
+version = "0.5.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3a8241f3ebb85c056b509d4327ad0358fbbba6ffb340bf388f26350aeda225b1"
+
+[[package]]
+name = "bitflags"
+version = "2.4.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ed570934406eb16438a4e976b1b4500774099c13b8cb96eec99f620f05090ddf"
+
+[[package]]
+name = "cc"
+version = "1.0.83"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f1174fb0b6ec23863f8b971027804a42614e347eafb0a95bf0b12cdae21fc4d0"
+dependencies = [
+ "libc",
 ]
 
 [[package]]
@@ -20,43 +111,542 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"
 
 [[package]]
-name = "compiler"
-version = "0.1.0"
+name = "color-eyre"
+version = "0.6.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5a667583cca8c4f8436db8de46ea8233c42a7d9ae424a82d338f2e4675229204"
+dependencies = [
+ "backtrace",
+ "color-spantrace",
+ "eyre",
+ "indenter",
+ "once_cell",
+ "owo-colors",
+ "tracing-error",
+]
 
 [[package]]
-name = "hashbrown"
-version = "0.13.2"
+name = "color-spantrace"
+version = "0.2.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "43a3c133739dddd0d2990f9a4bdf8eb4b21ef50e4851ca85ab661199821d510e"
+checksum = "cd6be1b2a7e382e2b98b43b2adcca6bb0e465af0bdd38123873ae61eb17a72c2"
 dependencies = [
- "ahash",
+ "once_cell",
+ "owo-colors",
+ "tracing-core",
+ "tracing-error",
+]
+
+[[package]]
+name = "const-random"
+version = "0.1.17"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5aaf16c9c2c612020bcfd042e170f6e32de9b9d75adb5277cdbbd2e2c8c8299a"
+dependencies = [
+ "const-random-macro",
+]
+
+[[package]]
+name = "const-random-macro"
+version = "0.1.16"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f9d839f2a20b0aee515dc581a6172f2321f96cab76c1a38a4c584a194955390e"
+dependencies = [
+ "getrandom",
+ "once_cell",
+ "tiny-keccak",
+]
+
+[[package]]
+name = "crunchy"
+version = "0.2.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7a81dae078cea95a014a339291cec439d2f232ebe854a9d672b796c6afafa9b7"
+
+[[package]]
+name = "eyre"
+version = "0.6.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7cd915d99f24784cdc19fd37ef22b97e3ff0ae756c7e492e9fbfe897d61e2aec"
+dependencies = [
+ "indenter",
+ "once_cell",
+]
+
+[[package]]
+name = "fnv"
+version = "1.0.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3f9eec918d3f24069decb9af1554cad7c880e2da24a9afd88aca000531ab82c1"
+
+[[package]]
+name = "getrandom"
+version = "0.2.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "190092ea657667030ac6a35e305e62fc4dd69fd98ac98631e5d3a2b1575a12b5"
+dependencies = [
+ "cfg-if",
+ "libc",
+ "wasi",
+]
+
+[[package]]
+name = "gimli"
+version = "0.28.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4271d37baee1b8c7e4b708028c57d816cf9d2434acb33a549475f78c181f6253"
+
+[[package]]
+name = "hbasm"
+version = "0.1.0"
+dependencies = [
+ "paste",
+ "rhai",
+ "with_builtin_macros",
+]
+
+[[package]]
+name = "hbbytecode"
+version = "0.1.0"
+dependencies = [
+ "paste",
+ "with_builtin_macros",
+]
+
+[[package]]
+name = "hblang"
+version = "0.1.0"
+dependencies = [
+ "hbbytecode",
+ "logos",
 ]
 
 [[package]]
 name = "hbvm"
 version = "0.1.0"
 dependencies = [
- "hashbrown",
- "log",
+ "hbbytecode",
 ]
 
 [[package]]
-name = "log"
-version = "0.4.17"
+name = "hbvm_aos_on_linux"
+version = "0.1.0"
+dependencies = [
+ "hbvm",
+ "nix",
+]
+
+[[package]]
+name = "hbxrt"
+version = "0.1.0"
+dependencies = [
+ "hbvm",
+ "memmap2",
+]
+
+[[package]]
+name = "indenter"
+version = "0.3.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "abb12e687cfb44aa40f41fc3978ef76448f9b6038cad6aef4259d3c095a2382e"
+checksum = "ce23b50ad8242c51a442f3ff322d56b02f08852c77e4c0b4d3fd684abc89c683"
+
+[[package]]
+name = "instant"
+version = "0.1.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7a5bbe824c507c5da5956355e86a746d82e0e1464f65d862cc5e71da70e94b2c"
 dependencies = [
  "cfg-if",
 ]
 
+[[package]]
+name = "lazy_static"
+version = "1.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e2abad23fbc42b3700f2f279844dc832adb2b2eb069b2df918f455c4e18cc646"
+
+[[package]]
+name = "libc"
+version = "0.2.153"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9c198f91728a82281a64e1f4f9eeb25d82cb32a5de251c6bd1b5154d63a8e7bd"
+
+[[package]]
+name = "logos"
+version = "0.13.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c000ca4d908ff18ac99b93a062cb8958d331c3220719c52e77cb19cc6ac5d2c1"
+dependencies = [
+ "logos-derive",
+]
+
+[[package]]
+name = "logos-codegen"
+version = "0.13.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "dc487311295e0002e452025d6b580b77bb17286de87b57138f3b5db711cded68"
+dependencies = [
+ "beef",
+ "fnv",
+ "proc-macro2",
+ "quote",
+ "regex-syntax",
+ "syn 2.0.48",
+]
+
+[[package]]
+name = "logos-derive"
+version = "0.13.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "dbfc0d229f1f42d790440136d941afd806bc9e949e2bcb8faa813b0f00d1267e"
+dependencies = [
+ "logos-codegen",
+]
+
+[[package]]
+name = "memchr"
+version = "2.7.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "523dc4f511e55ab87b694dc30d0f820d60906ef06413f93d4d7a1385599cc149"
+
+[[package]]
+name = "memmap2"
+version = "0.9.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fe751422e4a8caa417e13c3ea66452215d7d63e19e604f4980461212f3ae1322"
+dependencies = [
+ "libc",
+]
+
+[[package]]
+name = "miniz_oxide"
+version = "0.7.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9d811f3e15f28568be3407c8e7fdb6514c1cda3cb30683f15b6a1a1dc4ea14a7"
+dependencies = [
+ "adler",
+]
+
+[[package]]
+name = "nix"
+version = "0.27.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "2eb04e9c688eff1c89d72b407f168cf79bb9e867a9d3323ed6c01519eb9cc053"
+dependencies = [
+ "bitflags",
+ "cfg-if",
+ "libc",
+]
+
+[[package]]
+name = "num-traits"
+version = "0.2.18"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "da0df0e5185db44f69b44f26786fe401b6c293d1907744beaa7fa62b2e5a517a"
+dependencies = [
+ "autocfg",
+]
+
+[[package]]
+name = "object"
+version = "0.32.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a6a622008b6e321afc04970976f62ee297fdbaa6f95318ca343e3eebb9648441"
+dependencies = [
+ "memchr",
+]
+
 [[package]]
 name = "once_cell"
+version = "1.19.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3fdb12b2476b595f9358c5161aa467c2438859caa136dec86c26fdd2efe17b92"
+
+[[package]]
+name = "owo-colors"
+version = "3.5.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c1b04fb49957986fdce4d6ee7a65027d55d4b6d2265e5848bbb507b58ccfdb6f"
+
+[[package]]
+name = "paste"
+version = "1.0.14"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "de3145af08024dea9fa9914f381a17b8fc6034dfb00f3a84013f7ff43f29ed4c"
+
+[[package]]
+name = "pin-project-lite"
+version = "0.2.13"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8afb450f006bf6385ca15ef45d71d2288452bc3683ce2e2cacc0d18e4be60b58"
+
+[[package]]
+name = "proc-macro2"
+version = "1.0.78"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e2422ad645d89c99f8f3e6b88a9fdeca7fabeac836b1002371c4367c8f984aae"
+dependencies = [
+ "unicode-ident",
+]
+
+[[package]]
+name = "quote"
+version = "1.0.35"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "291ec9ab5efd934aaf503a6466c5d5251535d108ee747472c3977cc5acc868ef"
+dependencies = [
+ "proc-macro2",
+]
+
+[[package]]
+name = "regex-syntax"
+version = "0.6.29"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f162c6dd7b008981e4d40210aca20b4bd0f9b60ca9271061b07f78537722f2e1"
+
+[[package]]
+name = "rhai"
 version = "1.17.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b7e5500299e16ebb147ae15a00a942af264cf3688f47923b8fc2cd5858f23ad3"
+checksum = "f6273372244d04a8a4b0bec080ea1e710403e88c5d9d83f9808b2bfa64f0982a"
+dependencies = [
+ "ahash",
+ "bitflags",
+ "instant",
+ "num-traits",
+ "once_cell",
+ "rhai_codegen",
+ "smallvec",
+ "smartstring",
+ "thin-vec",
+]
+
+[[package]]
+name = "rhai_codegen"
+version = "2.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9db7f8dc4c9d48183a17ce550574c42995252b82d267eaca3fcd1b979159856c"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn 2.0.48",
+]
+
+[[package]]
+name = "rustc-demangle"
+version = "0.1.23"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d626bb9dae77e28219937af045c257c28bfd3f69333c512553507f5f9798cb76"
+
+[[package]]
+name = "serde"
+version = "1.0.196"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "870026e60fa08c69f064aa766c10f10b1d62db9ccd4d0abb206472bee0ce3b32"
+dependencies = [
+ "serde_derive",
+]
+
+[[package]]
+name = "serde_derive"
+version = "1.0.196"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "33c85360c95e7d137454dc81d9a4ed2b8efd8fbe19cee57357b32b9771fccb67"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn 2.0.48",
+]
+
+[[package]]
+name = "sharded-slab"
+version = "0.1.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f40ca3c46823713e0d4209592e8d6e826aa57e928f09752619fc696c499637f6"
+dependencies = [
+ "lazy_static",
+]
+
+[[package]]
+name = "smallvec"
+version = "1.13.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e6ecd384b10a64542d77071bd64bd7b231f4ed5940fba55e98c3de13824cf3d7"
+
+[[package]]
+name = "smartstring"
+version = "1.0.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3fb72c633efbaa2dd666986505016c32c3044395ceaf881518399d2f4127ee29"
+dependencies = [
+ "autocfg",
+ "static_assertions",
+ "version_check",
+]
+
+[[package]]
+name = "static_assertions"
+version = "1.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a2eb9349b6444b326872e140eb1cf5e7c522154d69e7a0ffb0fb81c06b37543f"
+
+[[package]]
+name = "syn"
+version = "1.0.109"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "72b64191b275b66ffe2469e8af2c1cfe3bafa67b529ead792a6d0160888b4237"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "unicode-ident",
+]
+
+[[package]]
+name = "syn"
+version = "2.0.48"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0f3531638e407dfc0814761abb7c00a5b54992b849452a0646b7f65c9f770f3f"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "unicode-ident",
+]
+
+[[package]]
+name = "thin-vec"
+version = "0.2.13"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a38c90d48152c236a3ab59271da4f4ae63d678c5d7ad6b7714d7cb9760be5e4b"
+
+[[package]]
+name = "thread_local"
+version = "1.1.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3fdd6f064ccff2d6567adcb3873ca630700f00b5ad3f060c25b5dcfd9a4ce152"
+dependencies = [
+ "cfg-if",
+ "once_cell",
+]
+
+[[package]]
+name = "tiny-keccak"
+version = "2.0.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "2c9d3793400a45f954c52e73d068316d76b6f4e36977e3fcebb13a2721e80237"
+dependencies = [
+ "crunchy",
+]
+
+[[package]]
+name = "tracing"
+version = "0.1.40"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c3523ab5a71916ccf420eebdf5521fcef02141234bbc0b8a49f2fdc4544364ef"
+dependencies = [
+ "pin-project-lite",
+ "tracing-core",
+]
+
+[[package]]
+name = "tracing-core"
+version = "0.1.32"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c06d3da6113f116aaee68e4d601191614c9053067f9ab7f6edbcb161237daa54"
+dependencies = [
+ "once_cell",
+ "valuable",
+]
+
+[[package]]
+name = "tracing-error"
+version = "0.2.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d686ec1c0f384b1277f097b2f279a2ecc11afe8c133c1aabf036a27cb4cd206e"
+dependencies = [
+ "tracing",
+ "tracing-subscriber",
+]
+
+[[package]]
+name = "tracing-subscriber"
+version = "0.3.18"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ad0f048c97dbd9faa9b7df56362b8ebcaa52adb06b498c050d2f4e32f90a7a8b"
+dependencies = [
+ "sharded-slab",
+ "thread_local",
+ "tracing-core",
+]
+
+[[package]]
+name = "unicode-ident"
+version = "1.0.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3354b9ac3fae1ff6755cb6db53683adb661634f67557942dea4facebec0fee4b"
+
+[[package]]
+name = "valuable"
+version = "0.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "830b7e5d4d90034032940e4ace0d9a9a057e7a45cd94e6c007832e39edb82f6d"
 
 [[package]]
 name = "version_check"
 version = "0.9.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "49874b5167b65d7193b8aba1567f5c7d93d001cafc34600cee003eda787e483f"
+
+[[package]]
+name = "wasi"
+version = "0.11.0+wasi-snapshot-preview1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9c8d87e72b64a3b4db28d11ce29237c246188f4f51057d65a7eab63b7987e423"
+
+[[package]]
+name = "with_builtin_macros"
+version = "0.0.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a59d55032495429b87f9d69954c6c8602e4d3f3e0a747a12dea6b0b23de685da"
+dependencies = [
+ "with_builtin_macros-proc_macros",
+]
+
+[[package]]
+name = "with_builtin_macros-proc_macros"
+version = "0.0.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "15bd7679c15e22924f53aee34d4e448c45b674feb6129689af88593e129f8f42"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn 1.0.109",
+]
+
+[[package]]
+name = "xtask"
+version = "0.1.0"
+dependencies = [
+ "argh",
+ "color-eyre",
+ "once_cell",
+]
+
+[[package]]
+name = "zerocopy"
+version = "0.7.32"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "74d4d3961e53fa4c9a25a8637fc2bfaf2595b3d3ae34875568a5cf64787716be"
+dependencies = [
+ "zerocopy-derive",
+]
+
+[[package]]
+name = "zerocopy-derive"
+version = "0.7.32"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9ce1b18ccd8e73a9321186f97e46f9f04b778851177567b1975109d26a08d2a6"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn 2.0.48",
+]

Cargo.toml

@@ -1,2 +1,3 @@
 [workspace]
-members = ["hbvm", "compiler"]
+resolver = "2"
+members = ["hbasm", "hbbytecode", "hblang", "hbvm", "hbvm_aos_on_linux", "hbxrt", "xtask"]

DESIGN_DOC.md

@@ -1,28 +0,0 @@
-# Math operations
-```
-MATH_OP
-    Add
-    Sub
-    Mul
-    Div
-    Mod
-```
-```
-MATH_TYPE
-    Unsigned
-    Signed
-    FloatingPoint
-```
-
-```
-MATH_OP_SIDES
-    Register Constant
-    Register Register
-    Constant Constant
-    Constant Register
-```
-`[MATH_OP] [MATH_OP_SIDES] [MATH_TYPE] [IMM_LHS] [IMM_RHS] [REG]`
-
-
-
-

assets/example.asm

@@ -1,4 +0,0 @@
-load 0 a0        ;; 05 00 A0
-load 10 a1       ;; 05 10 A1
-add a0 1 a0      ;; 01 A0 01 A0
-jump_neq a0 a1 0 ;; a1 A0 A1 0

assets/example.bytes

@@ -1,4 +0,0 @@
-load 10 A1
-load 0 A0
-add A0 1
-jump_less_than A0 A1 0

compiler/src/main.rs

@@ -1,5 +0,0 @@
-fn main() {
-    let prog = "load 1, A0
-    jump 0";
-    println!("Hello, world!");
-}

examples/asm/hello-name.hba

@@ -0,0 +1,100 @@
+jmp entry
+
+puts:
+-- Write string to console
+--  r2: [IN] *const u8  String pointer
+--  r3: [IN] usize      String length
+
+    li8     r1, 0x1         -- Write syscall
+    brc     r2, r3, 2       -- Copy parameters
+    li8     r2, 0x1         -- STDOUT
+    eca
+
+    jal     r0, r31, 0
+
+gets:
+-- Read string until end of buffer or LF
+--  r2: [IN] *mut u8    Buffer
+--  r3: [IN] usize      Buffer length
+
+    -- Register allocations:
+    --  r33: *mut u8    Buffer end
+    --  r34: u8         Immediate char
+    --  r35: u8         Const [0x0A = LF]
+
+    li8     r35, 0x0A
+    add64   r33, r2, r3
+
+    -- Setup syscall
+    li8     r2, 0x1         -- Stdin
+    cp      r3, r2
+    li8     r4, 0x1         -- Read one char
+
+    jeq     r3, r33, end
+    loop:
+        li8     r1, 0x1     -- Read syscall
+        eca
+        addi64  r3, r3, 1
+        ld      r34, r3, 0, 1
+        jeq     r34, r35, end
+        jne     r3, r33, loop
+    
+    end:
+        -- Set copied amount
+        sub64   r1, r33, r3
+        addi64  r1, -1
+        jal     r0, r31, 0
+
+alloc-pages:
+-- Allocate pages
+--  r1: [OUT] *mut u8   Pointer to page
+--  r2: [IN]  u16       Page count
+
+    muli16  r3, r2, 4096    -- page count
+    li8     r1, 0x9         -- mmap syscall
+    li8     r2, 0x0         -- no address set, kernel chosen
+    li8     r4, 0x2         -- PROT_WRITE
+    li8     r5, 0x20        -- MAP_ANONYMOUS
+    li64    r6, -1          -- Doesn't map file
+    li8     r7, 0x0         -- Doesn't map file
+    eca
+
+    jal     r0, r31, 0
+
+entry:
+-- Program entrypoint
+
+    -- Register allocations:
+    -- r32: *mut u8     Buffer
+    -- r36: usize       Read buffer length
+
+    -- Allocate one page (4096 KiB)
+    li8     r2, 1
+    jal     r31, 0, alloc-pages
+    cp      r32, r1
+
+    -- Print message
+    lra16   r2, r0, #enter-your-name
+    li8     r3, 17
+    jal     r31, r0, puts
+
+    -- Read name
+    cp      r2, r32
+    li16    r3, 4096
+    jal     r31, r0, gets
+    cp      r36, r1
+
+    -- Print your name is
+    lra16   r2, r0, #your-name-is
+    li8     r3, 15
+    jal     r31, r0, puts
+
+    -- And now print the name
+    cp      r2, r32
+    cp      r3, r36
+    jal     r31, r0, puts
+
+    tx
+
+#enter-your-name: "Enter your name: "
+#your-name-is   : "\nYour name is: "

hbasm/Cargo.toml

@@ -0,0 +1,9 @@
+[package]
+name = "hbasm"
+version = "0.1.0"
+edition = "2021"
+
+[dependencies]
+paste               = "1.0"
+rhai                = "1.16"
+with_builtin_macros = "0.0.3"

hbasm/examples/ableos/main.rhai

@@ -0,0 +1,13 @@
+import "hbasm/examples/ableos/std" as std;
+
+fn main(){
+    std::Error(":+)");
+    std::Warn("Your mom fell in a well!");
+    std::Info("Hello, world!");
+    std::Debug("ABC");
+    std::Trace("Trace Deez");
+
+    tx();
+}
+
+main();

hbasm/examples/ableos/std.rhai

@@ -0,0 +1,24 @@
+fn ipc_send(buffer_id, mem_addr, length){
+    // set the ecall
+    li8(r1, 3);
+    // Set the buffer ID to be the BufferID
+    li64(r2, buffer_id);
+    lra(r3, r0, mem_addr);
+    // set the length
+    li64(r4, length);
+    // ecall
+    eca(); 
+}
+
+private fn log(log_level, string){
+    let str = data::str(string);
+    ipc_send(1, str, str.len);
+}
+
+fn Error(string) {log(0, string);}
+fn Warn(string) {log(1, string);}
+fn Info(string) {log(2, string);}
+// Due to rhai limitations this cannot be debug
+// because of this all of the log levels are upper case
+fn Debug(string) {log(3, string);}
+fn Trace(string) {log(4, string);}

hbasm/examples/hello-linux.rhai

@@ -0,0 +1,9 @@
+let hello = data::str("Hello, world!");
+
+li8     (r1, 1);            // Write syscall
+li8     (r2, 1);            // Stdout FD
+lra16   (r3, r0, hello);    // String buffer
+li8     (r4, hello.len);    // String length
+eca     ();                 // System call
+
+tx      ();                 // End program

hbasm/examples/macros.rhai

@@ -0,0 +1,33 @@
+li8(r1, 69);
+li8(r2,  0);
+
+if_eq(r1, r2,
+    || puts("Equals!"),
+    || puts("Not equals!"),
+);
+
+
+tx(); // END OF MAIN
+
+/// Inline function – write text to stdout
+fn puts(string) {
+    let d = data::str(string);
+    li8   (r1, 1);     // Write syscall
+    li8   (r2, 1);     // Stdout handle
+    lra16 (r3, r0, d);
+    li64  (r4, d.len);
+    eca   ();
+}
+
+fn if_eq(a, b, thenblk, elseblk) {
+    let elselbl = declabel();
+    let endlbl  = declabel();
+
+    jne(a, b, elselbl);
+    thenblk.call();
+    jmp16(endlbl);
+
+    elselbl.here();
+    elseblk.call();
+    endlbl.here();
+}

hbasm/src/data.rs

@@ -0,0 +1,101 @@
+//! Data section inserts
+
+use {
+    crate::{object::SymbolRef, SharedObject},
+    rhai::{CustomType, Engine, FuncRegistration, ImmutableString, Module},
+};
+
+/// Generate insertions for data types
+///
+/// `gen_data_instructions!($module, $obj, [$type, …]);`
+/// - `$module`: Rhai module
+/// - `$obj`: Code object
+/// - `$type`: Type of single array item
+macro_rules! gen_data_insertions {
+    ($module:expr, $obj:expr, [$($ty:ident),* $(,)?] $(,)?) => {{
+        let (module, obj) = ($module, $obj);
+        $({
+            // Clone object to each function
+            let obj = ::std::rc::Rc::clone(obj);
+
+            FuncRegistration::new(stringify!($ty))
+                .with_namespace(rhai::FnNamespace::Global)
+                .set_into_module::<_, 1, false, _, true, _>(module, move |arr: ::rhai::Array| {
+                    let obj    = &mut *obj.borrow_mut();
+                    let symbol = obj.symbol($crate::object::Section::Data);
+
+                    // Reserve space for object so we don't resize it
+                    // all the time
+                    obj.sections
+                        .data
+                        .reserve(arr.len() * ::std::mem::size_of::<$ty>());
+
+                    // For every item…
+                    for item in arr {
+                        // … try do conversions from i32 to desired type
+                        //   and insert it.
+                        obj.sections.data.extend(
+                            match item.as_int() {
+                                Ok(num) => $ty::try_from(num).map_err(|_| "i64".to_owned()),
+                                Err(ty) => Err(ty.to_owned()),
+                            }
+                            .map_err(|err| {
+
+                                ::rhai::EvalAltResult::ErrorMismatchDataType(
+                                    stringify!($ty).to_owned(),
+                                    err,
+                                    ::rhai::Position::NONE,
+                                )
+                            })?
+                            .to_le_bytes(),
+                        );
+                    }
+
+                    Ok(DataRef {
+                        symbol,
+                        len: obj.sections.data.len() - symbol.0,
+                    })
+                });
+        })*
+    }};
+}
+
+/// Reference to entry in data section
+#[derive(Clone, Copy, Debug)]
+pub struct DataRef {
+    pub symbol: SymbolRef,
+    pub len:    usize,
+}
+
+impl CustomType for DataRef {
+    fn build(mut builder: rhai::TypeBuilder<Self>) {
+        builder
+            .with_name("DataRef")
+            .with_get("symbol", |this: &mut Self| this.symbol)
+            .with_get("len", |this: &mut Self| this.len as u64 as i64);
+    }
+}
+
+pub fn module(engine: &mut Engine, obj: SharedObject) -> Module {
+    let mut module = Module::new();
+
+    gen_data_insertions!(&mut module, &obj, [i8, i16, i32, i64]);
+
+    // Specialisation for strings, they should be
+    // inserted as plain UTF-8 arrays
+    FuncRegistration::new("str")
+        .with_namespace(rhai::FnNamespace::Global)
+        .set_into_module::<_, 1, false, _, true, _>(&mut module, move |s: ImmutableString| {
+            let obj = &mut *obj.borrow_mut();
+            let symbol = obj.symbol(crate::object::Section::Data);
+
+            obj.sections.data.extend(s.as_bytes());
+            Ok(DataRef {
+                symbol,
+                len: s.len(),
+            })
+        });
+
+    engine.build_type::<DataRef>();
+    module
+}

hbasm/src/ins.rs

@@ -0,0 +1,319 @@
+//! Functions for inserting instructions
+//! 
+//! Most of the code you see is just metaprogramming stuff.
+//! This ensures that adding new instructions won't need any
+//! specific changes and consistent behaviour.
+//!
+//! > I tried to comment stuff here, but I meanwhile forgor how it works.
+//!
+//! — Erin
+
+use {
+    crate::object::Object,
+    rhai::{FuncRegistration, Module},
+    std::{cell::RefCell, rc::Rc},
+};
+
+/// Operand types and their insertions
+pub mod optypes {
+    use {
+        crate::{
+            label::UnboundLabel,
+            object::{Object, RelocKey, RelocType, SymbolRef},
+        },
+        rhai::{Dynamic, EvalAltResult, ImmutableString, Position},
+    };
+
+    // These types represent operand types to be inserted
+    pub type R = u8;
+    pub type B = i8;
+    pub type H = i16;
+    pub type W = i32;
+    pub type D = i64;
+
+    pub type A = Dynamic;
+    pub type O = Dynamic;
+    pub type P = Dynamic;
+
+    /// Insert relocation into code
+    ///
+    /// - If integer, just write it to the code
+    /// - Otherwise insert entry into relocation table
+    ///   and fill zeroes
+    pub fn insert_reloc(
+        obj: &mut Object,
+        ty: RelocType,
+        val: &Dynamic,
+    ) -> Result<(), EvalAltResult> {
+        match () {
+            // Direct references – insert directly to table
+            _ if val.is::<SymbolRef>() => {
+                obj.relocation(RelocKey::Symbol(val.clone_cast::<SymbolRef>().0), ty)
+            }
+            _ if val.is::<UnboundLabel>() => {
+                obj.relocation(RelocKey::Symbol(val.clone_cast::<UnboundLabel>().0), ty)
+            }
+            _ if val.is::<DataRef>() => {
+                obj.relocation(RelocKey::Symbol(val.clone_cast::<DataRef>().symbol.0), ty)
+            }
+
+            // String (indirect) reference
+            _ if val.is_string() => {
+                obj.relocation(RelocKey::Label(val.clone_cast::<ImmutableString>()), ty)
+            }
+
+            // Manual offset
+            _ if val.is_int() => {
+                let int = val.clone_cast::<i64>();
+                match ty {
+                    RelocType::Rel32 => obj.sections.text.extend((int as i32).to_le_bytes()),
+                    RelocType::Rel16 => obj.sections.text.extend((int as i16).to_le_bytes()),
+                    RelocType::Abs64 => obj.sections.text.extend(int.to_le_bytes()),
+                }
+            }
+
+            _ => {
+                return Err(EvalAltResult::ErrorMismatchDataType(
+                    "SymbolRef, UnboundLabel, String or Int".to_owned(),
+                    val.type_name().to_owned(),
+                    Position::NONE,
+                ))
+            }
+        }
+
+        Ok(())
+    }
+
+    /// Generate macro for inserting item into the output object
+    ///
+    /// Pre-defines inserts for absolute address and relative offsets.
+    /// These are inserted with function [`insert_reloc`]
+    /// # le_bytes
+    /// `gen_insert!(le_bytes: [B, …]);`
+    ///
+    /// Takes sequence of operand types which should be inserted
+    /// by invoking `to_le_bytes` method on it.
+    macro_rules! gen_insert {
+        (le_bytes: [$($lety:ident),* $(,)?]) => {
+            /// `insert!($thing, $obj, $type)` where
+            /// - `$thing`: Value you want to insert
+            /// - `$obj`: Code object
+            /// - `$type`: Type of inserted value
+            ///
+            /// Eg. `insert!(69_u8, obj, B);`
+            macro_rules! insert {
+                $(($thing:expr, $obj: expr, $lety) => {
+                    $obj.sections.text.extend($thing.to_le_bytes());
+                };)*
+
+                ($thing:expr, $obj:expr, A) => {
+                    $crate::ins::optypes::insert_reloc(
+                        $obj,
+                        $crate::object::RelocType::Abs64,
+                        $thing
+                    )?
+                };
+                ($thing:expr, $obj:expr, O) => {
+                    $crate::ins::optypes::insert_reloc(
+                        $obj,
+                        $crate::object::RelocType::Rel32,
+                        $thing
+                    )?
+                };
+                ($thing:expr, $obj:expr, P) => {
+                    $crate::ins::optypes::insert_reloc(
+                        $obj,
+                        $crate::object::RelocType::Rel16,
+                        $thing
+                    )?
+                };
+            }
+        };
+    }
+
+    gen_insert!(le_bytes: [R, B, H, W, D]);
+
+    #[allow(clippy::single_component_path_imports)]
+    pub(super) use insert;
+
+    use crate::data::DataRef;
+}
+
+/// Rhai Types (types for function parameters as Rhai uses only 64bit signed integers)
+pub mod rity {
+    pub use super::optypes::{A, O, P, R};
+    pub type B = i64;
+    pub type H = i64;
+    pub type W = i64;
+    pub type D = i64;
+}
+
+/// Generic instruction (instruction of certain operands type) inserts
+pub mod generic {
+    use {crate::object::Object, rhai::EvalAltResult};
+
+    pub(super) fn convert_op<A, B>(from: A) -> Result<B, EvalAltResult>
+    where
+        B: TryFrom<A>,
+        <B as TryFrom<A>>::Error: std::error::Error + Sync + Send + 'static,
+    {
+        B::try_from(from).map_err(|e| {
+            EvalAltResult::ErrorSystem("Data conversion error".to_owned(), Box::new(e))
+        })
+    }
+
+    /// Generate opcode-generic instruction insert macro
+    macro_rules! gen_ins {
+        ($($($name:ident : $ty:ty),*;)*) => {
+            paste::paste! {
+                $(
+                    /// Instruction-generic opcode insertion function
+                    /// - `obj`: Code object
+                    /// - `opcode`: opcode, not checked if valid for instruction type
+                    /// - … for operands
+                    #[inline]
+                    pub fn [<$($ty:lower)*>](
+                        obj: &mut Object,
+                        opcode: u8,
+                        $($name: $crate::ins::optypes::$ty),*,
+                    ) -> Result<(), EvalAltResult> {
+                        // Push opcode
+                        obj.sections.text.push(opcode);
+
+                        // Insert based on type
+                        $($crate::ins::optypes::insert!(&$name, obj, $ty);)*
+                        Ok(())
+                    }
+                )*
+
+                /// Generate Rhai opcode-specific instruction insertion functions
+                ///
+                /// `gen_ins_fn!($obj, $opcode, $optype);` where:
+                /// - `$obj`: Code object
+                /// - `$opcode`: Opcode value
+                macro_rules! gen_ins_fn {
+                    $(
+                        ($obj:expr, $opcode:expr, [<$($ty)*>]) => {
+                            // Opcode-specific insertion function
+                            // - Parameters = operands
+                            move |$($name: $crate::ins::rity::$ty),*| {
+                                // Invoke generic function
+                                $crate::ins::generic::[<$($ty:lower)*>](
+                                    &mut *$obj.borrow_mut(),
+                                    $opcode,
+                                    $(
+                                        // Convert to desired type (from Rhai-provided values)
+                                        $crate::ins::generic::convert_op::<     
+                                            _,
+                                            $crate::ins::optypes::$ty
+                                        >($name)?
+                                    ),*
+                                )?;
+                                Ok(())
+                            }
+                        };
+
+                        // Internal-use: count args
+                        (@arg_count [<$($ty)*>]) => {
+                            { ["", $(stringify!($ty)),*].len() - 1 }
+                        };
+                    )*
+
+                    // Specialisation for no-operand instructions
+                    ($obj:expr, $opcode:expr, N) => {
+                        move || {
+                            $crate::ins::generic::n(&mut *$obj.borrow_mut(), $opcode);
+                            Ok(())
+                        }
+                    };
+
+                    // Internal-use specialisation: no-operand instructions
+                    (@arg_count N) => {
+                        { 0 }
+                    };
+                }
+            }
+        };
+    }
+
+    /// Specialisation for no-operand instructions – simply just push opcode
+    #[inline]
+    pub fn n(obj: &mut Object, opcode: u8) {
+        obj.sections.text.push(opcode);
+    }
+
+    // Generate opcode-generic instruction inserters
+    // (operand identifiers are arbitrary)
+    //
+    // New instruction types have to be added manually here
+    gen_ins! {
+        o0: R, o1: R;
+        o0: R, o1: R, o2: R;
+        o0: R, o1: R, o2: R, o3: R;
+        o0: R, o1: R, o2: B;
+        o0: R, o1: R, o2: H;
+        o0: R, o1: R, o2: W;
+        o0: R, o1: R, o2: D;
+        o0: R, o1: B;
+        o0: R, o1: H;
+        o0: R, o1: W;
+        o0: R, o1: D;
+        o0: R, o1: R, o2: A;
+        o0: R, o1: R, o2: A, o3: H;
+        o0: R, o1: R, o2: O, o3: H;
+        o0: R, o1: R, o2: P, o3: H;
+        o0: R, o1: R, o2: O;
+        o0: R, o1: R, o2: P;
+        o0: O;
+        o0: P;
+    }
+
+    #[allow(clippy::single_component_path_imports)]
+    pub(super) use gen_ins_fn;
+}
+
+/// Generate instructions from instruction table
+///
+/// ```ignore
+/// instructions!(($module, $obj) {
+///     // Data from instruction table
+///     $opcode, $mnemonic, $opty, $doc;
+///     …
+/// });
+/// ```
+/// - `$module`: Rhai module
+/// - `$obj`: Code object
+macro_rules! instructions {
+    (
+        ($module:expr, $obj:expr $(,)?)
+        { $($opcode:expr, $mnemonic:ident, $ops:tt, $doc:literal;)* }
+    ) => {{
+        let (module, obj) = ($module, $obj);
+        $({
+            // Object is shared across all functions
+            let obj = Rc::clone(&obj);
+
+            // Register newly generated function for each instruction
+            FuncRegistration::new(stringify!([<$mnemonic:lower>]))
+                .with_namespace(rhai::FnNamespace::Global)
+                .set_into_module::<_, { generic::gen_ins_fn!(@arg_count $ops) }, false, _, true, _>(
+                    module,
+                    generic::gen_ins_fn!(
+                        obj,
+                        $opcode,
+                        $ops
+                    )
+                );
+        })*
+    }};
+}
+
+/// Setup instruction insertors
+pub fn setup(module: &mut Module, obj: Rc<RefCell<Object>>) {
+    // Import instructions table and use it for generation
+    with_builtin_macros::with_builtin! {
+        let $spec = include_from_root!("../hbbytecode/instructions.in") in {
+            instructions!((module, obj) { $spec });
+        }
+    }
+}

hbasm/src/label.rs

@@ -0,0 +1,112 @@
+//! Stuff related to labels
+
+use {
+    crate::SharedObject,
+    rhai::{Engine, FuncRegistration, ImmutableString, Module},
+};
+
+/// Macro for creating functions for Rhai which
+/// is bit more friendly
+///
+/// ```ignore
+///     shdm_fns!{
+///         module: $module;
+///         shared: $shared => $shname;
+///         
+///         $vis fn $name($param_name: $param_ty, …) -> $ret { … }
+///         …
+///     }
+/// ```
+/// - `$module`: Rhai module
+/// - `$shared`: Data to be shared across the functions
+///     - `$shname`: The binding name inside functions
+/// - `$vis`: Function visibility for Rhai
+///     - Lowercased [`rhai::FnNamespace`] variants
+/// - `$name`: Function name
+/// - `$param_name`: Parameter name
+/// - `$param_ty`: Rust parameter type
+/// - `$ret`: Optional return type (otherwise infer)
+macro_rules! shdm_fns {
+    (
+        module: $module:expr;
+        shared: $shared:expr => $shname:ident;
+
+        $(
+            $vis:ident fn $name:ident($($param_name:ident: $param_ty:ty),*) $(-> $ret:ty)? $blk:block
+        )*
+    ) => {{
+        let module = $module;
+        let shared = $shared;
+        paste::paste! {
+            $({
+
+                let $shname = SharedObject::clone(&shared);
+    
+                FuncRegistration::new(stringify!($name))
+                    .with_namespace(rhai::FnNamespace::[<$vis:camel>])
+                    .set_into_module::<_, { ["", $(stringify!($param_name)),*].len() - 1 }, false, _, true, _>(
+                        module,
+                        move |$($param_name: $param_ty),*| $(-> $ret)? {
+                            let mut $shname = $shname.borrow_mut();
+                            $blk
+                        }
+                    );
+            })*
+        }
+    }};
+}
+
+/// Label without any place bound
+#[derive(Clone, Copy, Debug)]
+pub struct UnboundLabel(pub usize);
+
+pub fn setup(engine: &mut Engine, module: &mut Module, object: SharedObject) {
+    shdm_fns! {
+        module: module;
+        shared: object => obj;
+
+        // Insert unnamed label
+        global fn label() {
+            let symbol = obj.symbol(crate::object::Section::Text);
+            Ok(symbol)
+        }
+
+        // Insert string-labeled label
+        global fn label(label: ImmutableString) {
+            let symbol = obj.symbol(crate::object::Section::Text);
+            obj.labels.insert(label, symbol.0);
+
+            Ok(symbol)
+        }
+
+        // Declare unbound label (to be bound later)
+        global fn declabel() {
+            let index = obj.symbols.len();
+            obj.symbols.push(None);
+
+            Ok(UnboundLabel(index))
+        }
+
+        // Declare unbound label (to be bound later)
+        // with string label
+        global fn declabel(label: ImmutableString) {
+            let index = obj.symbols.len();
+            obj.symbols.push(None);
+            obj.labels.insert(label, index);
+
+            Ok(UnboundLabel(index))
+        }
+
+        // Set location for unbound label
+        global fn here(label: UnboundLabel) {
+            obj.symbols[label.0] = Some(crate::object::SymbolEntry {
+                location: crate::object::Section::Text,
+                offset:   obj.sections.text.len(),
+            });
+
+            Ok(())
+        }
+    }
+
+    engine.register_type_with_name::<UnboundLabel>("UnboundLabel");
+}

hbasm/src/lib.rs

@@ -0,0 +1,45 @@
+pub mod data;
+pub mod ins;
+pub mod label;
+pub mod linker;
+pub mod object;
+
+use {
+    object::Object,
+    rhai::{Engine, Module},
+    std::{cell::RefCell, rc::Rc},
+};
+
+type SharedObject = Rc<RefCell<Object>>;
+
+pub fn assembler(
+    linkout: &mut impl std::io::Write,
+    loader: impl FnOnce(&mut Engine) -> Result<(), Box<rhai::EvalAltResult>>,
+) -> Result<(), Box<dyn std::error::Error>> {
+    let mut engine = Engine::new();
+    let mut module = Module::new();
+    let obj = Rc::new(RefCell::new(Object::default()));
+    ins::setup(&mut module, Rc::clone(&obj));
+    label::setup(&mut engine, &mut module, Rc::clone(&obj));
+
+    // Registers
+    for n in 0_u8..=255 {
+        module.set_var(format!("r{n}"), n);
+    }
+
+    module.set_native_fn("reg", |n: i64| {
+        Ok(u8::try_from(n).map_err(|_| {
+            rhai::EvalAltResult::ErrorRuntime("Invalid register value".into(), rhai::Position::NONE)
+        })?)
+    });
+
+    module.set_native_fn("as_i64", |n: u8| Ok(n as i64));
+
+    let datamod = Rc::new(data::module(&mut engine, SharedObject::clone(&obj)));
+    engine.register_global_module(Rc::new(module));
+    engine.register_static_module("data", datamod);
+    engine.register_type_with_name::<object::SymbolRef>("SymbolRef");
+    loader(&mut engine)?;
+    linker::link(obj, linkout)?;
+    Ok(())
+}

hbasm/src/linker.rs

@@ -0,0 +1,47 @@
+//! Simple flat-bytecode linker
+
+use {
+    crate::{
+        object::{RelocKey, RelocType, Section},
+        SharedObject,
+    },
+    std::io::Write,
+};
+
+pub fn link(object: SharedObject, out: &mut impl Write) -> std::io::Result<()> {
+    let obj = &mut *object.borrow_mut();
+
+    // Walk relocation table entries
+    for (&loc, entry) in &obj.relocs {
+        let value = match &entry.key {
+            // Symbol – direct reference
+            RelocKey::Symbol(sym) => obj.symbols[*sym],
+
+            // Label – indirect label reference
+            RelocKey::Label(label) => obj.symbols[obj.labels[label]],
+        }
+        .ok_or_else(|| std::io::Error::other("Invalid symbol"))?;
+
+        let offset = match value.location {
+            // Text section is on the beginning
+            Section::Text => value.offset,
+
+            // Data section follows text section immediately
+            Section::Data => value.offset + obj.sections.text.len(),
+        };
+
+        // Insert address or calulate relative offset
+        match entry.ty {
+            RelocType::Rel32 => obj.sections.text[loc..loc + 4]
+                .copy_from_slice(&((offset as isize - loc as isize) as i32).to_le_bytes()),
+            RelocType::Rel16 => obj.sections.text[loc..loc + 2]
+                .copy_from_slice(&((offset as isize - loc as isize) as i16).to_le_bytes()),
+            RelocType::Abs64 => obj.sections.text[loc..loc + 8]
+                .copy_from_slice(&(offset as isize - loc as isize).to_le_bytes()),
+        }
+    }
+
+    // Write to output
+    out.write_all(&obj.sections.text)?;
+    out.write_all(&obj.sections.data)
+}

hbasm/src/main.rs

@@ -0,0 +1,8 @@
+use std::{io::stdout, path::PathBuf};
+
+fn main() -> Result<(), Box<dyn std::error::Error>> {
+    let path = PathBuf::from(std::env::args().nth(1).ok_or("Missing path")?);
+    hbasm::assembler(&mut stdout(), |engine| engine.run_file(path))?;
+    
+    Ok(())
+}

hbasm/src/object.rs

@@ -0,0 +1,94 @@
+//! Code object
+
+use {rhai::ImmutableString, std::collections::HashMap};
+
+/// Section tabel
+#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+pub enum Section {
+    Text,
+    Data,
+}
+
+/// Symbol entry (in what section, where)
+#[derive(Clone, Copy, Debug)]
+pub struct SymbolEntry {
+    pub location: Section,
+    pub offset:   usize,
+}
+
+/// Relocation table key
+#[derive(Clone, Debug)]
+pub enum RelocKey {
+    /// Direct reference
+    Symbol(usize),
+    /// Indirect reference
+    Label(ImmutableString),
+}
+
+/// Relocation type
+#[derive(Clone, Copy, Debug)]
+pub enum RelocType {
+    Rel32,
+    Rel16,
+    Abs64,
+}
+
+/// Relocation table entry
+#[derive(Clone, Debug)]
+pub struct RelocEntry {
+    pub key: RelocKey,
+    pub ty:  RelocType,
+}
+
+/// Object code
+#[derive(Clone, Debug, Default)]
+pub struct Sections {
+    pub text: Vec<u8>,
+    pub data: Vec<u8>,
+}
+
+/// Object
+#[derive(Clone, Debug, Default)]
+pub struct Object {
+    /// Vectors with sections
+    pub sections: Sections,
+    /// Symbol table
+    pub symbols:  Vec<Option<SymbolEntry>>,
+    /// Labels to symbols table
+    pub labels:   HashMap<ImmutableString, usize>,
+    /// Relocation table
+    pub relocs:   HashMap<usize, RelocEntry>,
+}
+
+#[derive(Clone, Copy, Debug)]
+#[repr(transparent)]
+pub struct SymbolRef(pub usize);
+
+impl Object {
+    /// Insert symbol at current location in specified section
+    pub fn symbol(&mut self, section: Section) -> SymbolRef {
+        let section_buf = match section {
+            Section::Text => &mut self.sections.text,
+            Section::Data => &mut self.sections.data,
+        };
+
+        self.symbols.push(Some(SymbolEntry {
+            location: section,
+            offset:   section_buf.len(),
+        }));
+
+        SymbolRef(self.symbols.len() - 1)
+    }
+
+    /// Insert to relocation table and write zeroes to code
+    pub fn relocation(&mut self, key: RelocKey, ty: RelocType) {
+        self.relocs
+            .insert(self.sections.text.len(), RelocEntry { key, ty });
+
+        self.sections.text.extend(match ty {
+            RelocType::Rel32 => &[0_u8; 4] as &[u8],
+            RelocType::Rel16 => &[0; 2],
+            RelocType::Abs64 => &[0; 8],
+        });
+    }
+}

hbbytecode/Cargo.toml

@@ -0,0 +1,8 @@
+[package]
+name = "hbbytecode"
+version = "0.1.0"
+edition = "2018"
+
+[dependencies]
+paste = "1.0.14"
+with_builtin_macros = "0.0.3"

hbbytecode/hbbytecode.h

@@ -0,0 +1,75 @@
+/* HoleyBytes Bytecode representation in C
+ * Requires C23 compiler or better
+ *
+ * Uses MSVC pack pragma extension,
+ * proved to work with Clang and GNU® GCC™.
+ */
+
+#pragma once
+#include <assert.h>
+#include <limits.h>
+#include <stdint.h>
+
+static_assert(CHAR_BIT == 8, "Cursed architectures are not supported");
+
+enum hbbc_Opcode: uint8_t {
+    hbbc_Op_UN    , hbbc_Op_TX   , hbbc_Op_NOP   , hbbc_Op_ADD   , hbbc_Op_SUB   , hbbc_Op_MUL   ,
+    hbbc_Op_AND   , hbbc_Op_OR   , hbbc_Op_XOR   , hbbc_Op_SL    , hbbc_Op_SR    , hbbc_Op_SRS   ,
+    hbbc_Op_CMP   , hbbc_Op_CMPU , hbbc_Op_DIR   , hbbc_Op_NEG   , hbbc_Op_NOT   , hbbc_Op_ADDI  , 
+    hbbc_Op_MULI  , hbbc_Op_ANDI , hbbc_Op_ORI   , hbbc_Op_XORI  , hbbc_Op_SLI   , hbbc_Op_SRI   ,
+    hbbc_Op_SRSI  , hbbc_Op_CMPI , hbbc_Op_CMPUI , hbbc_Op_CP    , hbbc_Op_SWA   , hbbc_Op_LI    ,
+    hhbc_Op_LRA   , hbbc_Op_LD   , hbbc_Op_ST    , hbbc_Op_LDR   , hhbc_Op_STR   , hbbc_Op_BMC   ,
+    hbbc_Op_BRC   , hbbc_Op_JMP  , hbbc_Op_JMPR  , hbbc_Op_JAL   , hbbc_Op_JALR  , hbbc_Op_JEQ   ,
+    hbbc_Op_JNE   , hbbc_Op_JLT  , hbbc_Op_JGT   , hbbc_Op_JLTU  , hbbc_Op_JGTU  , hbbc_Op_ECALL ,
+    hbbc_Op_ADDF  , hbbc_Op_SUBF , hbbc_Op_MULF  , hbbc_Op_DIRF  , hbbc_Op_FMAF  , hbbc_Op_NEGF  ,
+    hbbc_Op_ITF   , hbbc_Op_FTI  , hbbc_Op_ADDFI , hbbc_Op_MULFI ,
+} typedef hbbc_Opcode;
+
+static_assert(sizeof(hbbc_Opcode) == 1);
+
+#pragma pack(push, 1)
+struct hbbc_ParamBBBB
+    { uint8_t _0; uint8_t _1; uint8_t _2; uint8_t _3; }
+    typedef hbbc_ParamBBBB;
+    static_assert(sizeof(hbbc_ParamBBBB) == 32 / 8);
+
+struct hbbc_ParamBBB
+    { uint8_t _0; uint8_t _1; uint8_t _2; }
+    typedef hbbc_ParamBBB;
+    static_assert(sizeof(hbbc_ParamBBB) == 24 / 8);
+
+struct hbbc_ParamBBDH
+    { uint8_t _0; uint8_t _1; uint64_t _2; uint16_t _3; }
+    typedef hbbc_ParamBBDH;
+    static_assert(sizeof(hbbc_ParamBBDH) == 96 / 8);
+
+struct hbbc_ParamBBWH
+    { uint8_t _0; uint8_t _1; uint32_t _2; uint16_t _3; }
+    typedef hbbc_ParamBBWH;
+    static_assert(sizeof(hbbc_ParamBBWH) == 64 / 8);
+
+
+struct hbbc_ParamBBD
+    { uint8_t _0; uint8_t _1; uint64_t _2; }
+    typedef hbbc_ParamBBD;
+    static_assert(sizeof(hbbc_ParamBBD) == 80 / 8);
+
+struct hbbc_ParamBBW
+    { uint8_t _0; uint8_t _1; uint32_t _2; }
+    typedef hbbc_ParamBBW;
+    static_assert(sizeof(hbbc_ParamBBW) == 48 / 8);
+
+struct hbbc_ParamBB
+    { uint8_t _0; uint8_t _1; }
+    typedef hbbc_ParamBB;
+    static_assert(sizeof(hbbc_ParamBB) == 16 / 8);
+
+struct hbbc_ParamBD
+    { uint8_t _0; uint64_t _1; }
+    typedef hbbc_ParamBD;
+    static_assert(sizeof(hbbc_ParamBD) == 72 / 8);
+
+typedef uint64_t hbbc_ParamD;
+    static_assert(sizeof(hbbc_ParamD) == 64 / 8);
+
+#pragma pack(pop)

hbbytecode/instructions.in

@@ -0,0 +1,120 @@
+0x00, UN,       N,    "Cause an unreachable code trap"              ;
+0x01, TX,       N,    "Termiante execution"                         ;
+0x02, NOP,      N,    "Do nothing"                                  ;
+0x03, ADD8,     RRR,  "Addition (8b)"                               ;
+0x04, ADD16,    RRR,  "Addition (16b)"                              ;
+0x05, ADD32,    RRR,  "Addition (32b)"                              ;
+0x06, ADD64,    RRR,  "Addition (64b)"                              ;
+0x07, SUB8,     RRR,  "Subtraction (8b)"                            ;
+0x08, SUB16,    RRR,  "Subtraction (16b)"                           ;
+0x09, SUB32,    RRR,  "Subtraction (32b)"                           ;
+0x0A, SUB64,    RRR,  "Subtraction (64b)"                           ;
+0x0B, MUL8,     RRR,  "Multiplication (8b)"                         ;
+0x0C, MUL16,    RRR,  "Multiplication (16b)"                        ;
+0x0D, MUL32,    RRR,  "Multiplication (32b)"                        ;
+0x0E, MUL64,    RRR,  "Multiplication (64b)"                        ;
+0x0F, AND,      RRR,  "Bitand"                                      ;
+0x10, OR,       RRR,  "Bitor"                                       ;
+0x11, XOR,      RRR,  "Bitxor"                                      ;
+0x12, SLU8,     RRR,  "Unsigned left bitshift (8b)"                 ;
+0x13, SLU16,    RRR,  "Unsigned left bitshift (16b)"                ;
+0x14, SLU32,    RRR,  "Unsigned left bitshift (32b)"                ;
+0x15, SLU64,    RRR,  "Unsigned left bitshift (64b)"                ;
+0x16, SRU8,     RRR,  "Unsigned right bitshift (8b)"                ;
+0x17, SRU16,    RRR,  "Unsigned right bitshift (16b)"               ;
+0x18, SRU32,    RRR,  "Unsigned right bitshift (32b)"               ;
+0x19, SRU64,    RRR,  "Unsigned right bitshift (64b)"               ;
+0x1A, SRS8,     RRR,  "Signed right bitshift (8b)"                  ;
+0x1B, SRS16,    RRR,  "Signed right bitshift (16b)"                 ;
+0x1C, SRS32,    RRR,  "Signed right bitshift (32b)"                 ;
+0x1D, SRS64,    RRR,  "Signed right bitshift (64b)"                 ;
+0x1E, CMPU,     RRR,  "Unsigned comparsion"                         ;
+0x1F, CMPS,     RRR,  "Signed comparsion"                           ;
+0x20, DIRU8,    RRRR, "Merged divide-remainder (unsigned 8b)"       ;
+0x21, DIRU16,   RRRR, "Merged divide-remainder (unsigned 16b)"      ;
+0x22, DIRU32,   RRRR, "Merged divide-remainder (unsigned 32b)"      ;
+0x23, DIRU64,   RRRR, "Merged divide-remainder (unsigned 64b)"      ;
+0x24, DIRS8,    RRRR, "Merged divide-remainder (signed 8b)"         ;
+0x25, DIRS16,   RRRR, "Merged divide-remainder (signed 16b)"        ;
+0x26, DIRS32,   RRRR, "Merged divide-remainder (signed 32b)"        ;
+0x27, DIRS64,   RRRR, "Merged divide-remainder (signed 64b)"        ;
+0x28, NEG,      RR,   "Bit negation"                                ;
+0x29, NOT,      RR,   "Logical negation"                            ;
+0x2A, SXT8,     RR,   "Sign extend 8b to 64b"                       ;
+0x2B, SXT16,    RR,   "Sign extend 16b to 64b"                      ;
+0x2C, SXT32,    RR,   "Sign extend 32b to 64b"                      ;
+0x2D, ADDI8,    RRB,  "Addition with immediate (8b)"                ;
+0x2E, ADDI16,   RRH,  "Addition with immediate (16b)"               ;
+0x2F, ADDI32,   RRW,  "Addition with immediate (32b)"               ;
+0x30, ADDI64,   RRD,  "Addition with immediate (64b)"               ;
+0x31, MULI8,    RRB,  "Multiplication with immediate (8b)"          ;
+0x32, MULI16,   RRH,  "Multiplication with immediate (16b)"         ;
+0x33, MULI32,   RRW,  "Multiplication with immediate (32b)"         ;
+0x34, MULI64,   RRD,  "Multiplication with immediate (64b)"         ;
+0x35, ANDI,     RRD,  "Bitand with immediate"                       ;
+0x36, ORI,      RRD,  "Bitor with immediate"                        ;
+0x37, XORI,     RRD,  "Bitxor with immediate"                       ;
+0x38, SLUI8,    RRB,  "Unsigned left bitshift with immedidate (8b)" ;
+0x39, SLUI16,   RRB,  "Unsigned left bitshift with immedidate (16b)";
+0x3A, SLUI32,   RRB,  "Unsigned left bitshift with immedidate (32b)";
+0x3B, SLUI64,   RRB,  "Unsigned left bitshift with immedidate (64b)";
+0x3C, SRUI8,    RRB,  "Unsigned right bitshift with immediate (8b)" ;
+0x3D, SRUI16,   RRB,  "Unsigned right bitshift with immediate (16b)";
+0x3E, SRUI32,   RRB,  "Unsigned right bitshift with immediate (32b)";
+0x3F, SRUI64,   RRB,  "Unsigned right bitshift with immediate (64b)";
+0x40, SRSI8,    RRB,  "Signed right bitshift with immediate"        ;
+0x41, SRSI16,   RRB,  "Signed right bitshift with immediate"        ;
+0x42, SRSI32,   RRB,  "Signed right bitshift with immediate"        ;
+0x43, SRSI64,   RRB,  "Signed right bitshift with immediate"        ;
+0x44, CMPUI,    RRD,  "Unsigned compare with immediate"             ;
+0x45, CMPSI,    RRD,  "Signed compare with immediate"               ;
+0x46, CP,       RR,   "Copy register"                               ;
+0x47, SWA,      RR,   "Swap registers"                              ;
+0x48, LI8,      RB,   "Load immediate (8b)"                         ;
+0x49, LI16,     RH,   "Load immediate (16b)"                        ;
+0x4A, LI32,     RW,   "Load immediate (32b)"                        ;
+0x4B, LI64,     RD,   "Load immediate (64b)"                        ;
+0x4C, LRA,      RRO,  "Load relative address"                       ;
+0x4D, LD,       RRAH, "Load from absolute address"                  ;
+0x4E, ST,       RRAH, "Store to absolute address"                   ;
+0x4F, LDR,      RROH, "Load from relative address"                  ;
+0x50, STR,      RROH, "Store to relative address"                   ;
+0x51, BMC,      RRH,  "Copy block of memory"                        ;
+0x52, BRC,      RRB,  "Copy register block"                         ;
+0x53, JMP,      O,    "Relative jump"                               ;
+0x54, JAL,      RRO,  "Linking relative jump"                       ;
+0x55, JALA,     RRA,  "Linking absolute jump"                       ;
+0x56, JEQ,      RRP,  "Branch on equal"                             ;
+0x57, JNE,      RRP,  "Branch on nonequal"                          ;
+0x58, JLTU,     RRP,  "Branch on lesser-than (unsigned)"            ;
+0x59, JGTU,     RRP,  "Branch on greater-than (unsigned)"           ;
+0x5A, JLTS,     RRP,  "Branch on lesser-than (signed)"              ;
+0x5B, JGTS,     RRP,  "Branch on greater-than (signed)"             ;
+0x5C, ECA,      N,    "Environment call trap"                       ;
+0x5D, EBP,      N,    "Environment breakpoint"                      ;
+0x5E, FADD32,   RRR,  "Floating point addition (32b)"               ;
+0x5F, FADD64,   RRR,  "Floating point addition (64b)"               ;
+0x60, FSUB32,   RRR,  "Floating point subtraction (32b)"            ;
+0x61, FSUB64,   RRR,  "Floating point subtraction (64b)"            ;
+0x62, FMUL32,   RRR,  "Floating point multiply (32b)"               ;
+0x63, FMUL64,   RRR,  "Floating point multiply (64b)"               ;
+0x64, FDIV32,   RRR,  "Floating point division (32b)"               ;
+0x65, FDIV64,   RRR,  "Floating point division (64b)"               ;
+0x66, FMA32,    RRRR, "Float fused multiply-add (32b)"              ;
+0x67, FMA64,    RRRR, "Float fused multiply-add (64b)"              ;
+0x68, FINV32,   RR,   "Float reciprocal (32b)"                      ;
+0x69, FINV64,   RR,   "Float reciprocal (64b)"                      ;
+0x6A, FCMPLT32, RRR,  "Flaot compare less than (32b)"               ;
+0x6B, FCMPLT64, RRR,  "Flaot compare less than (64b)"               ;
+0x6C, FCMPGT32, RRR,  "Flaot compare greater than (32b)"            ;
+0x6D, FCMPGT64, RRR,  "Flaot compare greater than (64b)"            ;
+0x6E, ITF32,    RR,   "Int to 32 bit float"                         ;
+0x6F, ITF64,    RR,   "Int to 64 bit float"                         ;
+0x70, FTI32,    RRB,  "Float 32 to int"                             ;
+0x71, FTI64,    RRB,  "Float 64 to int"                             ;
+0x72, FC32T64,  RR,   "Float 64 to Float 32"                        ;
+0x73, FC64T32,  RRB,  "Float 32 to Float 64"                        ;
+0x74, LRA16,    RRP,  "Load relative immediate (16 bit)"            ;
+0x75, LDR16,    RRPH, "Load from relative address (16 bit)"         ;
+0x76, STR16,    RRPH, "Store to relative address (16 bit)"          ;
+0x77, JMP16,    P,    "Relative jump (16 bit)"                      ;

hbbytecode/src/lib.rs

@@ -0,0 +1,185 @@
+#![no_std]
+
+use core::convert::TryFrom;
+
+type OpR = u8;
+
+type OpA = u64;
+type OpO = i32;
+type OpP = i16;
+
+type OpB = u8;
+type OpH = u16;
+type OpW = u32;
+type OpD = u64;
+
+/// # Safety
+/// Has to be valid to be decoded from bytecode.
+pub unsafe trait BytecodeItem {}
+macro_rules! define_items {
+    ($($name:ident ($($nm:ident: $item:ident),* $(,)?)),* $(,)?) => {
+        $(
+            #[derive(Clone, Copy, Debug)]
+            #[repr(packed)]
+            pub struct $name($(pub $item),*);
+            unsafe impl BytecodeItem for $name {}
+
+            impl Encodable for $name {
+                fn encode(self, _buffer: &mut impl Buffer) {
+                    let Self($($nm),*) = self;
+                    $(
+                        for byte in $nm.to_le_bytes() {
+                            unsafe { _buffer.write(byte) };
+                        }
+                    )*
+                }
+
+                fn encode_len(self) -> usize {
+                    core::mem::size_of::<Self>()
+                }
+            }
+        )*
+    };
+}
+
+define_items! {
+    OpsRR   (a: OpR, b: OpR                ),
+    OpsRRR  (a: OpR, b: OpR, c: OpR        ),
+    OpsRRRR (a: OpR, b: OpR, c: OpR, d: OpR),
+    OpsRRB  (a: OpR, b: OpR, c: OpB        ),
+    OpsRRH  (a: OpR, b: OpR, c: OpH        ),
+    OpsRRW  (a: OpR, b: OpR, c: OpW        ),
+    OpsRRD  (a: OpR, b: OpR, c: OpD        ),
+    OpsRB   (a: OpR, b: OpB                ),
+    OpsRH   (a: OpR, b: OpH                ),
+    OpsRW   (a: OpR, b: OpW                ),
+    OpsRD   (a: OpR, b: OpD                ),
+    OpsRRA  (a: OpR, b: OpR, c: OpA        ),
+    OpsRRAH (a: OpR, b: OpR, c: OpA, d: OpH),
+    OpsRROH (a: OpR, b: OpR, c: OpO, d: OpH),
+    OpsRRPH (a: OpR, b: OpR, c: OpP, d: OpH),
+    OpsRRO  (a: OpR, b: OpR, c: OpO        ),
+    OpsRRP  (a: OpR, b: OpR, c: OpP        ),
+    OpsO    (a: OpO,                       ),
+    OpsP    (a: OpP,                       ),
+    OpsN    (                              ),
+}
+
+unsafe impl BytecodeItem for u8 {}
+
+::with_builtin_macros::with_builtin! {
+    let $spec = include_from_root!("instructions.in") in {
+        /// Invoke macro with bytecode definition
+        ///
+        /// # Format
+        /// ```text
+        /// Opcode, Mnemonic, Type, Docstring;
+        /// ```
+        ///
+        /// # Type
+        /// ```text
+        /// Types consist of letters meaning a single field
+        /// | Type | Size (B) | Meaning                 |
+        /// |:-----|:---------|:------------------------|
+        /// | N    | 0        | Empty                   |
+        /// | R    | 1        | Register                |
+        /// | A    | 8        | Absolute address        |
+        /// | O    | 4        | Relative address offset |
+        /// | P    | 2        | Relative address offset |
+        /// | B    | 1        | Immediate               |
+        /// | H    | 2        | Immediate               |
+        /// | W    | 4        | Immediate               |
+        /// | D    | 8        | Immediate               |
+        /// ```
+        #[macro_export]
+        macro_rules! invoke_with_def {
+            ($($macro:tt)*) => {
+                $($macro)*! { $spec }
+            };
+        }
+    }
+}
+
+pub trait Buffer {
+    fn reserve(&mut self, bytes: usize);
+    /// # Safety
+    /// Reserve needs to be called before this function, and only reserved amount can be written.
+    unsafe fn write(&mut self, byte: u8);
+}
+
+pub trait Encodable {
+    fn encode(self, buffer: &mut impl Buffer);
+    fn encode_len(self) -> usize;
+}
+
+macro_rules! gen_opcodes {
+    ($($opcode:expr, $mnemonic:ident, $ty:ident, $doc:literal;)*) => {
+        pub mod opcode {
+            $(
+                #[doc = $doc]
+                pub const $mnemonic: u8 = $opcode;
+            )*
+
+            paste::paste! {
+                #[derive(Clone, Copy, Debug)]
+                #[repr(u8)]
+                pub enum Op { $(
+                    [< $mnemonic:lower:camel >](super::[<Ops $ty>]),
+                )* }
+
+                impl Op {
+                    pub fn size(&self) -> usize {
+                        (match self {
+                            $(Self::[<$mnemonic:lower:camel>] { .. } => core::mem::size_of::<super::[<Ops $ty>]>(),)*
+                        }) + 1
+                    }
+                }
+
+                impl crate::Encodable for Op {
+                    fn encode(self, buffer: &mut impl crate::Buffer) {
+                        match self {
+                            $(
+                                Self::[< $mnemonic:lower:camel >](op) => {
+                                    unsafe { buffer.write($opcode) };
+                                    op.encode(buffer);
+                                }
+                            )*
+                        }
+                    }
+
+                    fn encode_len(self) -> usize {
+                        match self {
+                            $(
+                                Self::[< $mnemonic:lower:camel >](op) => {
+                                    1 + crate::Encodable::encode_len(op)
+                                }
+                            )*
+                        }
+                    }
+                }
+            }
+        }
+    };
+}
+
+/// Rounding mode
+#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+#[repr(u8)]
+pub enum RoundingMode {
+    NearestEven = 0,
+    Truncate = 1,
+    Up       = 2,
+    Down     = 3,
+}
+
+impl TryFrom<u8> for RoundingMode {
+    type Error = ();
+
+    fn try_from(value: u8) -> Result<Self, Self::Error> {
+        (value <= 3)
+            .then(|| unsafe { core::mem::transmute(value) })
+            .ok_or(())
+    }
+}
+
+invoke_with_def!(gen_opcodes);

hblang/Cargo.toml

@@ -1,8 +1,11 @@
 [package]
-name = "compiler"
+name = "hblang"
 version = "0.1.0"
 edition = "2021"
 
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
+hbbytecode = { version = "0.1.0", path = "../hbbytecode" }
+logos = "0.13.0"
+

hblang/src/codegen.rs

@@ -0,0 +1,595 @@
+use std::{iter::Cycle, ops::Range, usize};
+
+use crate::{
+    lexer::{self, Ty},
+    parser::{Exp, Function, Item, Literal, Struct, Type},
+};
+
+type Reg = u8;
+type Offset = i32;
+type Pushed = bool;
+type SlotIndex = usize;
+type Label = usize;
+type Data = usize;
+type Size = usize;
+
+//| Register   | Description         | Saver  |
+//|:-----------|:--------------------|:-------|
+//| r0         | Hard-wired zero     | N/A    |
+//| r1 - r2    | Return values       | Caller |
+//| r2 - r11   | Function parameters | Caller |
+//| r12 - r30  | General purpose     | Caller |
+//| r31        | Return address      | Caller |
+//| r32 - r253 | General purpose     | Callee |
+//| r254       | Stack pointer       | Callee |
+//| r255       | Thread pointer      | N/A    |
+
+struct RegAlloc {
+    pub regs:        Box<[Option<usize>; 256]>,
+    pub used:        Box<[bool; 256]>,
+    pub spill_cycle: Cycle<Range<u8>>,
+}
+
+impl RegAlloc {
+    const STACK_POINTER: Reg = 254;
+    const ZERO: Reg = 0;
+    const RETURN_ADDRESS: Reg = 31;
+
+    fn alloc_return(&mut self, slot: usize) -> Option<Reg> {
+        self.regs[1..2]
+            .iter_mut()
+            .position(|reg| {
+                if reg.is_none() {
+                    *reg = Some(slot);
+                    true
+                } else {
+                    false
+                }
+            })
+            .map(|reg| reg as Reg + 1)
+    }
+
+    fn alloc_general(&mut self, slot: usize) -> Option<Reg> {
+        self.regs[32..254]
+            .iter_mut()
+            .zip(&mut self.used[32..254])
+            .position(|(reg, used)| {
+                if reg.is_none() {
+                    *reg = Some(slot);
+                    *used = true;
+                    true
+                } else {
+                    false
+                }
+            })
+            .map(|reg| reg as Reg + 32)
+    }
+
+    fn free(&mut self, reg: Reg) {
+        assert!(self.regs[reg as usize].take().is_some());
+    }
+
+    fn is_used(&self, reg: Reg) -> bool {
+        self.regs[reg as usize].is_some()
+    }
+
+    fn spill(&mut self, for_slot: usize) -> (Reg, Option<usize>) {
+        let to_spill = self.spill_cycle.next().unwrap();
+        let slot = self.spill_specific(to_spill, for_slot);
+        (to_spill as Reg + 32, slot)
+    }
+
+    fn spill_specific(&mut self, reg: Reg, for_slot: usize) -> Option<usize> {
+        self.regs[reg as usize].replace(for_slot)
+    }
+
+    fn restore(&mut self, reg: Reg, slot: usize) -> usize {
+        self.regs[reg as usize].replace(slot).unwrap()
+    }
+
+    fn alloc_specific(&mut self, reg: u8, to: usize) {
+        assert!(self.regs[reg as usize].replace(to).is_none());
+    }
+
+    fn alloc_specific_in_reg(&mut self, reg: InReg, to: usize) {
+        match reg {
+            InReg::Single(r) => self.alloc_specific(r, to),
+            InReg::Pair(r1, r2) => {
+                self.alloc_specific(r1, to);
+                self.alloc_specific(r2, to);
+            }
+        }
+    }
+}
+
+pub struct ParamAlloc {
+    reg_range: Range<Reg>,
+    stack:     Offset,
+}
+
+impl ParamAlloc {
+    fn new() -> Self {
+        Self {
+            stack:     8, // return adress is in callers stack frame
+            reg_range: 2..12,
+        }
+    }
+
+    fn alloc(&mut self, size: usize) -> SlotValue {
+        match self.try_alloc_regs(size) {
+            Some(reg) => reg,
+            None => {
+                let stack = self.stack;
+                self.stack += size as Offset;
+                SlotValue::Stack(stack)
+            }
+        }
+    }
+
+    fn try_alloc_regs(&mut self, size: usize) -> Option<SlotValue> {
+        let mut needed = size.div_ceil(8);
+        if needed > 2 {
+            needed = 1; // passed by ref
+        }
+
+        if self.reg_range.len() < needed {
+            return None;
+        }
+
+        match needed {
+            1 => {
+                let reg = self.reg_range.start;
+                self.reg_range.start += 1;
+                Some(SlotValue::Reg(InReg::Single(reg)))
+            }
+            2 => {
+                let reg = self.reg_range.start;
+                self.reg_range.start += 2;
+                Some(SlotValue::Reg(InReg::Pair(reg, reg + 1)))
+            }
+            _ => unreachable!(),
+        }
+    }
+}
+
+impl Default for RegAlloc {
+    fn default() -> Self {
+        Self {
+            regs:        Box::new([None; 256]),
+            used:        Box::new([false; 256]),
+            spill_cycle: (32..254).cycle(),
+        }
+    }
+}
+
+struct Variable {
+    name:     String,
+    location: usize,
+}
+
+#[derive(Clone, Copy)]
+struct SlotId {
+    // index into slot stack
+    index:  SlotIndex,
+    // temorary offset carried over when eg. accessing fields
+    offset: Offset,
+    // this means we can mutate the value as part of computation
+    owned:  bool,
+}
+
+impl SlotId {
+    fn base(location: usize) -> Self {
+        Self {
+            index:  location,
+            offset: 0,
+            owned:  true,
+        }
+    }
+
+    fn borrowed(self) -> Self {
+        Self {
+            owned: false,
+            ..self
+        }
+    }
+}
+
+struct Slot {
+    ty:    Type,
+    value: SlotValue,
+}
+
+#[repr(transparent)]
+struct InstBuffer {
+    buffer: Vec<u8>,
+}
+
+impl InstBuffer {
+    fn new(vec: &mut Vec<u8>) -> &mut Self {
+        unsafe { &mut *(vec as *mut Vec<u8> as *mut Self) }
+    }
+}
+
+impl hbbytecode::Buffer for InstBuffer {
+    fn reserve(&mut self, bytes: usize) {
+        self.buffer.reserve(bytes);
+    }
+
+    unsafe fn write(&mut self, byte: u8) {
+        self.buffer.push(byte);
+    }
+}
+
+#[derive(Clone, Copy)]
+enum InReg {
+    Single(Reg),
+    // if one of the registes is allocated, the other is too, ALWAYS
+    // with the same slot
+    Pair(Reg, Reg),
+}
+
+#[derive(Clone, Copy)]
+enum Spill {
+    Reg(InReg),
+    Stack(Offset), // relative to frame end (rsp if nothing was pushed)
+}
+
+#[derive(Clone, Copy)]
+enum SlotValue {
+    Reg(InReg),
+    Stack(Offset), // relative to frame start (rbp)
+    Imm(u64),
+    Spilled(Spill, SlotIndex),
+}
+
+pub struct Value {
+    store:  ValueStore,
+    offset: Offset,
+}
+
+#[derive(Clone, Copy)]
+enum ValueStore {
+    Reg(InReg),
+    Stack(Offset, Pushed),
+    Imm(u64),
+}
+
+impl From<SlotValue> for ValueStore {
+    fn from(value: SlotValue) -> Self {
+        match value {
+            SlotValue::Reg(reg) => ValueStore::Reg(reg),
+            SlotValue::Stack(offset) => ValueStore::Stack(offset, false),
+            SlotValue::Imm(imm) => ValueStore::Imm(imm),
+            SlotValue::Spilled(spill, _) => match spill {
+                Spill::Reg(reg) => ValueStore::Reg(reg),
+                Spill::Stack(offset) => ValueStore::Stack(offset, true),
+            },
+        }
+    }
+}
+
+pub struct LabelReloc {
+    pub label:  Label,
+    pub offset: usize,
+}
+
+pub struct DataReloc {
+    pub data:   Data,
+    pub offset: usize,
+}
+
+#[must_use]
+pub struct Frame {
+    pub slot_count: usize,
+    pub var_count:  usize,
+}
+
+enum Instr {
+    BinOp(lexer::Op, Value, Value),
+    Move(Size, Value, Value),
+    Push(Reg),
+    Jump(Label),
+    Call(String),
+    JumpIfZero(Value, Label),
+}
+
+#[derive(Default)]
+pub struct Generator<'a> {
+    ast: &'a [Item],
+
+    func_labels: Vec<(String, Label)>,
+
+    stack_size:  Offset,
+    pushed_size: Offset,
+
+    regs:      RegAlloc,
+    variables: Vec<Variable>,
+    slots:     Vec<Slot>,
+
+    labels:       Vec<Option<usize>>,
+    label_relocs: Vec<LabelReloc>,
+
+    data:        Vec<Option<usize>>,
+    data_relocs: Vec<DataReloc>,
+
+    code_section: Vec<u8>,
+    data_section: Vec<u8>,
+
+    instrs: Vec<Instr>,
+}
+
+impl<'a> Generator<'a> {
+    fn generate(mut self) -> Vec<u8> {
+        for item in self.ast {
+            let Item::Function(f) = item else { continue };
+            self.generate_function(f);
+        }
+
+        self.link()
+    }
+
+    fn generate_function(&mut self, f: &Function) {
+        let frame = self.push_frame();
+
+        let mut param_alloc = ParamAlloc::new();
+
+        for param in f.args.iter() {
+            let param_size = self.size_of(&param.ty);
+            let value = param_alloc.alloc(param_size);
+            let slot = self.add_slot(param.ty.clone(), value);
+            if let SlotValue::Reg(reg) = value {
+                self.regs.alloc_specific_in_reg(reg, slot);
+            }
+            self.add_variable(param.name.clone(), slot);
+        }
+
+        for stmt in f.body.iter() {
+            assert!(self
+                .generate_expr(Some(Type::Builtin(Ty::Void)), stmt)
+                .is_none());
+        }
+
+        self.pop_frame(frame);
+    }
+
+    fn generate_expr(&mut self, expected: Option<Type>, expr: &Exp) -> Option<SlotId> {
+        let value = match expr {
+            Exp::Literal(lit) => SlotId::base(match lit {
+                Literal::Int(i) => self.add_slot(expected.clone().unwrap(), SlotValue::Imm(*i)),
+                Literal::Bool(b) => {
+                    self.add_slot(Type::Builtin(Ty::Bool), SlotValue::Imm(*b as u64))
+                }
+            }),
+            Exp::Variable(ident) => {
+                SlotId::base(self.lookup_variable(ident).unwrap().location).borrowed()
+            }
+            Exp::Call { name, args } => self.generate_call(expected.clone(), name, args),
+            Exp::Ctor { name, fields } => todo!(),
+            Exp::Index { base, index } => todo!(),
+            Exp::Field { base, field } => todo!(),
+            Exp::Unary { op, exp } => todo!(),
+            Exp::Binary { op, left, right } => todo!(),
+            Exp::If { cond, then, else_ } => todo!(),
+            Exp::Let { name, ty, value } => todo!(),
+            Exp::For {
+                init,
+                cond,
+                step,
+                block,
+            } => todo!(),
+            Exp::Block(_) => todo!(),
+            Exp::Return(_) => todo!(),
+            Exp::Break => todo!(),
+            Exp::Continue => todo!(),
+        };
+
+        if let Some(expected) = expected {
+            let actual = self.slots[value.index].ty.clone();
+            assert_eq!(expected, actual);
+        }
+
+        Some(value)
+    }
+
+    fn generate_call(&mut self, expected: Option<Type>, name: &str, args: &[Exp]) -> SlotId {
+        let frame = self.push_frame();
+        let func = self.lookup_function(name);
+
+        let mut arg_alloc = ParamAlloc::new();
+        for (arg, param) in args.iter().zip(&func.args) {
+            let arg_slot = self.generate_expr(Some(param.ty.clone()), arg).unwrap();
+            let arg_size = self.size_of(&param.ty);
+            let param_slot = arg_alloc.alloc(arg_size);
+            self.set_temporarly(arg_slot, param_slot);
+        }
+
+        self.instrs.push(Instr::Call(name.to_owned()));
+
+        todo!()
+    }
+
+    fn set_temporarly(&mut self, from: SlotId, to: SlotValue) {
+        let to = self.make_mutable(to, from.index);
+        let to_slot = self.add_slot(self.slots[from.index].ty.clone(), to);
+        self.emit_move(from, SlotId::base(to_slot));
+    }
+
+    fn make_mutable(&mut self, target: SlotValue, by: SlotIndex) -> SlotValue {
+        match target {
+            SlotValue::Reg(in_reg) => {
+                self.regs.alloc_specific_in_reg(in_reg, by);
+                target
+            }
+            SlotValue::Spilled(Spill::Reg(in_reg), slot) => {
+                let new_val = SlotValue::Spilled(
+                    match in_reg {
+                        InReg::Single(reg) => Spill::Stack(self.emmit_push(reg)),
+                        InReg::Pair(r1, r2) => {
+                            self.emmit_push(r2);
+                            Spill::Stack(self.emmit_push(r1))
+                        }
+                    },
+                    slot,
+                );
+                let new_slot = self.add_slot(self.slots[slot].ty.clone(), new_val);
+                SlotValue::Spilled(Spill::Reg(in_reg), new_slot)
+            }
+            _ => unreachable!(),
+        }
+    }
+
+    fn emmit_push(&mut self, reg: Reg) -> Offset {
+        self.pushed_size += 8;
+        self.instrs.push(Instr::Push(reg));
+        self.pushed_size
+    }
+
+    fn emit_move(&mut self, from: SlotId, to: SlotId) {
+        let size = self.size_of(&self.slots[from.index].ty);
+        let other_size = self.size_of(&self.slots[to.index].ty);
+        assert_eq!(size, other_size);
+
+        self.instrs.push(Instr::Move(
+            size,
+            self.slot_to_value(from),
+            self.slot_to_value(to),
+        ));
+    }
+
+    fn slot_to_value(&self, slot: SlotId) -> Value {
+        let slot_val = &self.slots[slot.index];
+        Value {
+            store:  slot_val.value.into(),
+            offset: slot.offset,
+        }
+    }
+
+    fn size_of(&self, ty: &Type) -> Size {
+        match ty {
+            Type::Builtin(ty) => match ty {
+                Ty::U8 | Ty::I8 | Ty::Bool => 1,
+                Ty::U16 | Ty::I16 => 2,
+                Ty::U32 | Ty::I32 => 4,
+                Ty::U64 | Ty::I64 => 8,
+                Ty::Void => 0,
+            },
+            Type::Struct(name) => self
+                .lookup_struct(name)
+                .fields
+                .iter()
+                .map(|field| self.size_of(&field.ty))
+                .sum(),
+            Type::Pinter(_) => 8,
+        }
+    }
+}
+
+impl<'a> Generator<'a> {
+    fn add_variable(&mut self, name: String, location: usize) {
+        self.variables.push(Variable { name, location });
+    }
+
+    fn add_slot(&mut self, ty: Type, value: SlotValue) -> usize {
+        let slot = self.slots.len();
+        self.slots.push(Slot { ty, value });
+        slot
+    }
+
+    fn link(mut self) -> Vec<u8> {
+        for reloc in self.label_relocs {
+            let label = self.labels[reloc.label].unwrap();
+            let offset = reloc.offset;
+            let target = label - offset;
+            let target_bytes = u64::to_le_bytes(target as u64);
+            self.code_section[offset..offset + 8].copy_from_slice(&target_bytes);
+        }
+
+        for reloc in self.data_relocs {
+            let data = self.data[reloc.data].unwrap();
+            let offset = reloc.offset;
+            let target = data;
+            let target_bytes = u64::to_le_bytes((target + self.code_section.len()) as u64);
+            self.data_section[offset..offset + 8].copy_from_slice(&target_bytes);
+        }
+
+        self.code_section.extend_from_slice(&self.data_section);
+        self.code_section
+    }
+
+    fn lookup_func_label(&mut self, name: &str) -> Label {
+        if let Some(label) = self.func_labels.iter().find(|(n, _)| n == name) {
+            return label.1;
+        }
+
+        panic!("Function not found: {}", name);
+    }
+
+    fn declare_label(&mut self) -> Label {
+        self.labels.push(None);
+        self.labels.len() - 1
+    }
+
+    fn define_label(&mut self, label: Label) {
+        self.labels[label] = Some(self.code_section.len());
+    }
+
+    fn declare_data(&mut self) -> Data {
+        self.data.push(None);
+        self.data.len() - 1
+    }
+
+    fn define_data(&mut self, data: Data, bytes: &[u8]) {
+        self.data[data] = Some(self.data.len());
+        self.data_section.extend_from_slice(bytes);
+    }
+
+    fn lookup_struct(&self, name: &str) -> &Struct {
+        self.lookup_item(name)
+            .map(|item| match item {
+                Item::Struct(s) => s,
+                _ => panic!("Not a struct: {}", name),
+            })
+            .expect("Struct not found")
+    }
+
+    fn lookup_function(&self, name: &str) -> &'a Function {
+        self.lookup_item(name)
+            .map(|item| match item {
+                Item::Function(f) => f,
+                _ => panic!("Not a function: {}", name),
+            })
+            .expect("Function not found")
+    }
+
+    fn lookup_item(&self, name: &str) -> Option<&'a Item> {
+        self.ast.iter().find(|item| match item {
+            Item::Import(_) => false,
+            Item::Struct(s) => s.name == name,
+            Item::Function(f) => f.name == name,
+        })
+    }
+
+    fn lookup_variable(&self, name: &str) -> Option<&Variable> {
+        self.variables.iter().find(|variable| variable.name == name)
+    }
+
+    fn push_frame(&mut self) -> Frame {
+        Frame {
+            slot_count: self.slots.len(),
+            var_count:  self.variables.len(),
+        }
+    }
+
+    fn pop_frame(&mut self, frame: Frame) {
+        self.slots.truncate(frame.slot_count);
+        self.variables.truncate(frame.var_count);
+    }
+}
+
+pub fn generate(ast: &[Item]) -> Vec<u8> {
+    Generator {
+        ast,
+        ..Default::default()
+    }
+    .generate()
+}

hblang/src/lexer.rs

@@ -0,0 +1,151 @@
+use logos::Logos;
+
+macro_rules! gen_token {
+    ($name:ident {
+        keywords: {
+            $($keyword:ident = $lit:literal,)*
+        },
+        operators: $op_name:ident {
+            $($prec:literal: {$(
+                $op:ident = $op_lit:literal,
+            )*},)*
+        },
+        types: $ty_type:ident {
+            $($ty:ident = $ty_lit:literal,)*
+        },
+        regexes: {
+            $($regex:ident = $regex_lit:literal,)*
+        },
+    }) => {
+        #[derive(Debug, Clone, PartialEq, Eq, Copy, Logos)]
+        #[logos(skip "[ \t\n]+")]
+        pub enum $name {
+            $(#[token($lit)] $keyword,)*
+            $($(#[token($op_lit, |_| $op_name::$op)])*)*
+            Op($op_name),
+            $(#[token($ty_lit, |_| $ty_type::$ty)])*
+            Ty($ty_type),
+            $(#[regex($regex_lit)] $regex,)*
+        }
+
+        #[derive(Debug, Clone, PartialEq, Eq, Copy)]
+        pub enum $op_name {
+            $($($op,)*)*
+        }
+
+        #[derive(Debug, Clone, PartialEq, Eq, Copy)]
+        pub enum $ty_type {
+            $($ty,)*
+        }
+
+        impl $op_name {
+            pub fn prec(&self) -> u8 {
+                match self {
+                    $($($op_name::$op => $prec,)*)*
+                }
+            }
+        }
+    };
+}
+
+gen_token! {
+    TokenKind {
+        keywords: {
+            Use = "use",
+            Fn = "fn",
+            Let = "let",
+            If = "if",
+            Else = "else",
+            For = "for",
+            Return = "return",
+            Break = "break",
+            Continue = "continue",
+            Struct = "struct",
+
+            True = "true",
+            False = "false",
+
+            LBrace = "{",
+            RBrace = "}",
+            LParen = "(",
+            RParen = ")",
+            LBracket = "[",
+            RBracket = "]",
+
+            Colon = ":",
+            Semicolon = ";",
+            Comma = ",",
+            Dot = ".",
+        },
+        operators: Op {
+            14: {
+                Assign = "=",
+                AddAssign = "+=",
+                SubAssign = "-=",
+                MulAssign = "*=",
+                DivAssign = "/=",
+                ModAssign = "%=",
+                AndAssign = "&=",
+                OrAssign = "|=",
+                XorAssign = "^=",
+                ShlAssign = "<<=",
+                ShrAssign = ">>=",
+            },
+            12: {
+                Or = "||",
+            },
+            11: {
+                And = "&&",
+            },
+            10: {
+                Bor = "|",
+            },
+            9: {
+                Xor = "^",
+            },
+            8: {
+                Band = "&",
+            },
+            7: {
+                Eq = "==",
+                Neq = "!=",
+            },
+            6: {
+                Lt = "<",
+                Gt = ">",
+                Le = "<=",
+                Ge = ">=",
+            },
+            5: {
+                Shl = "<<",
+                Shr = ">>",
+            },
+            4: {
+                Add = "+",
+                Sub = "-",
+            },
+            3: {
+                Mul = "*",
+                Div = "/",
+                Mod = "%",
+            },
+        },
+        types: Ty {
+            U8 = "u8",
+            U16 = "u16",
+            U32 = "u32",
+            U64 = "u64",
+            I8 = "i8",
+            I16 = "i16",
+            I32 = "i32",
+            I64 = "i64",
+            Bool = "bool",
+            Void = "void",
+        },
+        regexes: {
+            Ident = "[a-zA-Z_][a-zA-Z0-9_]*",
+            String = r#""([^"\\]|\\.)*""#,
+            Number = "[0-9]+",
+        },
+    }
+}

hblang/src/lib.rs

@@ -0,0 +1,6 @@
+#![allow(dead_code)]
+
+mod codegen;
+mod lexer;
+mod parser;
+mod typechk;

hblang/src/parser.rs

@@ -0,0 +1,566 @@
+use {core::panic, std::iter};
+
+use std::array;
+
+use logos::{Lexer, Logos};
+
+use crate::lexer::{Op, TokenKind, Ty};
+
+#[derive(Clone, Debug)]
+pub enum Item {
+    Import(String),
+    Struct(Struct),
+    Function(Function),
+}
+
+#[derive(Clone, Debug, PartialEq, Eq)]
+pub enum Type {
+    Builtin(Ty),
+    Struct(String),
+    Pinter(Box<Type>),
+}
+
+#[derive(Clone, Debug)]
+pub struct Struct {
+    pub name:   String,
+    pub fields: Vec<Field>,
+}
+
+#[derive(Clone, Debug)]
+pub struct Field {
+    pub name: String,
+    pub ty:   Type,
+}
+
+#[derive(Clone, Debug)]
+pub struct Function {
+    pub name: String,
+    pub args: Vec<Arg>,
+    pub ret:  Type,
+    pub body: Vec<Exp>,
+}
+
+#[derive(Clone, Debug)]
+pub struct Arg {
+    pub name: String,
+    pub ty:   Type,
+}
+
+#[derive(Clone, Debug)]
+pub struct CtorField {
+    pub name:  String,
+    pub value: Exp,
+}
+
+#[derive(Clone, Debug)]
+pub enum Exp {
+    Literal(Literal),
+    Variable(String),
+    Call {
+        name: String,
+        args: Vec<Exp>,
+    },
+    Ctor {
+        name:   Option<Box<Exp>>,
+        fields: Vec<CtorField>,
+    },
+    Index {
+        base:  Box<Exp>,
+        index: Box<Exp>,
+    },
+    Field {
+        base:  Box<Exp>,
+        field: String,
+    },
+    Unary {
+        op:  Op,
+        exp: Box<Exp>,
+    },
+    Binary {
+        op:    Op,
+        left:  Box<Exp>,
+        right: Box<Exp>,
+    },
+    If {
+        cond:  Box<Exp>,
+        then:  Box<Exp>,
+        else_: Option<Box<Exp>>,
+    },
+    Let {
+        name:  String,
+        ty:    Option<Type>,
+        value: Box<Exp>,
+    },
+    For {
+        init:  Option<Box<Exp>>,
+        cond:  Option<Box<Exp>>,
+        step:  Option<Box<Exp>>,
+        block: Box<Exp>,
+    },
+    Block(Vec<Exp>),
+    Return(Option<Box<Exp>>),
+    Break,
+    Continue,
+}
+
+#[derive(Clone, Debug)]
+pub enum Literal {
+    Int(u64),
+    Bool(bool),
+}
+
+#[derive(Debug, PartialEq, Clone)]
+pub struct Token {
+    pub kind:  TokenKind,
+    pub span:  std::ops::Range<usize>,
+    pub value: String,
+}
+
+struct Parser<'a> {
+    next_token: Option<Token>,
+    lexer:      logos::Lexer<'a, TokenKind>,
+}
+
+impl<'a> Parser<'a> {
+    pub fn new(input: &'a str) -> Self {
+        let mut lexer = TokenKind::lexer(input);
+        let next_token = Self::next_token(&mut lexer);
+        Self { next_token, lexer }
+    }
+
+    pub fn next(&mut self) -> Option<Token> {
+        let token = self.next_token.clone();
+        self.next_token = Self::next_token(&mut self.lexer);
+        token
+    }
+
+    pub fn next_token(lexer: &mut Lexer<TokenKind>) -> Option<Token> {
+        lexer.next().map(|r| {
+            r.map(|e| Token {
+                kind:  e,
+                span:  lexer.span(),
+                value: lexer.slice().to_owned(),
+            })
+            .unwrap_or_else(|e| {
+                let (line, col) = Self::pos_to_line_col_low(lexer.source(), lexer.span().start);
+                println!("Lexer error: {}:{}: {:?}", line, col, e);
+                std::process::exit(1);
+            })
+        })
+    }
+
+    pub fn pos_to_line_col(&self, pos: usize) -> (usize, usize) {
+        Self::pos_to_line_col_low(self.lexer.source(), pos)
+    }
+
+    pub fn pos_to_line_col_low(source: &str, pos: usize) -> (usize, usize) {
+        let line = source[..pos].lines().count();
+        let col = source[..pos].lines().last().map(|l| l.len()).unwrap_or(0);
+        (line, col)
+    }
+
+    pub fn expect(&mut self, kind: TokenKind) -> Token {
+        let token = self.expect_any();
+        if token.kind == kind {
+            token
+        } else {
+            let (line, col) = self.pos_to_line_col(token.span.start);
+            panic!(
+                "Expected {:?} at {}:{}, found {:?}",
+                kind, line, col, token.kind
+            )
+        }
+    }
+
+    pub fn expect_any(&mut self) -> Token {
+        self.next().unwrap_or_else(|| panic!("Unexpected EOF"))
+    }
+
+    pub fn peek(&self) -> Option<&Token> {
+        self.next_token.as_ref()
+    }
+
+    pub fn try_advance(&mut self, kind: TokenKind) -> bool {
+        if self.peek().is_some_and(|t| t.kind == kind) {
+            self.next();
+            true
+        } else {
+            false
+        }
+    }
+
+    pub fn parse(&mut self) -> Vec<Item> {
+        iter::from_fn(|| self.parse_item()).collect()
+    }
+
+    fn parse_item(&mut self) -> Option<Item> {
+        let token = self.next()?;
+        match token.kind {
+            TokenKind::Struct => Some(self.parse_struct()),
+            TokenKind::Fn => Some(self.parse_function()),
+            TokenKind::Use => Some(Item::Import(self.expect(TokenKind::String).value)),
+            tkn => {
+                let (line, col) = self.pos_to_line_col(token.span.start);
+                panic!("Unexpected {:?} at {}:{}", tkn, line, col)
+            }
+        }
+    }
+
+    fn parse_struct(&mut self) -> Item {
+        let name = self.expect(TokenKind::Ident).value;
+        self.expect(TokenKind::LBrace);
+        let fields = self.sequence(TokenKind::Comma, TokenKind::RBrace, Self::parse_field);
+        Item::Struct(Struct { name, fields })
+    }
+
+    fn parse_field(&mut self) -> Field {
+        let name = self.expect(TokenKind::Ident).value;
+        self.expect(TokenKind::Colon);
+        let ty = self.type_();
+
+        Field { name, ty }
+    }
+
+    fn type_(&mut self) -> Type {
+        let token = self.next().unwrap();
+        match token.kind {
+            TokenKind::Ty(ty) => Type::Builtin(ty),
+            TokenKind::Ident => Type::Struct(token.value),
+            TokenKind::Op(Op::Band) => {
+                let ty = self.type_();
+                Type::Pinter(Box::new(ty))
+            }
+            tkn => {
+                let (line, col) = self.pos_to_line_col(token.span.start);
+                panic!("Unexpected {:?} at {}:{}", tkn, line, col)
+            }
+        }
+    }
+
+    fn parse_function(&mut self) -> Item {
+        let name = self.expect(TokenKind::Ident).value;
+        self.expect(TokenKind::LParen);
+        let args = self.sequence(TokenKind::Comma, TokenKind::RParen, Self::parse_arg);
+        self.expect(TokenKind::Colon);
+        let ret = self.type_();
+        Item::Function(Function {
+            name,
+            args,
+            ret,
+            body: self.parse_block(),
+        })
+    }
+
+    fn parse_arg(&mut self) -> Arg {
+        let name = self.expect(TokenKind::Ident).value;
+        self.expect(TokenKind::Colon);
+        let ty = self.type_();
+        self.try_advance(TokenKind::Comma);
+        Arg { name, ty }
+    }
+
+    fn parse_expr(&mut self) -> Exp {
+        self.parse_binary_expr(255)
+    }
+
+    fn parse_binary_expr(&mut self, min_prec: u8) -> Exp {
+        let mut lhs = self.parse_unit_expr();
+
+        while let Some(TokenKind::Op(op)) = self.peek().map(|t| t.kind) {
+            let prec = op.prec();
+            if prec > min_prec {
+                break;
+            }
+
+            self.next();
+            let rhs = self.parse_binary_expr(prec);
+
+            lhs = Exp::Binary {
+                op,
+                left: Box::new(lhs),
+                right: Box::new(rhs),
+            };
+        }
+
+        lhs
+    }
+
+    fn parse_unit_expr(&mut self) -> Exp {
+        let token = self.next().unwrap();
+        let mut expr = match token.kind {
+            TokenKind::True => Exp::Literal(Literal::Bool(true)),
+            TokenKind::False => Exp::Literal(Literal::Bool(false)),
+            TokenKind::Ident => Exp::Variable(token.value),
+            TokenKind::LBrace => {
+                Exp::Block(self.sequence(TokenKind::Semicolon, TokenKind::LBrace, Self::parse_expr))
+            }
+            TokenKind::LParen => {
+                let expr = self.parse_expr();
+                self.expect(TokenKind::RParen);
+                expr
+            }
+            TokenKind::Number => {
+                let value = token.value.parse().unwrap();
+                Exp::Literal(Literal::Int(value))
+            }
+            TokenKind::Let => {
+                let name = self.expect(TokenKind::Ident).value;
+                let ty = self.try_advance(TokenKind::Colon).then(|| self.type_());
+                self.expect(TokenKind::Op(Op::Assign));
+                let value = self.parse_expr();
+                Exp::Let {
+                    name,
+                    ty,
+                    value: Box::new(value),
+                }
+            }
+            TokenKind::If => {
+                let cond = self.parse_expr();
+                let then = Exp::Block(self.parse_block());
+                let else_ = self
+                    .try_advance(TokenKind::Else)
+                    .then(|| {
+                        if self.peek().is_some_and(|t| t.kind == TokenKind::If) {
+                            self.parse_expr()
+                        } else {
+                            Exp::Block(self.parse_block())
+                        }
+                    })
+                    .map(Box::new);
+                Exp::If {
+                    cond: Box::new(cond),
+                    then: Box::new(then),
+                    else_,
+                }
+            }
+            TokenKind::For => {
+                let params =
+                    self.sequence(TokenKind::Semicolon, TokenKind::LBrace, Self::parse_expr);
+                let mut exprs = Vec::new();
+                while !self.try_advance(TokenKind::RBrace) {
+                    exprs.push(self.parse_expr());
+                    self.try_advance(TokenKind::Semicolon);
+                }
+                let block = Exp::Block(exprs);
+                let len = params.len();
+                let mut exprs = params.into_iter();
+                let [init, consd, step] = array::from_fn(|_| exprs.next());
+                match len {
+                    0 => Exp::For {
+                        init:  None,
+                        cond:  None,
+                        step:  None,
+                        block: Box::new(block),
+                    },
+                    1 => Exp::For {
+                        init:  None,
+                        cond:  init.map(Box::new),
+                        step:  None,
+                        block: Box::new(block),
+                    },
+                    3 => Exp::For {
+                        init:  init.map(Box::new),
+                        cond:  consd.map(Box::new),
+                        step:  step.map(Box::new),
+                        block: Box::new(block),
+                    },
+                    _ => {
+                        let (line, col) = self.pos_to_line_col(token.span.start);
+                        panic!("Invalid loop syntax at {}:{}, loop accepts 1 (while), 0 (loop), or 3 (for) statements separated by semicolon", line, col)
+                    }
+                }
+            }
+            TokenKind::Return => {
+                let value = self
+                    .peek()
+                    .is_some_and(|t| {
+                        !matches!(
+                            t.kind,
+                            TokenKind::Semicolon
+                                | TokenKind::RBrace
+                                | TokenKind::RParen
+                                | TokenKind::Comma
+                        )
+                    })
+                    .then(|| Box::new(self.parse_expr()));
+                Exp::Return(value)
+            }
+            TokenKind::Op(op) => Exp::Unary {
+                op,
+                exp: Box::new(self.parse_expr()),
+            },
+            TokenKind::Dot => {
+                let token = self.expect_any();
+                match token.kind {
+                    TokenKind::LBrace => {
+                        let fields = self.sequence(
+                            TokenKind::Comma,
+                            TokenKind::RBrace,
+                            Self::parse_ctor_field,
+                        );
+                        Exp::Ctor { name: None, fields }
+                    }
+                    tkn => {
+                        let (line, col) = self.pos_to_line_col(token.span.start);
+                        panic!("Unexpected {:?} at {}:{}", tkn, line, col)
+                    }
+                }
+            }
+
+            TokenKind::Ty(_)
+            | TokenKind::String
+            | TokenKind::Use
+            | TokenKind::Break
+            | TokenKind::Continue
+            | TokenKind::Struct
+            | TokenKind::RBrace
+            | TokenKind::RParen
+            | TokenKind::LBracket
+            | TokenKind::RBracket
+            | TokenKind::Colon
+            | TokenKind::Semicolon
+            | TokenKind::Comma
+            | TokenKind::Fn
+            | TokenKind::Else => {
+                let (line, col) = self.pos_to_line_col(token.span.start);
+                panic!("Unexpected {:?} at {}:{}", token.kind, line, col)
+            }
+        };
+
+        loop {
+            match self.peek().map(|t| t.kind) {
+                Some(TokenKind::LParen) => {
+                    self.next();
+                    expr = Exp::Call {
+                        name: match expr {
+                            Exp::Variable(name) => name,
+                            _ => {
+                                let (line, col) = self.pos_to_line_col(token.span.start);
+                                panic!("Expected function name at {}:{}", line, col)
+                            }
+                        },
+                        args: self.sequence(TokenKind::Comma, TokenKind::RParen, Self::parse_expr),
+                    };
+                }
+                Some(TokenKind::LBracket) => {
+                    self.next();
+                    let index = self.parse_expr();
+                    self.expect(TokenKind::RBracket);
+                    expr = Exp::Index {
+                        base:  Box::new(expr),
+                        index: Box::new(index),
+                    };
+                }
+                Some(TokenKind::Dot) => {
+                    self.next();
+
+                    let token = self.expect_any();
+                    match token.kind {
+                        TokenKind::Ident => {
+                            expr = Exp::Field {
+                                base:  Box::new(expr),
+                                field: token.value,
+                            };
+                        }
+                        TokenKind::LBrace => {
+                            let fields = self.sequence(
+                                TokenKind::Comma,
+                                TokenKind::RBrace,
+                                Self::parse_ctor_field,
+                            );
+                            expr = Exp::Ctor {
+                                name: Some(Box::new(expr)),
+                                fields,
+                            };
+                        }
+                        tkn => {
+                            let (line, col) = self.pos_to_line_col(token.span.start);
+                            panic!("Unexpected {:?} at {}:{}", tkn, line, col)
+                        }
+                    }
+                }
+                _ => break expr,
+            }
+        }
+    }
+
+    pub fn parse_ctor_field(&mut self) -> CtorField {
+        let name = self.expect(TokenKind::Ident).value;
+        self.expect(TokenKind::Colon);
+        let value = self.parse_expr();
+        CtorField { name, value }
+    }
+
+    pub fn parse_block(&mut self) -> Vec<Exp> {
+        self.expect(TokenKind::LBrace);
+        let mut exprs = Vec::new();
+        while !self.try_advance(TokenKind::RBrace) {
+            exprs.push(self.parse_expr());
+            self.try_advance(TokenKind::Semicolon);
+        }
+        exprs
+    }
+
+    pub fn sequence<T>(
+        &mut self,
+        sep: TokenKind,
+        term: TokenKind,
+        mut parser: impl FnMut(&mut Self) -> T,
+    ) -> Vec<T> {
+        let mut items = Vec::new();
+        while !self.try_advance(term) {
+            items.push(parser(self));
+            if self.try_advance(term) {
+                break;
+            }
+            self.expect(sep);
+        }
+        items
+    }
+}
+
+pub fn parse(input: &str) -> Vec<Item> {
+    Parser::new(input).parse()
+}
+
+#[cfg(test)]
+mod test {
+    #[test]
+    fn sanity() {
+        let input = r#"
+            struct Foo {
+                x: i32,
+                y: i32,
+            }
+
+            fn main(): void {
+                let foo = Foo.{ x: 1, y: 2 };
+                if foo.x > 0 {
+                    return foo.x;
+                } else {
+                    return foo.y;
+                }
+                for i < 10 {
+                    i = i + 1;
+                }
+                for let i = 0; i < 10; i = i + 1 {
+                    i = i + 1;
+                }
+                i + 1 * 3 / 4 % 5 == 2 + 3 - 4 * 5 / 6 % 7;
+                fomething();
+                pahum(&foo);
+                lupa(*soo);
+                return foo.x + foo.y;
+            }
+
+            fn lupa(x: i32): i32 {
+                return x;
+            }
+
+            fn pahum(x: &Foo): void {
+                return;
+            }
+        "#;
+        let _ = super::parse(input);
+    }
+}

hblang/src/typechk.rs

@@ -0,0 +1,20 @@
+use crate::lexer::Ty;
+
+#[derive(Clone, Debug)]
+pub enum Type {
+    Builtin(Ty),
+    Struct(StructType),
+    Pointer(Box<Type>),
+}
+
+#[derive(Clone, Debug)]
+pub struct StructType {
+    pub name:   String,
+    pub fields: Vec<Field>,
+}
+
+#[derive(Clone, Debug)]
+pub struct Field {
+    pub name: String,
+    pub ty:   Type,
+}

hbvm/Cargo.toml

@@ -3,8 +3,10 @@ name = "hbvm"
 version = "0.1.0"
 edition = "2021"
 
-# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
+[features]
+default = ["alloc"]
+alloc   = []
+nightly = []
 
 [dependencies]
-log = "*"
-hashbrown = "0.13.2"
+hbbytecode = { path = "../hbbytecode" }

hbvm/fuzz/.gitignore

@@ -0,0 +1,5 @@
+target
+artifacts
+corpus
+coverage
+Cargo.lock

hbvm/fuzz/Cargo.toml

@@ -0,0 +1,30 @@
+[package]
+name = "hbvm-fuzz"
+version = "0.0.0"
+publish = false
+edition = "2021"
+
+[package.metadata]
+cargo-fuzz = true
+
+[dependencies]
+libfuzzer-sys = "0.4"
+
+[dependencies.hbvm]
+path = ".."
+
+[dependencies.hbbytecode]
+path = "../../hbbytecode"
+
+# Prevent this from interfering with workspaces
+[workspace]
+members = ["."]
+
+[profile.release]
+debug = 1
+
+[[bin]]
+name = "vm"
+path = "fuzz_targets/vm.rs"
+test = false
+doc = false

hbvm/fuzz/fuzz_targets/vm.rs

@@ -0,0 +1,82 @@
+#![no_main]
+
+use {
+    hbvm::{
+        mem::{
+            softpaging::{
+                paging::{PageTable, Permission},
+                HandlePageFault, PageSize, SoftPagedMem,
+            },
+            Address, MemoryAccessReason,
+        },
+        Vm,
+    },
+    libfuzzer_sys::fuzz_target,
+};
+
+fuzz_target!(|data: &[u8]| {
+    let mut vm = unsafe {
+        Vm::<_, 16384>::new(
+            SoftPagedMem::<_, true> {
+                pf_handler: TestTrapHandler,
+                program:    data,
+                root_pt:    Box::into_raw(Default::default()),
+                icache:     Default::default(),
+            },
+            Address::new(4),
+        )
+    };
+
+    // Alloc and map some memory
+    let pages = [
+        alloc_and_map(&mut vm.memory, 0),
+        alloc_and_map(&mut vm.memory, 4096),
+    ];
+
+    // Run VM
+    let _ = vm.run();
+
+    // Unmap and dealloc the memory
+    for (i, page) in pages.into_iter().enumerate() {
+        unmap_and_dealloc(&mut vm.memory, page, i as u64 * 4096);
+    }
+
+    let _ = unsafe { Box::from_raw(vm.memory.root_pt) };
+});
+
+fn alloc_and_map(memory: &mut SoftPagedMem<TestTrapHandler>, at: u64) -> *mut u8 {
+    let ptr = Box::into_raw(Box::<Page>::default()).cast();
+    unsafe {
+        memory
+            .map(ptr, Address::new(at), Permission::Write, PageSize::Size4K)
+            .unwrap()
+    };
+    ptr
+}
+
+fn unmap_and_dealloc(memory: &mut SoftPagedMem<TestTrapHandler>, ptr: *mut u8, from: u64) {
+    memory.unmap(Address::new(from)).unwrap();
+    let _ = unsafe { Box::from_raw(ptr.cast::<Page>()) };
+}
+
+#[repr(align(4096))]
+struct Page([u8; 4096]);
+impl Default for Page {
+    fn default() -> Self {
+        unsafe { std::mem::MaybeUninit::zeroed().assume_init() }
+    }
+}
+
+struct TestTrapHandler;
+impl HandlePageFault for TestTrapHandler {
+    fn page_fault(
+        &mut self,
+        _: MemoryAccessReason,
+        _: &mut PageTable,
+        _: Address,
+        _: PageSize,
+        _: *mut u8,
+    ) -> bool {
+        false
+    }
+}

hbvm/src/bmc.rs

@@ -0,0 +1,141 @@
+//! Block memory copier state machine
+
+use {
+    super::{mem::MemoryAccessReason, Memory, VmRunError},
+    crate::mem::Address,
+    core::{mem::MaybeUninit, task::Poll},
+};
+
+/// Buffer size (defaults to 4 KiB, a smallest page size on most platforms)
+const BUF_SIZE: usize = 4096;
+
+/// Buffer of possibly uninitialised bytes, aligned to [`BUF_SIZE`]
+#[repr(align(4096))]
+struct AlignedBuf([MaybeUninit<u8>; BUF_SIZE]);
+
+/// State for block memory copy
+pub struct BlockCopier {
+    /// Source address
+    src:       Address,
+    /// Destination address
+    dst:       Address,
+    /// How many buffer sizes to copy?
+    n_buffers: usize,
+    /// …and what remainds after?
+    rem:       usize,
+}
+
+impl BlockCopier {
+    /// Construct a new one
+    #[inline]
+    pub fn new(src: Address, dst: Address, count: usize) -> Self {
+        Self {
+            src,
+            dst,
+            n_buffers: count / BUF_SIZE,
+            rem: count % BUF_SIZE,
+        }
+    }
+
+    /// Copy one block
+    ///
+    /// # Safety
+    /// - Same as for [`Memory::load`] and [`Memory::store`]
+    pub unsafe fn poll(&mut self, memory: &mut impl Memory) -> Poll<Result<(), BlkCopyError>> {
+        // Safety: Assuming uninit of array of MaybeUninit is sound
+        let mut buf = AlignedBuf(unsafe { MaybeUninit::uninit().assume_init() });
+
+        // We have at least one buffer size to copy
+        if self.n_buffers != 0 {
+            if let Err(e) = unsafe {
+                act(
+                    memory,
+                    self.src,
+                    self.dst,
+                    buf.0.as_mut_ptr().cast(),
+                    BUF_SIZE,
+                )
+            } {
+                return Poll::Ready(Err(e));
+            }
+
+            // Bump source and destination address
+            self.src += BUF_SIZE;
+            self.dst += BUF_SIZE;
+
+            self.n_buffers -= 1;
+
+            return if self.n_buffers + self.rem == 0 {
+                // If there is nothing left, we are done
+                Poll::Ready(Ok(()))
+            } else {
+                // Otherwise let's advice to run it again
+                Poll::Pending
+            };
+        }
+
+        if self.rem != 0 {
+            if let Err(e) = unsafe {
+                act(
+                    memory,
+                    self.src,
+                    self.dst,
+                    buf.0.as_mut_ptr().cast(),
+                    self.rem,
+                )
+            } {
+                return Poll::Ready(Err(e));
+            }
+        }
+
+        Poll::Ready(Ok(()))
+    }
+}
+
+/// Load to buffer and store from buffer
+#[inline]
+unsafe fn act(
+    memory: &mut impl Memory,
+    src: Address,
+    dst: Address,
+    buf: *mut u8,
+    count: usize,
+) -> Result<(), BlkCopyError> {
+    unsafe {
+        // Load to buffer
+        memory
+            .load(src, buf, count)
+            .map_err(|super::mem::LoadError(addr)| BlkCopyError {
+                access_reason: MemoryAccessReason::Load,
+                addr,
+            })?;
+
+        // Store from buffer
+        memory
+            .store(dst, buf, count)
+            .map_err(|super::mem::StoreError(addr)| BlkCopyError {
+                access_reason: MemoryAccessReason::Store,
+                addr,
+            })?;
+    }
+
+    Ok(())
+}
+
+/// Error occured when copying a block of memory
+#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+pub struct BlkCopyError {
+    /// Kind of access
+    access_reason: MemoryAccessReason,
+    /// VM Address
+    addr: Address,
+}
+
+impl From<BlkCopyError> for VmRunError {
+    fn from(value: BlkCopyError) -> Self {
+        match value.access_reason {
+            MemoryAccessReason::Load => Self::LoadAccessEx(value.addr),
+            MemoryAccessReason::Store => Self::StoreAccessEx(value.addr),
+        }
+    }
+}

hbvm/src/bytecode/mod.rs

@@ -1,2 +0,0 @@
-pub mod ops;
-pub mod types;

hbvm/src/bytecode/ops.rs

@@ -1,68 +0,0 @@
-#[repr(u8)]
-pub enum Operations {
-    NOP = 0,
-
-    ADD = 1,
-    SUB = 2,
-    MUL = 3,
-    DIV = 4,
-    MOD = 5,
-
-    AND = 6,
-    OR = 7,
-    XOR = 8,
-    NOT = 9,
-
-    // LOADs a memory address/constant into a register
-    LOAD = 15,
-    // STOREs a register/constant into a memory address
-    STORE = 16,
-
-    MapPage = 17,
-    UnmapPage = 18,
-
-    // SHIFT LEFT 16 A0
-    Shift = 20,
-
-    JUMP = 100,
-    JumpCond = 101,
-    RET = 103,
-
-    EnviromentCall = 255,
-}
-
-pub enum PageMapTypes {
-    // Have the host make a new VMPage
-    VMPage = 0,
-    // Ask the host to map a RealPage into memory
-    RealPage = 1,
-}
-
-pub enum MathOpSubTypes {
-    Unsigned = 0,
-    Signed = 1,
-    FloatingPoint = 2,
-}
-
-pub enum MathOpSides {
-    RegisterConstant = 0,
-    RegisterRegister = 1,
-    ConstantConstant = 2,
-    ConstantRegister = 3,
-}
-
-pub enum RWSubTypes {
-    AddrToReg = 0,
-    RegToAddr,
-    ConstToReg,
-    ConstToAddr,
-}
-
-pub enum JumpConditionals {
-    Equal = 0,
-    NotEqual = 1,
-    LessThan = 2,
-    LessThanOrEqualTo = 3,
-    GreaterThan = 4,
-    GreaterThanOrEqualTo = 5,
-}

hbvm/src/bytecode/types.rs

@@ -1,9 +0,0 @@
-
-pub const CONST_U8: u8 = 0x00;
-pub const CONST_I8: i8 = 0x01;
-
-pub const CONST_U64: u8 = 0x02;
-pub const CONST_I64: u8 = 0x03;
-pub const CONST_F64: u8 = 0x04;
-
-pub const ADDRESS: u8 = 0x05;

hbvm/src/engine/call_stack.rs

@@ -1,6 +0,0 @@
-use alloc::vec::Vec;
-
-pub type CallStack = Vec<FnCall>;
-pub struct FnCall {
-    pub ret: usize,
-}

hbvm/src/engine/config.rs

@@ -1,13 +0,0 @@
-pub struct EngineConfig {
-    pub call_stack_depth: usize,
-    pub quantum: u32,
-}
-
-impl EngineConfig {
-    pub fn default() -> Self {
-        Self {
-            call_stack_depth: 32,
-            quantum: 0,
-        }
-    }
-}

hbvm/src/engine/enviroment_calls.rs

@@ -1,3 +0,0 @@
-use super::Engine;
-
-pub type EnviromentCall = fn(&mut Engine) -> Result<&mut Engine, u64>;

hbvm/src/engine/mod.rs

@@ -1,100 +0,0 @@
-pub mod call_stack;
-pub mod config;
-pub mod enviroment_calls;
-pub mod regs;
-#[cfg(test)]
-pub mod tests;
-
-use {
-    self::call_stack::CallStack,
-    crate::{memory, HaltStatus, RuntimeErrors},
-    alloc::vec::Vec,
-    config::EngineConfig,
-    log::trace,
-    regs::Registers,
-};
-// pub const PAGE_SIZE: usize = 8192;
-
-pub struct RealPage {
-    pub ptr: *mut u8,
-}
-
-#[derive(Debug, Clone, Copy)]
-pub struct VMPage {
-    pub data: [u8; 8192],
-}
-impl VMPage {
-    pub fn new() -> Self {
-        Self {
-            data: [0; 4096 * 2],
-        }
-    }
-}
-
-pub enum Page {
-    VMPage(VMPage),
-    RealPage(RealPage),
-}
-impl Page {
-    pub fn data(&self) -> [u8; 4096 * 2] {
-        match self {
-            Page::VMPage(vmpage) => vmpage.data,
-            Page::RealPage(_) => {
-                unimplemented!("Memmapped hw page not yet supported")
-            }
-        }
-    }
-}
-
-pub fn empty_enviroment_call(engine: &mut Engine) -> Result<&mut Engine, u64> {
-    trace!("Registers {:?}", engine.registers);
-    Err(0)
-}
-
-pub struct Engine {
-    pub index:     usize,
-    pub program:   Vec<u8>,
-    pub registers: Registers,
-    pub config:    EngineConfig,
-
-    /// BUG: This DOES NOT account for overflowing
-    pub last_timer_count: u32,
-    pub timer_callback: Option<fn() -> u32>,
-    pub memory: memory::Memory,
-    pub enviroment_call_table: [Option<EnviromentCall>; 256],
-    pub call_stack: CallStack,
-}
-use crate::engine::enviroment_calls::EnviromentCall;
-impl Engine {
-    pub fn set_timer_callback(&mut self, func: fn() -> u32) {
-        self.timer_callback = Some(func);
-    }
-    pub fn set_register(&mut self, register: u8, value: u64) {}
-}
-
-impl Engine {
-    pub fn new(program: Vec<u8>) -> Self {
-        let mut mem = memory::Memory::new();
-        for (addr, byte) in program.clone().into_iter().enumerate() {
-            let _ = mem.set_addr8(addr as u64, byte);
-        }
-        trace!("{:?}", mem.read_addr8(0));
-        let ecall_table: [Option<EnviromentCall>; 256] = [None; 256];
-        Self {
-            index: 0,
-            program,
-            registers: Registers::new(),
-            config: EngineConfig::default(),
-            last_timer_count: 0,
-            timer_callback: None,
-            enviroment_call_table: ecall_table,
-            memory: mem,
-            call_stack: Vec::new(),
-        }
-    }
-
-    pub fn dump(&self) {}
-    pub fn run(&mut self) -> Result<HaltStatus, RuntimeErrors> {
-        Ok(HaltStatus::Halted)
-    }
-}

hbvm/src/engine/regs.rs

@@ -1,32 +0,0 @@
-#[rustfmt::skip]
-#[derive(Debug, Clone, Copy)]
-pub struct Registers {
-    pub a0: u64, pub b0: u64, pub c0: u64, pub d0: u64, pub e0: u64, pub f0: u64,
-    pub a1: u64, pub b1: u64, pub c1: u64, pub d1: u64, pub e1: u64, pub f1: u64,
-    pub a2: u64, pub b2: u64, pub c2: u64, pub d2: u64, pub e2: u64, pub f2: u64,
-    pub a3: u64, pub b3: u64, pub c3: u64, pub d3: u64, pub e3: u64, pub f3: u64,
-    pub a4: u64, pub b4: u64, pub c4: u64, pub d4: u64, pub e4: u64, pub f4: u64,
-    pub a5: u64, pub b5: u64, pub c5: u64, pub d5: u64, pub e5: u64, pub f5: u64,
-    pub a6: u64, pub b6: u64, pub c6: u64, pub d6: u64, pub e6: u64, pub f6: u64,
-    pub a7: u64, pub b7: u64, pub c7: u64, pub d7: u64, pub e7: u64, pub f7: u64,
-    pub a8: u64, pub b8: u64, pub c8: u64, pub d8: u64, pub e8: u64, pub f8: u64,
-    pub a9: u64, pub b9: u64, pub c9: u64, pub d9: u64, pub e9: u64, pub f9: u64,
-}
-
-impl Registers {
-    #[rustfmt::skip]
-    pub fn new() -> Self{
-        Self {
-            a0: 0, b0: 0, c0: 0, d0: 0, e0: 0, f0: 0,
-            a1: 0, b1: 0, c1: 0, d1: 0, e1: 0, f1: 0,
-            a2: 0, b2: 0, c2: 0, d2: 0, e2: 0, f2: 0,
-            a3: 0, b3: 0, c3: 0, d3: 0, e3: 0, f3: 0,
-            a4: 0, b4: 0, c4: 0, d4: 0, e4: 0, f4: 0,
-            a5: 0, b5: 0, c5: 0, d5: 0, e5: 0, f5: 0,
-            a6: 0, b6: 0, c6: 0, d6: 0, e6: 0, f6: 0,
-            a7: 0, b7: 0, c7: 0, d7: 0, e7: 0, f7: 0,
-            a8: 0, b8: 0, c8: 0, d8: 0, e8: 0, f8: 0,
-            a9: 0, b9: 0, c9: 0, d9: 0, e9: 0, f9: 0,
-        }
-    }
-}

hbvm/src/engine/tests/mod.rs

@@ -1,125 +0,0 @@
-use {
-    super::Engine,
-    crate::{HaltStatus, RuntimeErrors},
-    alloc::vec,
-    RuntimeErrors::*,
-};
-
-#[test]
-fn invalid_program() {
-    let prog = vec![1, 0];
-    let mut eng = Engine::new(prog);
-    let ret = eng.run();
-    assert_eq!(ret, Err(InvalidOpcodePair(1, 0)));
-}
-
-#[test]
-fn empty_program() {
-    let prog = vec![];
-    let mut eng = Engine::new(prog);
-    let ret = eng.run();
-    assert_eq!(ret, Ok(HaltStatus::Halted));
-}
-
-#[test]
-fn max_quantum_reached() {
-    let prog = vec![0, 0, 0, 0];
-    let mut eng = Engine::new(prog);
-    eng.set_timer_callback(|| {
-        return 1;
-    });
-    eng.config.quantum = 1;
-    let ret = eng.run();
-    assert_eq!(ret, Ok(HaltStatus::Running));
-}
-
-#[test]
-fn jump_out_of_bounds() {
-    use crate::bytecode::ops::Operations::JUMP;
-    let prog = vec![JUMP as u8, 0, 0, 0, 0, 0, 0, 1, 0];
-    let mut eng = Engine::new(prog);
-    let ret = eng.run();
-    assert_eq!(ret, Err(InvalidJumpAddress(256)));
-}
-
-#[test]
-fn invalid_system_call() {
-    let prog = vec![255, 0];
-    let mut eng = Engine::new(prog);
-    let ret = eng.run();
-    assert_eq!(ret, Err(InvalidSystemCall(0)));
-}
-
-#[test]
-fn add_u8() {
-    use crate::bytecode::ops::{MathOpSides::ConstantConstant, Operations::ADD};
-
-    let prog = vec![ADD as u8, ConstantConstant as u8, 100, 98, 0xA0];
-    let mut eng = Engine::new(prog);
-    let _ = eng.run();
-    assert_eq!(eng.registers.a0, 2);
-}
-
-#[test]
-fn sub_u8() {
-    use crate::bytecode::ops::Operations::SUB;
-
-    let prog = vec![SUB as u8];
-    let mut eng = Engine::new(prog);
-    let _ = eng.run();
-    assert_eq!(eng.registers.a0, 1);
-}
-#[test]
-fn mul_u8() {
-    use crate::bytecode::ops::{MathOpSides::ConstantConstant, Operations::MUL};
-
-    let prog = vec![MUL as u8, ConstantConstant as u8, 1, 2, 0xA0];
-    let mut eng = Engine::new(prog);
-    let _ = eng.run();
-    assert_eq!(eng.registers.a0, 2);
-}
-
-#[test]
-fn div_u8() {
-    use crate::bytecode::ops::Operations::DIV;
-
-    let prog = vec![DIV as u8];
-    let mut eng = Engine::new(prog);
-    let _ = eng.run();
-    assert_eq!(eng.registers.a0, 2);
-}
-
-#[test]
-fn set_register() {
-    let prog = alloc::vec![];
-    let mut eng = Engine::new(prog);
-    eng.set_register(0xA0, 1);
-    assert_eq!(eng.registers.a0, 1);
-}
-
-#[test]
-fn load_u8() {
-    use crate::bytecode::ops::{Operations::LOAD, RWSubTypes::AddrToReg};
-
-    let prog = vec![LOAD as u8, AddrToReg as u8, 0, 0, 0, 0, 0, 0, 1, 0, 0xA0];
-    let mut eng = Engine::new(prog);
-    let ret = eng.memory.set_addr8(256, 1);
-    assert_eq!(ret, Ok(()));
-    let _ = eng.run();
-    assert_eq!(eng.registers.a0, 1);
-}
-#[test]
-fn set_memory_8() {
-    let prog = vec![];
-    let mut eng = Engine::new(prog);
-    let ret = eng.memory.set_addr8(256, 1);
-    assert_eq!(ret, Ok(()));
-}
-
-#[test]
-fn set_memory_64() {
-    let prog = vec![];
-    let mut eng = Engine::new(prog);
-    let ret = eng.memory.set_addr64(256, 1);
-    assert_eq!(ret, Ok(()));
-}

hbvm/src/float/aarch64.rs

@@ -0,0 +1,49 @@
+use {core::arch::asm, hbbytecode::RoundingMode};
+
+macro_rules! fnsdef {
+    {$(
+        $(#[$attr:meta])*
+        $vis:vis fn $name:ident[$inreg:ident -> $outreg:ident]($from:ident -> $to:ident): $ins:literal;
+    )*} => {$(
+        $(#[$attr])*
+        $vis fn $name(val: $from, mode: RoundingMode) -> $to {
+            let result: $to;
+            unsafe {
+                if mode == RoundingMode::NearestEven {
+                    return;
+                }
+
+                let fpcr: u64;
+                unsafe { asm!("mrs {}, fpcr", out(reg) fpcr) };
+
+                let fpcr_new = fpcr & !(0b11 << 22)
+                    | (match mode {
+                        RoundingMode::NearestEven => 0b00,
+                        RoundingMode::Truncate => 0b11,
+                        RoundingMode::Up => 0b01,
+                        RoundingMode::Down => 0b10,
+                    }) << 22;
+
+                unsafe { asm!("msr fpcr, {}", in(reg) fpcr_new) };
+                asm!(
+                    $ins,
+                    out($outreg) result,
+                    in($inreg)   val,
+                );
+                unsafe { asm!("msr fpcr, {}", in(reg) fpcr) };
+            }
+            result
+        }
+    )*};
+}
+
+fnsdef! {
+    /// Convert [`f64`] to [`f32`] with chosen rounding mode
+    pub fn conv64to32[vreg -> vreg](f64 -> f32): "fcvt {:s}, {:d}";
+
+    /// Convert [`f32`] to [`i64`] with chosen rounding mode
+    pub fn f32toint[vreg -> reg](f32 -> i64): "fcvtzs {}, {:s}";
+
+    /// Convert [`f64`] to [`i64`] with chosen rounding mode
+    pub fn f64toint[vreg -> reg](f64 -> i64): "fcvtzs {}, {:d}";
+}

hbvm/src/float/mod.rs

@@ -0,0 +1,31 @@
+macro_rules! arch_specific {
+    {
+        $({$($cfg:tt)*} : $mod:ident;)*
+    } => {
+        $(
+            #[cfg($($cfg)*)]
+            mod $mod;
+
+            #[cfg($($cfg)*)]
+            pub use $mod::*;
+
+            #[cfg($($cfg)*)]
+            pub const FL_ARCH_SPECIFIC_SUPPORTED: bool = true;
+        )*
+        
+        #[cfg(not(any($($($cfg)*),*)))]
+        mod unsupported;
+
+        #[cfg(not(any($($($cfg)*),*)))]
+        pub use unsupported::*;
+
+        #[cfg(not(any($($($cfg)*),*)))]
+        pub const FL_ARCH_SPECIFIC_SUPPORTED: bool = false;
+    };
+}
+
+arch_specific! {
+    {target_arch = "x86_64" }: x86_64;
+    {target_arch = "riscv64"}: riscv64;
+    {target_arch = "aarch64"}: aarch64;
+}

hbvm/src/float/riscv64.rs

@@ -0,0 +1,59 @@
+use {core::arch::asm, hbbytecode::RoundingMode};
+
+macro_rules! roundm_op_litmode_internal {
+    ($ins:literal, $in:expr, $out:expr => $outy:ident, $mode:literal $(,)?) => {
+        asm!(
+            concat!($ins, " {}, {}, ", $mode),
+            out($outy) $out,
+            in(freg)   $in,
+        )
+    };
+}
+
+macro_rules! gen_roundm_op_litmode {
+    [$($ty:ident => $reg:ident),* $(,)?] => {
+        macro_rules! roundm_op_litmode {
+            $(
+                ($ins:literal, $in:expr, $out:expr => $ty, $mode:literal) => {
+                    roundm_op_litmode_internal!($ins, $in, $out => $reg, $mode)
+                };
+            )*
+        }
+    };
+}
+
+gen_roundm_op_litmode![
+    f32 => freg,
+    f64 => freg,
+    i64 => reg,
+];
+
+macro_rules! fnsdef {
+    {$(
+        $(#[$attr:meta])*
+        $vis:vis fn $name:ident($from:ident -> $to:ident): $ins:literal;
+    )*} => {$(
+        $(#[$attr])*
+        $vis fn $name(val: $from, mode: RoundingMode) -> $to {
+            let result: $to;
+            unsafe {
+                match mode {
+                    RoundingMode::NearestEven => roundm_op_litmode!($ins, val, result => $to, "rne"),
+                    RoundingMode::Truncate    => roundm_op_litmode!($ins, val, result => $to, "rtz"),
+                    RoundingMode::Up          => roundm_op_litmode!($ins, val, result => $to, "rup"),
+                    RoundingMode::Down        => roundm_op_litmode!($ins, val, result => $to, "rdn"),
+                }
+            }
+            result
+        }
+    )*};
+}
+
+fnsdef! {
+    /// Convert [`f64`] to [`f32`] with chosen rounding mode
+    pub fn conv64to32(f64 -> f32): "fcvt.s.d";
+    /// Convert [`f32`] to [`i64`] with chosen rounding mode
+    pub fn f32toint(f32 -> i64): "fcvt.l.s";
+    /// Convert [`f64`] to [`i64`] with chosen rounding mode
+    pub fn f64toint(f64 -> i64): "fcvt.l.d";
+}

hbvm/src/float/unsupported.rs

@@ -0,0 +1,16 @@
+use hbbytecode::RoundingMode;
+
+#[inline(always)]
+pub fn conv64to32(_: f64, _: RoundingMode) -> f32 {
+    f32::NAN
+}
+
+#[inline(always)]
+pub fn f32toint(_: f32, _: RoundingMode) -> i64 {
+    i64::MAX
+}
+
+#[inline(always)]
+pub fn f64toint(_: f64, _: RoundingMode) -> i64 {
+    i64::MAX
+}

hbvm/src/float/x86_64.rs

@@ -0,0 +1,64 @@
+use {
+    core::arch::{asm, x86_64 as arin},
+    hbbytecode::RoundingMode,
+};
+
+macro_rules! gen_op {
+    [$($ty:ident => $reg:ident),* $(,)?] => {
+        macro_rules! op {
+            $(
+                ($ins:literal, $in:expr, $out:expr => $ty) => {
+                    asm!(concat!($ins, " {}, {}"), out($reg) $out, in(xmm_reg) $in)
+                };
+            )*
+        }
+    };
+}
+
+gen_op![
+    f32 => xmm_reg,
+    f64 => xmm_reg,
+    i64 => reg,
+];
+
+macro_rules! fnsdef {
+    {$(
+        $(#[$attr:meta])*
+        $vis:vis fn $name:ident($from:ident -> $to:ident): $ins:literal;
+    )*} => {$(
+        $(#[$attr])*
+        $vis fn $name(val: $from, mode: RoundingMode) -> $to {
+            let result: $to;
+            unsafe {
+                let mut mxcsr = 0_u32;
+                'a: {
+                    asm!("stmxcsr [{}]", in(reg) &mut mxcsr);
+                    asm!(
+                        "ldmxcsr [{}]",
+                        in(reg) &(mxcsr & !arin::_MM_ROUND_MASK | match mode {
+                            RoundingMode::NearestEven => break 'a,
+                            RoundingMode::Truncate => arin::_MM_ROUND_TOWARD_ZERO,
+                            RoundingMode::Up => arin::_MM_ROUND_UP,
+                            RoundingMode::Down => arin::_MM_ROUND_DOWN,
+                        })
+                    );
+                }
+
+                op!($ins, val, result => $to);
+
+                // Set MXCSR to original value
+                asm!("ldmxcsr [{}]", in(reg) &mxcsr);
+            }
+            result
+        }
+    )*};
+}
+
+fnsdef! {
+    /// Convert [`f64`] to [`f32`] with chosen rounding mode
+    pub fn conv64to32(f64 -> f32): "cvtsd2ss";
+    /// Convert [`f32`] to [`i64`] with chosen rounding mode
+    pub fn f32toint(f32 -> i64): "cvttss2si";
+    /// Convert [`f64`] to [`i64`] with chosen rounding mode
+    pub fn f64toint(f64 -> i64): "cvttsd2si";
+}

hbvm/src/lib.rs

@@ -1,23 +1,143 @@
+//! HoleyBytes Virtual Machine
+//!
+//! # Alloc feature
+//! - Enabled by default
+//! - Provides mapping / unmapping, as well as [`Default`] and [`Drop`]
+//!   implementations for soft-paged memory implementation
+
+// # General safety notice:
+// - Validation has to assure there is 256 registers (r0 - r255)
+// - Instructions have to be valid as specified (values and sizes)
+// - Mapped pages should be at least 4 KiB
+
 #![no_std]
+#![cfg_attr(feature = "nightly", feature(fn_align))]
+#![deny(unsafe_op_in_unsafe_fn)]
+
+#[cfg(feature = "alloc")]
 extern crate alloc;
 
-pub mod bytecode;
-pub mod engine;
-pub mod memory;
+pub mod mem;
+pub mod value;
 
-#[derive(Debug, PartialEq)]
-pub enum RuntimeErrors {
-    InvalidOpcodePair(u8, u8),
-    RegisterTooSmall,
-    HostError(u64),
-    PageNotMapped(u64),
-    InvalidJumpAddress(u64),
-    InvalidSystemCall(u8),
+mod bmc;
+mod float;
+mod utils;
+mod vmrun;
+
+pub use float::FL_ARCH_SPECIFIC_SUPPORTED;
+
+use {
+    bmc::BlockCopier,
+    mem::{Address, Memory},
+    value::{Value, ValueVariant},
+};
+
+/// HoleyBytes Virtual Machine
+pub struct Vm<Mem, const TIMER_QUOTIENT: usize> {
+    /// Holds 256 registers
+    ///
+    /// Writing to register 0 is considered undefined behaviour
+    /// in terms of HoleyBytes program execution
+    pub registers: [Value; 256],
+
+    /// Memory implementation
+    pub memory: Mem,
+
+    /// Program counter
+    pub pc: Address,
+
+    /// Program timer
+    timer: usize,
+
+    /// Saved block copier
+    copier: Option<BlockCopier>,
 }
 
-// If you solve the halting problem feel free to remove this
-#[derive(PartialEq, Debug)]
-pub enum HaltStatus {
-    Halted,
-    Running,
+impl<Mem, const TIMER_QUOTIENT: usize> Vm<Mem, TIMER_QUOTIENT>
+where
+    Mem: Memory,
+{
+    /// Create a new VM with program and trap handler
+    ///
+    /// # Safety
+    /// Program code has to be validated
+    pub unsafe fn new(memory: Mem, entry: Address) -> Self {
+        Self {
+            registers: [Value::from(0_u64); 256],
+            memory,
+            pc: entry,
+            timer: 0,
+            copier: None,
+        }
+    }
+
+    /// Read register
+    #[inline(always)]
+    pub fn read_reg(&self, n: u8) -> Value {
+        unsafe { *self.registers.get_unchecked(n as usize) }
+    }
+
+    /// Write a register.
+    /// Writing to register 0 is no-op.
+    #[inline(always)]
+    pub fn write_reg<T: ValueVariant>(&mut self, n: u8, value: T) {
+        if n != 0 {
+            unsafe {
+                core::ptr::copy_nonoverlapping(
+                    (&value as *const T).cast::<u8>(),
+                    self.registers.as_mut_ptr().add(n.into()).cast::<u8>(),
+                    core::mem::size_of::<T>(),
+                );
+            };
+        }
+    }
+}
+
+/// Virtual machine halt error
+#[derive(Copy, Clone, Debug, PartialEq, Eq)]
+#[repr(u8)]
+pub enum VmRunError {
+    /// Tried to execute invalid instruction
+    InvalidOpcode(u8),
+
+    /// Unhandled load access exception
+    LoadAccessEx(Address),
+
+    /// Unhandled instruction load access exception
+    ProgramFetchLoadEx(Address),
+
+    /// Unhandled store access exception
+    StoreAccessEx(Address),
+
+    /// Register out-of-bounds access
+    RegOutOfBounds,
+
+    /// Address out-of-bounds
+    AddrOutOfBounds,
+
+    /// Reached unreachable code
+    Unreachable,
+
+    /// Invalid operand
+    InvalidOperand,
+
+    /// Unimplemented feature
+    Unimplemented,
+}
+
+/// Virtual machine halt ok
+#[derive(Copy, Clone, Debug, PartialEq, Eq)]
+pub enum VmRunOk {
+    /// Program has eached its end
+    End,
+
+    /// Program was interrupted by a timer
+    Timer,
+
+    /// Environment call
+    Ecall,
+
+    /// Breakpoint
+    Breakpoint,
 }

hbvm/src/main.rs

@@ -1,31 +0,0 @@
-use hbvm::{
-    bytecode::ops::{Operations::*, RWSubTypes::*},
-    engine::Engine,
-    RuntimeErrors,
-};
-
-fn main() -> Result<(), RuntimeErrors> {
-    // TODO: Grab program from cmdline
-    #[rustfmt::skip]
-        let prog: Vec<u8> = vec![
-            NOP as u8,            
-            JUMP as u8, 0, 0, 0, 0, 0, 0, 0, 0,
-        ];
-
-    let mut eng = Engine::new(prog);
-    // eng.set_timer_callback(time);
-    eng.enviroment_call_table[10] = Some(print_fn);
-    eng.run()?;
-    eng.dump();
-    println!("{:#?}", eng.registers);
-
-    Ok(())
-}
-
-pub fn time() -> u32 {
-    9
-}
-pub fn print_fn(engine: &mut Engine) -> Result<&mut Engine, u64> {
-    println!("hello");
-    Ok(engine)
-}

hbvm/src/mem/addr.rs

@@ -0,0 +1,131 @@
+//! Virtual(?) memory address
+
+use {
+    crate::utils::impl_display,
+    core::{fmt::Debug, ops},
+};
+
+/// Memory address
+#[derive(Clone, Copy, PartialEq, Eq, PartialOrd, Ord)]
+pub struct Address(u64);
+impl Address {
+    /// A null address
+    pub const NULL: Self = Self(0);
+
+    /// Saturating integer addition. Computes self + rhs, saturating at the numeric bounds instead of overflowing.
+    #[inline]
+    pub fn saturating_add<T: AddressOp>(self, rhs: T) -> Self {
+        Self(self.0.saturating_add(rhs.cast_u64()))
+    }
+
+    /// Saturating integer subtraction. Computes self - rhs, saturating at the numeric bounds instead of overflowing.
+    #[inline]
+    pub fn saturating_sub<T: AddressOp>(self, rhs: T) -> Self {
+        Self(self.0.saturating_sub(rhs.cast_u64()))
+    }
+
+    /// Wrapping integer addition. Computes self + rhs, wrapping the numeric bounds.
+    #[inline]
+    pub fn wrapping_add<T: AddressOp>(self, rhs: T) -> Self {
+        Self(self.0.wrapping_add(rhs.cast_u64()))
+    }
+
+    /// Wrapping integer subtraction. Computes self + rhs, wrapping the numeric bounds.
+    #[inline]
+    pub fn wrapping_sub<T: AddressOp>(self, rhs: T) -> Self {
+        Self(self.0.wrapping_sub(rhs.cast_u64()))
+    }
+
+    /// Cast or if smaller, truncate to [`usize`]
+    pub fn truncate_usize(self) -> usize {
+        self.0 as _
+    }
+
+    /// Get inner value
+    #[inline(always)]
+    pub fn get(self) -> u64 {
+        self.0
+    }
+
+    /// Construct new address
+    #[inline(always)]
+    pub fn new(val: u64) -> Self {
+        Self(val)
+    }
+
+    /// Do something with inner value
+    #[inline(always)]
+    pub fn map(self, f: impl Fn(u64) -> u64) -> Self {
+        Self(f(self.0))
+    }
+}
+
+impl_display!(for Address =>
+    |Address(a)| "{a:0x}"
+);
+
+impl<T: AddressOp> ops::Add<T> for Address {
+    type Output = Self;
+
+    #[inline]
+    fn add(self, rhs: T) -> Self::Output {
+        Self(self.0.wrapping_add(rhs.cast_u64()))
+    }
+}
+
+impl<T: AddressOp> ops::Sub<T> for Address {
+    type Output = Self;
+
+    #[inline]
+    fn sub(self, rhs: T) -> Self::Output {
+        Self(self.0.wrapping_sub(rhs.cast_u64()))
+    }
+}
+
+impl<T: AddressOp> ops::AddAssign<T> for Address {
+    fn add_assign(&mut self, rhs: T) {
+        self.0 = self.0.wrapping_add(rhs.cast_u64())
+    }
+}
+
+impl<T: AddressOp> ops::SubAssign<T> for Address {
+    fn sub_assign(&mut self, rhs: T) {
+        self.0 = self.0.wrapping_sub(rhs.cast_u64())
+    }
+}
+
+impl From<Address> for u64 {
+    #[inline(always)]
+    fn from(value: Address) -> Self {
+        value.0
+    }
+}
+
+impl Debug for Address {
+    fn fmt(&self, f: &mut core::fmt::Formatter<'_>) -> core::fmt::Result {
+        write!(f, "[{:0x}]", self.0)
+    }
+}
+
+/// Can perform address operations with
+pub trait AddressOp {
+    /// Cast to u64, truncating or extending
+    fn cast_u64(self) -> u64;
+}
+
+macro_rules! impl_address_ops_u(($($ty:ty),* $(,)?) => {
+    $(impl AddressOp for $ty {
+        #[inline(always)]
+        fn cast_u64(self) -> u64 { self as _ }
+    })*
+});
+
+macro_rules! impl_address_ops_i(($($ty:ty),* $(,)?) => {
+    $(impl AddressOp for $ty {
+        #[inline(always)]
+        fn cast_u64(self) -> u64 { self as i64 as u64 }
+    })*
+});
+
+impl_address_ops_u!(u8, u16, u32, u64, usize);
+impl_address_ops_i!(i8, i16, i32, i64, isize);

hbvm/src/mem/mod.rs

@@ -0,0 +1,80 @@
+//! Memory implementations
+
+pub mod softpaging;
+
+pub(crate) mod addr;
+
+pub use addr::Address;
+
+use crate::utils::impl_display;
+
+/// Load-store memory access
+pub trait Memory {
+    /// Load data from memory on address
+    ///
+    /// # Safety
+    /// - Shall not overrun the buffer
+    unsafe fn load(
+        &mut self,
+        addr: Address,
+        target: *mut u8,
+        count: usize,
+    ) -> Result<(), LoadError>;
+
+    /// Store data to memory on address
+    ///
+    /// # Safety
+    /// - Shall not overrun the buffer
+    unsafe fn store(
+        &mut self,
+        addr: Address,
+        source: *const u8,
+        count: usize,
+    ) -> Result<(), StoreError>;
+
+    /// Read from program memory to execute
+    ///
+    /// # Safety
+    /// - Data read have to be valid
+    unsafe fn prog_read<T: Copy>(&mut self, addr: Address) -> T;
+}
+
+/// Unhandled load access trap
+#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+pub struct LoadError(pub Address);
+impl_display!(for LoadError =>
+    |LoadError(a)| "Load access error at address {a}",
+);
+
+/// Unhandled store access trap
+#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+pub struct StoreError(pub Address);
+impl_display!(for StoreError =>
+    |StoreError(a)| "Load access error at address {a}",
+);
+
+/// Reason to access memory
+#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+pub enum MemoryAccessReason {
+    /// Memory was accessed for load (read)
+    Load,
+    /// Memory was accessed for store (write)
+    Store,
+}
+
+impl_display!(for MemoryAccessReason => match {
+    Self::Load  => const "Load";
+    Self::Store => const "Store";
+});
+
+impl From<LoadError> for crate::VmRunError {
+    fn from(value: LoadError) -> Self {
+        Self::LoadAccessEx(value.0)
+    }
+}
+
+impl From<StoreError> for crate::VmRunError {
+    fn from(value: StoreError) -> Self {
+        Self::StoreAccessEx(value.0)
+    }
+}

hbvm/src/mem/softpaging/icache.rs

@@ -0,0 +1,111 @@
+//! Program instruction cache
+
+use crate::mem::Address;
+
+use {
+    super::{lookup::AddrPageLookuper, paging::PageTable, PageSize},
+    core::{
+        mem::{size_of, MaybeUninit},
+        ptr::{copy_nonoverlapping, NonNull},
+    },
+};
+
+/// Instruction cache
+#[derive(Clone, Debug)]
+pub struct ICache {
+    /// Current page address base
+    base: Address,
+    /// Curent page pointer
+    data: Option<NonNull<u8>>,
+    /// Current page size
+    size: PageSize,
+    /// Address mask
+    mask: u64,
+}
+
+impl Default for ICache {
+    fn default() -> Self {
+        Self {
+            base: Address::NULL,
+            data: Default::default(),
+            size: PageSize::Size4K,
+            mask: Default::default(),
+        }
+    }
+}
+
+impl ICache {
+    /// Fetch instruction from cache
+    ///
+    /// # Safety
+    /// `T` should be valid to read from instruction memory
+    pub(super) unsafe fn fetch<T>(
+        &mut self,
+        addr: Address,
+        root_pt: *const PageTable,
+    ) -> Option<T> {
+        let mut ret = MaybeUninit::<T>::uninit();
+
+        let pbase = self
+            .data
+            .or_else(|| unsafe { self.fetch_page(self.base + self.size, root_pt) })?;
+
+        // Get address base
+        let base = addr.map(|x| x & self.mask);
+
+        // Base not matching, fetch anew
+        if base != self.base {
+            unsafe { self.fetch_page(base, root_pt) }?;
+        };
+
+        let offset = addr.get() & !self.mask;
+        let requ_size = size_of::<T>();
+
+        // Page overflow
+        let rem = (offset as usize)
+            .saturating_add(requ_size)
+            .saturating_sub(self.size as _);
+        let first_copy = requ_size.saturating_sub(rem);
+
+        // Copy non-overflowing part
+        unsafe { copy_nonoverlapping(pbase.as_ptr(), ret.as_mut_ptr().cast::<u8>(), first_copy) };
+
+        // Copy overflow
+        if rem != 0 {
+            let pbase = unsafe { self.fetch_page(self.base + self.size, root_pt) }?;
+
+            // Unlikely, unsupported scenario
+            if rem > self.size as _ {
+                return None;
+            }
+
+            unsafe {
+                copy_nonoverlapping(
+                    pbase.as_ptr(),
+                    ret.as_mut_ptr().cast::<u8>().add(first_copy),
+                    rem,
+                )
+            };
+        }
+
+        Some(unsafe { ret.assume_init() })
+    }
+
+    /// Fetch a page
+    unsafe fn fetch_page(&mut self, addr: Address, pt: *const PageTable) -> Option<NonNull<u8>> {
+        let res = AddrPageLookuper::new(addr, 0, pt).next()?.ok()?;
+        if !super::perm_check::executable(res.perm) {
+            return None;
+        }
+
+        (self.size, self.mask) = match res.size {
+            4096 => (PageSize::Size4K, !((1 << 8) - 1)),
+            2097152 => (PageSize::Size2M, !((1 << (8 * 2)) - 1)),
+            1073741824 => (PageSize::Size1G, !((1 << (8 * 3)) - 1)),
+            _ => return None,
+        };
+        self.data = Some(NonNull::new(res.ptr)?);
+        self.base = addr.map(|x| x & self.mask);
+        self.data
+    }
+}

hbvm/src/mem/softpaging/lookup.rs

@@ -0,0 +1,126 @@
+//! Address lookup
+
+use crate::mem::addr::Address;
+
+use super::{
+    addr_extract_index,
+    paging::{PageTable, Permission},
+    PageSize,
+};
+
+/// Good result from address split
+pub struct AddrPageLookupOk {
+    /// Virtual address
+    pub vaddr: Address,
+
+    /// Pointer to the start for perform operation
+    pub ptr: *mut u8,
+
+    /// Size to the end of page / end of desired size
+    pub size: usize,
+
+    /// Page permission
+    pub perm: Permission,
+}
+
+/// Errornous address split result
+pub struct AddrPageLookupError {
+    /// Address of failure
+    pub addr: Address,
+
+    /// Requested page size
+    pub size: PageSize,
+}
+
+/// Address splitter into pages
+pub struct AddrPageLookuper {
+    /// Current address
+    addr: Address,
+
+    /// Size left
+    size: usize,
+
+    /// Page table
+    pagetable: *const PageTable,
+}
+
+impl AddrPageLookuper {
+    /// Create a new page lookuper
+    #[inline]
+    pub const fn new(addr: Address, size: usize, pagetable: *const PageTable) -> Self {
+        Self {
+            addr,
+            size,
+            pagetable,
+        }
+    }
+
+    /// Bump address by size X
+    pub fn bump(&mut self, page_size: PageSize) {
+        self.addr += page_size;
+        self.size = self.size.saturating_sub(page_size as _);
+    }
+}
+
+impl Iterator for AddrPageLookuper {
+    type Item = Result<AddrPageLookupOk, AddrPageLookupError>;
+
+    fn next(&mut self) -> Option<Self::Item> {
+        // The end, everything is fine
+        if self.size == 0 {
+            return None;
+        }
+
+        let (base, perm, size, offset) = 'a: {
+            let mut current_pt = self.pagetable;
+
+            // Walk the page table
+            for lvl in (0..5).rev() {
+                // Get an entry
+                unsafe {
+                    let entry = (*current_pt)
+                        .table
+                        .get_unchecked(addr_extract_index(self.addr, lvl));
+
+                    let ptr = entry.ptr();
+                    match entry.permission() {
+                        // No page → page fault
+                        Permission::Empty => {
+                            return Some(Err(AddrPageLookupError {
+                                addr: self.addr,
+                                size: PageSize::from_lvl(lvl)?,
+                            }))
+                        }
+
+                        // Node → proceed waking
+                        Permission::Node => current_pt = ptr as _,
+
+                        // Leaf → return relevant data
+                        perm => {
+                            break 'a (
+                                // Pointer in host memory
+                                ptr as *mut u8,
+                                perm,
+                                PageSize::from_lvl(lvl)?,
+                                // In-page offset
+                                addr_extract_index(self.addr, lvl),
+                            );
+                        }
+                    }
+                }
+            }
+            return None; // Reached the end (should not happen)
+        };
+
+        // Get available byte count in the selected page with offset
+        let avail = (size as usize).saturating_sub(offset).clamp(0, self.size);
+        self.bump(size);
+
+        Some(Ok(AddrPageLookupOk {
+            vaddr: self.addr,
+            ptr: unsafe { base.add(offset) }, // Return pointer to the start of region
+            size: avail,
+            perm,
+        }))
+    }
+}

hbvm/src/mem/softpaging/mapping.rs

@@ -0,0 +1,172 @@
+//! Automatic memory mapping
+
+use crate::{mem::addr::Address, utils::impl_display};
+
+use {
+    super::{
+        addr_extract_index,
+        paging::{PageTable, Permission, PtEntry, PtPointedData},
+        PageSize, SoftPagedMem,
+    },
+    alloc::boxed::Box,
+};
+
+impl<'p, A, const OUT_PROG_EXEC: bool> SoftPagedMem<'p, A, OUT_PROG_EXEC> {
+    /// Maps host's memory into VM's memory
+    ///
+    /// # Safety
+    /// - Your faith in the gods of UB
+    ///     - Addr-san claims it's fine but who knows is she isn't lying :ferrisSus:
+    ///     - Alright, Miri-sama is also fine with this, who knows why
+    pub unsafe fn map(
+        &mut self,
+        host: *mut u8,
+        target: Address,
+        perm: Permission,
+        pagesize: PageSize,
+    ) -> Result<(), MapError> {
+        let mut current_pt = self.root_pt;
+
+        // Decide on what level depth are we going
+        let lookup_depth = match pagesize {
+            PageSize::Size4K => 0,
+            PageSize::Size2M => 1,
+            PageSize::Size1G => 2,
+        };
+
+        // Walk pagetable levels
+        for lvl in (lookup_depth + 1..5).rev() {
+            let entry = unsafe {
+                (*current_pt)
+                    .table
+                    .get_unchecked_mut(addr_extract_index(target, lvl))
+            };
+
+            let ptr = entry.ptr();
+            match entry.permission() {
+                // Still not on target and already seeing empty entry?
+                // No worries! Let's create one (allocates).
+                Permission::Empty => {
+                    // Increase children count
+                    unsafe { *current_pt }.childen += 1;
+
+                    let table = Box::into_raw(Box::new(PtPointedData {
+                        pt: PageTable::default(),
+                    }));
+
+                    unsafe { core::ptr::write(entry, PtEntry::new(table, Permission::Node)) };
+                    current_pt = table as _;
+                }
+                // Continue walking
+                Permission::Node => current_pt = ptr as _,
+
+                // There is some entry on place of node
+                _ => return Err(MapError::PageOnNode),
+            }
+        }
+
+        let node = unsafe {
+            (*current_pt)
+                .table
+                .get_unchecked_mut(addr_extract_index(target, lookup_depth))
+        };
+
+        // Check if node is not mapped
+        if node.permission() != Permission::Empty {
+            return Err(MapError::AlreadyMapped);
+        }
+
+        // Write entry
+        unsafe {
+            (*current_pt).childen += 1;
+            core::ptr::write(node, PtEntry::new(host.cast(), perm));
+        }
+
+        Ok(())
+    }
+
+    /// Unmaps pages from VM's memory
+    ///
+    /// If errors, it only means there is no entry to unmap and in most cases
+    /// just should be ignored.
+    pub fn unmap(&mut self, addr: Address) -> Result<(), NothingToUnmap> {
+        let mut current_pt = self.root_pt;
+        let mut page_tables = [core::ptr::null_mut(); 5];
+
+        // Walk page table in reverse
+        for lvl in (0..5).rev() {
+            let entry = unsafe {
+                (*current_pt)
+                    .table
+                    .get_unchecked_mut(addr_extract_index(addr, lvl))
+            };
+
+            let ptr = entry.ptr();
+            match entry.permission() {
+                // Nothing is there, throw an error, not critical!
+                Permission::Empty => return Err(NothingToUnmap),
+                // Node – Save to visited pagetables and continue walking
+                Permission::Node => {
+                    page_tables[lvl as usize] = entry;
+                    current_pt = ptr as _
+                }
+                // Page entry – zero it out!
+                // Zero page entry is completely valid entry with
+                // empty permission - no UB here!
+                _ => unsafe {
+                    core::ptr::write_bytes(entry, 0, 1);
+                    break;
+                },
+            }
+        }
+
+        // Now walk in order visited page tables
+        for entry in page_tables.into_iter() {
+            // Level not visited, skip.
+            if entry.is_null() {
+                continue;
+            }
+
+            unsafe {
+                let children = &mut (*(*entry).ptr()).pt.childen;
+                *children -= 1; // Decrease children count
+
+                // If there are no children, deallocate.
+                if *children == 0 {
+                    let _ = Box::from_raw((*entry).ptr() as *mut PageTable);
+
+                    // Zero visited entry
+                    core::ptr::write_bytes(entry, 0, 1);
+                } else {
+                    break;
+                }
+            }
+        }
+
+        Ok(())
+    }
+}
+
+/// Error mapping
+#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+pub enum MapError {
+    /// Entry was already mapped
+    AlreadyMapped,
+    /// When walking a page entry was
+    /// encounterd.
+    PageOnNode,
+}
+
+impl_display!(for MapError => match {
+    Self::AlreadyMapped => "There is already a page mapped on specified address";
+    Self::PageOnNode    => "There was a page mapped on the way instead of node";
+});
+
+/// There was no entry in page table to unmap
+///
+/// No worry, don't panic, nothing bad has happened,
+/// but if you are 120% sure there should be something,
+/// double-check your addresses.
+#[derive(Clone, Copy, Debug)]
+pub struct NothingToUnmap;
+impl_display!(for NothingToUnmap => "There is no entry to unmap");

hbvm/src/mem/softpaging/mod.rs

@@ -0,0 +1,282 @@
+//! Platform independent, software paged memory implementation
+
+pub mod icache;
+pub mod lookup;
+pub mod paging;
+
+#[cfg(feature = "alloc")]
+pub mod mapping;
+
+use {
+    super::{addr::Address, LoadError, Memory, MemoryAccessReason, StoreError},
+    core::mem::size_of,
+    icache::ICache,
+    lookup::{AddrPageLookupError, AddrPageLookupOk, AddrPageLookuper},
+    paging::{PageTable, Permission},
+};
+
+/// HoleyBytes software paged memory
+///
+/// - `OUT_PROG_EXEC`: set to `false` to disable executing program
+///   not contained in initially provided program, even the pages
+///   are executable
+#[derive(Clone, Debug)]
+pub struct SoftPagedMem<'p, PfH, const OUT_PROG_EXEC: bool = true> {
+    /// Root page table
+    pub root_pt:    *mut PageTable,
+    /// Page fault handler
+    pub pf_handler: PfH,
+    /// Program memory segment
+    pub program:    &'p [u8],
+    /// Program instruction cache
+    pub icache:     ICache,
+}
+
+impl<'p, PfH: HandlePageFault, const OUT_PROG_EXEC: bool> Memory
+    for SoftPagedMem<'p, PfH, OUT_PROG_EXEC>
+{
+    /// Load value from an address
+    ///
+    /// # Safety
+    /// Applies same conditions as for [`core::ptr::copy_nonoverlapping`]
+    unsafe fn load(
+        &mut self,
+        addr: Address,
+        target: *mut u8,
+        count: usize,
+    ) -> Result<(), LoadError> {
+        self.memory_access(
+            MemoryAccessReason::Load,
+            addr,
+            target,
+            count,
+            perm_check::readable,
+            |src, dst, count| unsafe { core::ptr::copy_nonoverlapping(src, dst, count) },
+        )
+        .map_err(LoadError)
+    }
+
+    /// Store value to an address
+    ///
+    /// # Safety
+    /// Applies same conditions as for [`core::ptr::copy_nonoverlapping`]
+    unsafe fn store(
+        &mut self,
+        addr: Address,
+        source: *const u8,
+        count: usize,
+    ) -> Result<(), StoreError> {
+        self.memory_access(
+            MemoryAccessReason::Store,
+            addr,
+            source.cast_mut(),
+            count,
+            perm_check::writable,
+            |dst, src, count| unsafe { core::ptr::copy_nonoverlapping(src, dst, count) },
+        )
+        .map_err(StoreError)
+    }
+
+    #[inline(always)]
+    unsafe fn prog_read<T>(&mut self, addr: Address) -> T {
+        if OUT_PROG_EXEC && addr.truncate_usize() > self.program.len() {
+            return unsafe { self.icache.fetch::<T>(addr, self.root_pt) }
+                .unwrap_or_else(|| unsafe { core::mem::zeroed() });
+        }
+
+        let addr = addr.truncate_usize();
+        self.program
+            .get(addr..addr + size_of::<T>())
+            .map(|x| unsafe { x.as_ptr().cast::<T>().read() })
+            .unwrap_or_else(|| unsafe { core::mem::zeroed() })
+    }
+}
+
+impl<'p, PfH: HandlePageFault, const OUT_PROG_EXEC: bool> SoftPagedMem<'p, PfH, OUT_PROG_EXEC> {
+    // Everyone behold, the holy function, the god of HBVM memory accesses!
+
+    /// Split address to pages, check their permissions and feed pointers with offset
+    /// to a specified function.
+    ///
+    /// If page is not found, execute page fault trap handler.
+    #[allow(clippy::too_many_arguments)] // Silence peasant
+    fn memory_access(
+        &mut self,
+        reason: MemoryAccessReason,
+        src: Address,
+        mut dst: *mut u8,
+        len: usize,
+        permission_check: fn(Permission) -> bool,
+        action: fn(*mut u8, *mut u8, usize),
+    ) -> Result<(), Address> {
+        // Memory load from program section
+        let (src, len) = if src.truncate_usize() < self.program.len() as _ {
+            // Allow only loads
+            if reason != MemoryAccessReason::Load {
+                return Err(src);
+            }
+
+            // Determine how much data to copy from here
+            let to_copy = len.clamp(0, self.program.len().saturating_sub(src.truncate_usize()));
+
+            // Perform action
+            action(
+                unsafe { self.program.as_ptr().add(src.truncate_usize()).cast_mut() },
+                dst,
+                to_copy,
+            );
+
+            // Return shifted from what we've already copied
+            (
+                src.saturating_add(to_copy as u64),
+                len.saturating_sub(to_copy),
+            )
+        } else {
+            (src, len) // Nothing weird!
+        };
+
+        // Nothing to copy? Don't bother doing anything, bail.
+        if len == 0 {
+            return Ok(());
+        }
+
+        // Create new splitter
+        let mut pspl = AddrPageLookuper::new(src, len, self.root_pt);
+        loop {
+            match pspl.next() {
+                // Page is found
+                Some(Ok(AddrPageLookupOk {
+                    vaddr,
+                    ptr,
+                    size,
+                    perm,
+                })) => {
+                    if !permission_check(perm) {
+                        return Err(vaddr);
+                    }
+
+                    // Perform specified memory action and bump destination pointer
+                    action(ptr, dst, size);
+                    dst = unsafe { dst.add(size) };
+                }
+                // No page found
+                Some(Err(AddrPageLookupError { addr, size })) => {
+                    // Attempt to execute page fault handler
+                    if self.pf_handler.page_fault(
+                        reason,
+                        unsafe { &mut *self.root_pt },
+                        addr,
+                        size,
+                        dst,
+                    ) {
+                        // Shift the splitter address
+                        pspl.bump(size);
+
+                        // Bump dst pointer
+                        dst = unsafe { dst.add(size as _) };
+                    } else {
+                        return Err(addr); // Unhandleable, VM will yield.
+                    }
+                }
+                // No remaining pages, we are done!
+                None => return Ok(()),
+            }
+        }
+    }
+}
+
+/// Extract index in page table on specified level
+///
+/// The level shall not be larger than 4, otherwise
+/// the output of the function is unspecified (yes, it can also panic :)
+pub fn addr_extract_index(addr: Address, lvl: u8) -> usize {
+    debug_assert!(lvl <= 4);
+    let addr = addr.get();
+    usize::try_from((addr >> (lvl * 8 + 12)) & ((1 << 8) - 1)).expect("?conradluget a better CPU")
+}
+
+/// Page size
+#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+pub enum PageSize {
+    /// 4 KiB page (on level 0)
+    Size4K = 4096,
+
+    /// 2 MiB page (on level 1)
+    Size2M = 1024 * 1024 * 2,
+
+    /// 1 GiB page (on level 2)
+    Size1G = 1024 * 1024 * 1024,
+}
+
+impl PageSize {
+    /// Convert page table level to size of page
+    const fn from_lvl(lvl: u8) -> Option<Self> {
+        match lvl {
+            0 => Some(PageSize::Size4K),
+            1 => Some(PageSize::Size2M),
+            2 => Some(PageSize::Size1G),
+            _ => None,
+        }
+    }
+}
+
+impl core::ops::Add<PageSize> for Address {
+    type Output = Self;
+
+    #[inline(always)]
+    fn add(self, rhs: PageSize) -> Self::Output {
+        self + (rhs as u64)
+    }
+}
+
+impl core::ops::AddAssign<PageSize> for Address {
+    #[inline(always)]
+    fn add_assign(&mut self, rhs: PageSize) {
+        *self = Self::new(self.get().wrapping_add(rhs as u64));
+    }
+}
+
+/// Permisison checks
+pub mod perm_check {
+    use super::paging::Permission;
+
+    /// Page is readable
+    #[inline(always)]
+    pub const fn readable(perm: Permission) -> bool {
+        matches!(
+            perm,
+            Permission::Readonly | Permission::Write | Permission::Exec
+        )
+    }
+
+    /// Page is writable
+    #[inline(always)]
+    pub const fn writable(perm: Permission) -> bool {
+        matches!(perm, Permission::Write)
+    }
+
+    /// Page is executable
+    #[inline(always)]
+    pub const fn executable(perm: Permission) -> bool {
+        matches!(perm, Permission::Exec)
+    }
+}
+
+/// Handle VM traps
+pub trait HandlePageFault {
+    /// Handle page fault
+    ///
+    /// Return true if handling was sucessful,
+    /// otherwise the program will be interrupted and will
+    /// yield an error.
+    fn page_fault(
+        &mut self,
+        reason: MemoryAccessReason,
+        pagetable: &mut PageTable,
+        vaddr: Address,
+        size: PageSize,
+        dataptr: *mut u8,
+    ) -> bool
+    where
+        Self: Sized;
+}

hbvm/src/mem/softpaging/paging.rs

@@ -0,0 +1,86 @@
+//! Page table and associated structures implementation
+
+use core::{fmt::Debug, mem::MaybeUninit};
+
+/// Page entry permission
+#[derive(Clone, Copy, Debug, Default, PartialEq, Eq)]
+#[repr(u8)]
+pub enum Permission {
+    /// No page present
+    #[default]
+    Empty,
+    /// Points to another pagetable
+    Node,
+    /// Page is read only
+    Readonly,
+    /// Page is readable and writable
+    Write,
+    /// Page is readable and executable
+    Exec,
+}
+
+/// Page table entry
+#[derive(Clone, Copy, Default, PartialEq, Eq)]
+pub struct PtEntry(u64);
+impl PtEntry {
+    /// Create new
+    ///
+    /// # Safety
+    /// - `ptr` has to point to valid data and shall not be deallocated
+    ///    troughout the entry lifetime
+    #[inline]
+    pub unsafe fn new(ptr: *mut PtPointedData, permission: Permission) -> Self {
+        Self(ptr as u64 | permission as u64)
+    }
+
+    /// Get permission
+    #[inline]
+    pub fn permission(&self) -> Permission {
+        unsafe { core::mem::transmute(self.0 as u8 & 0b111) }
+    }
+
+    /// Get pointer to the data (leaf) or next page table (node)
+    #[inline]
+    pub fn ptr(&self) -> *mut PtPointedData {
+        (self.0 & !((1 << 12) - 1)) as _
+    }
+}
+
+impl Debug for PtEntry {
+    fn fmt(&self, f: &mut core::fmt::Formatter<'_>) -> core::fmt::Result {
+        f.debug_struct("PtEntry")
+            .field("ptr", &self.ptr())
+            .field("permission", &self.permission())
+            .finish()
+    }
+}
+
+/// Page table
+#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+#[repr(align(4096))]
+pub struct PageTable {
+    /// How much entries are in use
+    pub childen: u8,
+    /// Entries
+    pub table:   [PtEntry; 256],
+}
+
+impl Default for PageTable {
+    fn default() -> Self {
+        // SAFETY: It's fine, zeroed page table entry is valid (= empty)
+        Self {
+            childen: 0,
+            table:   unsafe { MaybeUninit::zeroed().assume_init() },
+        }
+    }
+}
+
+/// Data page table entry can possibly point to
+#[derive(Clone, Copy)]
+#[repr(C, align(4096))]
+pub union PtPointedData {
+    /// Node - next page table
+    pub pt:   PageTable,
+    /// Leaf
+    pub page: u8,
+}

hbvm/src/memory.rs

@@ -1,70 +0,0 @@
-use crate::engine::VMPage;
-
-use {
-    crate::{engine::Page, RuntimeErrors},
-    alloc::vec::Vec,
-    hashbrown::HashMap,
-    log::trace,
-};
-
-pub struct Memory {
-    inner: HashMap<u64, Page>,
-}
-
-impl Memory {
-    pub fn new() -> Self {
-        Self {
-            inner: HashMap::new(),
-        }
-        //
-    }
-
-    pub fn map_vec(&mut self, address: u64, vec: Vec<u8>) {
-        panic!("Mapping vectors into pages is not supported yet");
-    }
-}
-
-impl Memory {
-    pub fn read_addr8(&mut self, address: u64) -> Result<u8, RuntimeErrors> {
-        let (page, offset) = addr_to_page(address);
-        trace!("page {} offset {}", page, offset);
-        match self.inner.get(&page) {
-            Some(page) => {
-                let val = page.data()[offset as usize];
-                trace!("Value {}", val);
-                Ok(val)
-            }
-            None => {
-                trace!("page not mapped");
-                Err(RuntimeErrors::PageNotMapped(page))
-            }
-        }
-    }
-    pub fn read_addr64(&mut self, address: u64) -> u64 {
-        unimplemented!()
-    }
-
-    pub fn set_addr8(&mut self, address: u64, value: u8) -> Result<(), RuntimeErrors> {
-        let (page, offset) = addr_to_page(address);
-        let ret: Option<(&u64, &mut Page)> = self.inner.get_key_value_mut(&page);
-        match ret {
-            Some((_, page)) => {
-                page.data()[offset as usize] = value;
-            }
-            None => {
-                let mut pg = VMPage::new();
-                pg.data[offset as usize] = value;
-                self.inner.insert(page, Page::VMPage(pg));
-                trace!("Mapped page {}", page);
-            }
-        }
-        Ok(())
-    }
-    pub fn set_addr64(&mut self, address: u64, value: u64) -> Result<(), RuntimeErrors> {
-        unimplemented!()
-    }
-}
-
-fn addr_to_page(addr: u64) -> (u64, u64) {
-    (addr / 8192, addr % 8192)
-}

hbvm/src/utils.rs

@@ -0,0 +1,53 @@
+macro_rules! impl_display {
+    (for $ty:ty => $(|$selfty:pat_param|)? $fmt:literal $(, $($param:expr),+)? $(,)?) => {
+        impl ::core::fmt::Display for $ty {
+            fn fmt(&self, f: &mut ::core::fmt::Formatter<'_>) -> ::core::fmt::Result {
+                $(let $selfty = self;)?
+                write!(f, $fmt, $($param),*)
+            }
+        }
+    };
+
+    (for $ty:ty => $str:literal) => {
+        impl ::core::fmt::Display for $ty {
+            fn fmt(&self, f: &mut ::core::fmt::Formatter<'_>) -> ::core::fmt::Result {
+                f.write_str($str)
+            }
+        }
+    };
+
+    (for $ty:ty => match {$(
+        $bind:pat => $($const:ident)? $fmt:literal $(,$($params:tt)*)?;
+    )*}) => {
+        impl ::core::fmt::Display for $ty {
+            fn fmt(&self, f: &mut ::core::fmt::Formatter<'_>) -> ::core::fmt::Result {
+                match self {
+                    $(
+                        $bind => $crate::utils::internal::impl_display_match_fragment!($($const,)? f, $fmt $(, $($params)*)?)
+                    ),*
+                }
+            }
+        }
+    }
+}
+
+#[doc(hidden)]
+pub(crate) mod internal {
+    macro_rules! impl_display_match_fragment {
+        (const, $f:expr, $lit:literal) => {
+            $f.write_str($lit)
+        };
+
+        ($f:expr, $fmt:literal $(, $($params:tt)*)?) => {
+            write!($f, $fmt, $($($params)*)?)
+        };
+    }
+
+    pub(crate) use impl_display_match_fragment;
+}
+
+macro_rules! static_assert(($expr:expr $(,)?) => {
+    const _: [(); !$expr as usize] = [];
+});
+
+pub(crate) use {impl_display, static_assert};

hbvm/src/value.rs

@@ -0,0 +1,88 @@
+//! HoleyBytes register value definition
+
+use crate::utils::static_assert;
+
+/// Define [`Value`] »union« (it's fake)
+///
+/// # Safety
+/// Its variants have to be sound to byte-reinterpretate
+/// between each other. Otherwise the behaviour is undefined.
+macro_rules! value_def {
+    ($($ty:ident),* $(,)?) => {
+        /// HBVM register value
+        #[derive(Copy, Clone)]
+        #[repr(transparent)]
+        pub struct Value(pub u64);
+
+        $(
+            impl From<$ty> for Value {
+                #[inline]
+                fn from(value: $ty) -> Self {
+                    let mut new = core::mem::MaybeUninit::<u64>::zeroed();
+                    unsafe {
+                        new.as_mut_ptr().cast::<$ty>().write(value);
+                        Self(new.assume_init())
+                    }
+                }
+            }
+
+            static_assert!(core::mem::size_of::<$ty>() <= core::mem::size_of::<Value>());
+
+            impl private::Sealed for $ty {}
+            unsafe impl ValueVariant for $ty {}
+        )*
+    };
+}
+
+impl Value {
+    /// Byte reinterpret value to target variant
+    #[inline]
+    pub fn cast<V: ValueVariant>(self) -> V {
+        unsafe { core::mem::transmute_copy(&self.0) }
+    }
+}
+
+/// # Safety
+/// - N/A, not to be implemented manually
+pub unsafe trait ValueVariant: private::Sealed + Copy + Into<Value> {}
+impl private::Sealed for Value {}
+unsafe impl ValueVariant for Value {}
+
+mod private {
+    pub trait Sealed {}
+}
+
+value_def!(u8, u16, u32, u64, i8, i16, i32, i64, f32, f64);
+static_assert!(core::mem::size_of::<Value>() == 8);
+
+impl core::fmt::Debug for Value {
+    fn fmt(&self, f: &mut core::fmt::Formatter<'_>) -> core::fmt::Result {
+        // Print formatted as hexadecimal, unsigned integer
+        write!(f, "{:x}", self.cast::<u64>())
+    }
+}
+
+pub(crate) trait CheckedDivRem {
+    fn checked_div(self, other: Self) -> Option<Self>
+    where
+        Self: Sized;
+    fn checked_rem(self, other: Self) -> Option<Self>
+    where
+        Self: Sized;
+}
+
+macro_rules! impl_checked_div_rem {
+    ($($ty:ty),* $(,)?) => {
+        $(impl CheckedDivRem for $ty {
+            #[inline(always)]
+            fn checked_div(self, another: Self) -> Option<Self>
+                { self.checked_div(another) }
+
+            #[inline(always)]
+            fn checked_rem(self, another: Self) -> Option<Self>
+                { self.checked_rem(another) }
+        })*
+    };
+}
+
+impl_checked_div_rem!(u8, u16, u32, u64, i8, i16, i32, i64);

hbvm/src/vmrun.rs

@@ -0,0 +1,578 @@
+//! Welcome to the land of The Great Dispatch Loop
+//!
+//! Have fun
+
+use {
+    super::{bmc::BlockCopier, mem::Memory, value::ValueVariant, Vm, VmRunError, VmRunOk},
+    crate::{
+        mem::{addr::AddressOp, Address},
+        value::CheckedDivRem,
+    },
+    core::{cmp::Ordering, ops},
+    hbbytecode::{
+        OpsN, OpsO, OpsP, OpsRB, OpsRD, OpsRH, OpsRR, OpsRRA, OpsRRAH, OpsRRB, OpsRRD, OpsRRH,
+        OpsRRO, OpsRROH, OpsRRP, OpsRRPH, OpsRRR, OpsRRRR, OpsRW, RoundingMode,
+    },
+};
+
+macro_rules! handler {
+    ($self:expr, |$ty:ident ($($ident:pat),* $(,)?)| $expr:expr) => {{
+        #[allow(unused_unsafe)]
+        let $ty($($ident),*) = unsafe { $self.decode::<$ty>() };
+        #[allow(clippy::no_effect)] let e = $expr;
+        $self.bump_pc::<$ty>();
+        e
+    }};
+}
+
+impl<Mem, const TIMER_QUOTIENT: usize> Vm<Mem, TIMER_QUOTIENT>
+where
+    Mem: Memory,
+{
+    /// Execute program
+    ///
+    /// Program can return [`VmRunError`] if a trap handling failed
+    #[cfg_attr(feature = "nightly", repr(align(4096)))]
+    pub fn run(&mut self) -> Result<VmRunOk, VmRunError> {
+        use hbbytecode::opcode::*;
+        loop {
+            // Big match
+            //
+            // Contribution guide:
+            // - Zero register shall never be overwitten. It's value has to always be 0.
+            //     - Prefer `Self::read_reg` and `Self::write_reg` functions
+            // - Try to use `handler!` macro for decoding and then bumping program counter
+            // - Prioritise speed over code size
+            //     - Memory is cheap, CPUs not that much
+            // - Do not heap allocate at any cost
+            //     - Yes, user-provided trap handler may allocate,
+            //       but that is not our »fault«.
+            // - Unsafe is kinda must, but be sure you have validated everything
+            //     - Your contributions have to pass sanitizers, fuzzer and Miri
+            // - Strictly follow the spec
+            //     - The spec does not specify how you perform actions, in what order,
+            //       just that the observable effects have to be performed in order and
+            //       correctly.
+            // - Yes, we assume you run 64 bit CPU. Else ?conradluget a better CPU
+            //   sorry 8 bit fans, HBVM won't run on your Speccy :(
+            unsafe {
+                match self.memory.prog_read::<u8>(self.pc as _) {
+                    UN => {
+                        self.bump_pc::<OpsN>();
+                        return Err(VmRunError::Unreachable);
+                    }
+                    TX => {
+                        self.bump_pc::<OpsN>();
+                        return Ok(VmRunOk::End);
+                    }
+                    NOP => handler!(self, |OpsN()| ()),
+                    ADD8 => self.binary_op(u8::wrapping_add),
+                    ADD16 => self.binary_op(u16::wrapping_add),
+                    ADD32 => self.binary_op(u32::wrapping_add),
+                    ADD64 => self.binary_op(u64::wrapping_add),
+                    SUB8 => self.binary_op(u8::wrapping_sub),
+                    SUB16 => self.binary_op(u16::wrapping_sub),
+                    SUB32 => self.binary_op(u32::wrapping_sub),
+                    SUB64 => self.binary_op(u64::wrapping_sub),
+                    MUL8 => self.binary_op(u8::wrapping_mul),
+                    MUL16 => self.binary_op(u16::wrapping_mul),
+                    MUL32 => self.binary_op(u32::wrapping_mul),
+                    MUL64 => self.binary_op(u64::wrapping_mul),
+                    AND => self.binary_op::<u64>(ops::BitAnd::bitand),
+                    OR => self.binary_op::<u64>(ops::BitOr::bitor),
+                    XOR => self.binary_op::<u64>(ops::BitXor::bitxor),
+                    SLU8 => self.binary_op_shift::<u8>(u8::wrapping_shl),
+                    SLU16 => self.binary_op_shift::<u16>(u16::wrapping_shl),
+                    SLU32 => self.binary_op_shift::<u32>(u32::wrapping_shl),
+                    SLU64 => self.binary_op_shift::<u64>(u64::wrapping_shl),
+                    SRU8 => self.binary_op_shift::<u8>(u8::wrapping_shr),
+                    SRU16 => self.binary_op_shift::<u16>(u16::wrapping_shr),
+                    SRU32 => self.binary_op_shift::<u32>(u32::wrapping_shr),
+                    SRU64 => self.binary_op_shift::<u64>(u64::wrapping_shr),
+                    SRS8 => self.binary_op_shift::<i8>(i8::wrapping_shr),
+                    SRS16 => self.binary_op_shift::<i16>(i16::wrapping_shr),
+                    SRS32 => self.binary_op_shift::<i32>(i32::wrapping_shr),
+                    SRS64 => self.binary_op_shift::<i64>(i64::wrapping_shr),
+                    CMPU => handler!(self, |OpsRRR(tg, a0, a1)| self.cmp(
+                        tg,
+                        a0,
+                        self.read_reg(a1).cast::<u64>()
+                    )),
+                    CMPS => handler!(self, |OpsRRR(tg, a0, a1)| self.cmp(
+                        tg,
+                        a0,
+                        self.read_reg(a1).cast::<i64>()
+                    )),
+                    DIRU8 => self.dir::<u8>(),
+                    DIRU16 => self.dir::<u16>(),
+                    DIRU32 => self.dir::<u32>(),
+                    DIRU64 => self.dir::<u64>(),
+                    DIRS8 => self.dir::<i8>(),
+                    DIRS16 => self.dir::<i16>(),
+                    DIRS32 => self.dir::<i32>(),
+                    DIRS64 => self.dir::<i64>(),
+                    NEG => handler!(self, |OpsRR(tg, a0)| {
+                        // Bit negation
+                        self.write_reg(tg, !self.read_reg(a0).cast::<u64>())
+                    }),
+                    NOT => handler!(self, |OpsRR(tg, a0)| {
+                        // Logical negation
+                        self.write_reg(tg, u64::from(self.read_reg(a0).cast::<u64>() == 0));
+                    }),
+                    SXT8 => handler!(self, |OpsRR(tg, a0)| {
+                        self.write_reg(tg, self.read_reg(a0).cast::<i8>() as i64)
+                    }),
+                    SXT16 => handler!(self, |OpsRR(tg, a0)| {
+                        self.write_reg(tg, self.read_reg(a0).cast::<i16>() as i64)
+                    }),
+                    SXT32 => handler!(self, |OpsRR(tg, a0)| {
+                        self.write_reg(tg, self.read_reg(a0).cast::<i32>() as i64)
+                    }),
+                    ADDI8 => self.binary_op_imm(u8::wrapping_add),
+                    ADDI16 => self.binary_op_imm(u16::wrapping_add),
+                    ADDI32 => self.binary_op_imm(u32::wrapping_add),
+                    ADDI64 => self.binary_op_imm(u64::wrapping_add),
+                    MULI8 => self.binary_op_imm(u8::wrapping_mul),
+                    MULI16 => self.binary_op_imm(u16::wrapping_mul),
+                    MULI32 => self.binary_op_imm(u32::wrapping_mul),
+                    MULI64 => self.binary_op_imm(u64::wrapping_mul),
+                    ANDI => self.binary_op_imm::<u64>(ops::BitAnd::bitand),
+                    ORI => self.binary_op_imm::<u64>(ops::BitOr::bitor),
+                    XORI => self.binary_op_imm::<u64>(ops::BitXor::bitxor),
+                    SLUI8 => self.binary_op_ims::<u8>(u8::wrapping_shl),
+                    SLUI16 => self.binary_op_ims::<u16>(u16::wrapping_shl),
+                    SLUI32 => self.binary_op_ims::<u32>(u32::wrapping_shl),
+                    SLUI64 => self.binary_op_ims::<u64>(u64::wrapping_shl),
+                    SRUI8 => self.binary_op_ims::<u8>(u8::wrapping_shr),
+                    SRUI16 => self.binary_op_ims::<u16>(u16::wrapping_shr),
+                    SRUI32 => self.binary_op_ims::<u32>(u32::wrapping_shr),
+                    SRUI64 => self.binary_op_ims::<u64>(u64::wrapping_shr),
+                    SRSI8 => self.binary_op_ims::<i8>(i8::wrapping_shr),
+                    SRSI16 => self.binary_op_ims::<i16>(i16::wrapping_shr),
+                    SRSI32 => self.binary_op_ims::<i32>(i32::wrapping_shr),
+                    SRSI64 => self.binary_op_ims::<i64>(i64::wrapping_shr),
+                    CMPUI => handler!(self, |OpsRRD(tg, a0, imm)| { self.cmp(tg, a0, imm) }),
+                    CMPSI => handler!(self, |OpsRRD(tg, a0, imm)| { self.cmp(tg, a0, imm as i64) }),
+                    CP => handler!(self, |OpsRR(tg, a0)| self.write_reg(tg, self.read_reg(a0))),
+                    SWA => handler!(self, |OpsRR(r0, r1)| {
+                        // Swap registers
+                        match (r0, r1) {
+                            (0, 0) => (),
+                            (dst, 0) | (0, dst) => self.write_reg(dst, 0_u64),
+                            (r0, r1) => {
+                                core::ptr::swap(
+                                    self.registers.get_unchecked_mut(usize::from(r0)),
+                                    self.registers.get_unchecked_mut(usize::from(r1)),
+                                );
+                            }
+                        }
+                    }),
+                    LI8 => handler!(self, |OpsRB(tg, imm)| self.write_reg(tg, imm)),
+                    LI16 => handler!(self, |OpsRH(tg, imm)| self.write_reg(tg, imm)),
+                    LI32 => handler!(self, |OpsRW(tg, imm)| self.write_reg(tg, imm)),
+                    LI64 => handler!(self, |OpsRD(tg, imm)| self.write_reg(tg, imm)),
+                    LRA => handler!(self, |OpsRRO(tg, reg, off)| self.write_reg(
+                        tg,
+                        self.pcrel(off, 3)
+                            .wrapping_add(self.read_reg(reg).cast::<i64>())
+                            .get(),
+                    )),
+                    // Load. If loading more than register size, continue on adjecent registers
+                    LD => handler!(self, |OpsRRAH(dst, base, off, count)| self
+                        .load(dst, base, off, count)?),
+                    // Store. Same rules apply as to LD
+                    ST => handler!(self, |OpsRRAH(dst, base, off, count)| self
+                        .store(dst, base, off, count)?),
+                    LDR => handler!(self, |OpsRROH(dst, base, off, count)| self.load(
+                        dst,
+                        base,
+                        self.pcrel(off, 3).get(),
+                        count
+                    )?),
+                    STR => handler!(self, |OpsRROH(dst, base, off, count)| self.store(
+                        dst,
+                        base,
+                        self.pcrel(off, 3).get(),
+                        count
+                    )?),
+                    BMC => {
+                        // Block memory copy
+                        match if let Some(copier) = &mut self.copier {
+                            // There is some copier, poll.
+                            copier.poll(&mut self.memory)
+                        } else {
+                            // There is none, make one!
+                            let OpsRRH(src, dst, count) = self.decode();
+
+                            self.copier = Some(BlockCopier::new(
+                                Address::new(self.read_reg(src).cast()),
+                                Address::new(self.read_reg(dst).cast()),
+                                count as _,
+                            ));
+
+                            self.copier
+                                .as_mut()
+                                .unwrap_unchecked() // SAFETY: We just assigned there
+                                .poll(&mut self.memory)
+                        } {
+                            // We are done, shift program counter
+                            core::task::Poll::Ready(Ok(())) => {
+                                self.copier = None;
+                                self.bump_pc::<OpsRRH>();
+                            }
+                            // Error, shift program counter (for consistency)
+                            // and yield error
+                            core::task::Poll::Ready(Err(e)) => {
+                                return Err(e.into());
+                            }
+                            // Not done yet, proceed to next cycle
+                            core::task::Poll::Pending => (),
+                        }
+                    }
+                    BRC => handler!(self, |OpsRRB(src, dst, count)| {
+                        // Block register copy
+                        if src.checked_add(count).is_none() || dst.checked_add(count).is_none() {
+                            return Err(VmRunError::RegOutOfBounds);
+                        }
+
+                        core::ptr::copy(
+                            self.registers.get_unchecked(usize::from(src)),
+                            self.registers.get_unchecked_mut(usize::from(dst)),
+                            usize::from(count),
+                        );
+                    }),
+                    JMP => {
+                        let OpsO(off) = self.decode();
+                        self.pc = self.pc.wrapping_add(off).wrapping_add(1);
+                    }
+                    JAL => {
+                        // Jump and link. Save PC after this instruction to
+                        // specified register and jump to reg + relative offset.
+                        let OpsRRO(save, reg, offset) = self.decode();
+
+                        self.write_reg(save, self.pc.get());
+                        self.pc = self
+                            .pcrel(offset, 3)
+                            .wrapping_add(self.read_reg(reg).cast::<i64>());
+                    }
+                    JALA => {
+                        // Jump and link. Save PC after this instruction to
+                        // specified register and jump to reg
+                        let OpsRRA(save, reg, offset) = self.decode();
+
+                        self.write_reg(save, self.pc.get());
+                        self.pc =
+                            Address::new(self.read_reg(reg).cast::<u64>().wrapping_add(offset));
+                    }
+                    // Conditional jumps, jump only to immediates
+                    JEQ => self.cond_jmp::<u64>(Ordering::Equal),
+                    JNE => {
+                        let OpsRRP(a0, a1, ja) = self.decode();
+                        if self.read_reg(a0).cast::<u64>() != self.read_reg(a1).cast::<u64>() {
+                            self.pc = self.pcrel(ja, 3);
+                        } else {
+                            self.bump_pc::<OpsRRP>();
+                        }
+                    }
+                    JLTS => self.cond_jmp::<u64>(Ordering::Less),
+                    JGTS => self.cond_jmp::<u64>(Ordering::Greater),
+                    JLTU => self.cond_jmp::<i64>(Ordering::Less),
+                    JGTU => self.cond_jmp::<i64>(Ordering::Greater),
+                    ECA => {
+                        // So we don't get timer interrupt after ECALL
+                        if TIMER_QUOTIENT != 0 {
+                            self.timer = self.timer.wrapping_add(1);
+                        }
+
+                        self.bump_pc::<OpsN>();
+                        return Ok(VmRunOk::Ecall);
+                    }
+                    EBP => {
+                        self.bump_pc::<OpsN>();
+                        return Ok(VmRunOk::Breakpoint);
+                    }
+                    FADD32 => self.binary_op::<f32>(ops::Add::add),
+                    FADD64 => self.binary_op::<f64>(ops::Add::add),
+                    FSUB32 => self.binary_op::<f32>(ops::Sub::sub),
+                    FSUB64 => self.binary_op::<f64>(ops::Sub::sub),
+                    FMUL32 => self.binary_op::<f32>(ops::Mul::mul),
+                    FMUL64 => self.binary_op::<f64>(ops::Mul::mul),
+                    FDIV32 => self.binary_op::<f32>(ops::Div::div),
+                    FDIV64 => self.binary_op::<f64>(ops::Div::div),
+                    FMA32 => self.fma::<f32>(),
+                    FMA64 => self.fma::<f64>(),
+                    FINV32 => handler!(self, |OpsRR(tg, reg)| self
+                        .write_reg(tg, 1. / self.read_reg(reg).cast::<f32>())),
+                    FINV64 => handler!(self, |OpsRR(tg, reg)| self
+                        .write_reg(tg, 1. / self.read_reg(reg).cast::<f64>())),
+                    FCMPLT32 => self.fcmp::<f32>(Ordering::Less),
+                    FCMPLT64 => self.fcmp::<f64>(Ordering::Less),
+                    FCMPGT32 => self.fcmp::<f32>(Ordering::Greater),
+                    FCMPGT64 => self.fcmp::<f64>(Ordering::Greater),
+                    ITF32 => handler!(self, |OpsRR(tg, reg)| self
+                        .write_reg(tg, self.read_reg(reg).cast::<i64>() as f32)),
+                    ITF64 => handler!(self, |OpsRR(tg, reg)| self
+                        .write_reg(tg, self.read_reg(reg).cast::<i64>() as f64)),
+                    FTI32 => handler!(self, |OpsRRB(tg, reg, mode)| self.write_reg(
+                        tg,
+                        crate::float::f32toint(
+                            self.read_reg(reg).cast::<f32>(),
+                            RoundingMode::try_from(mode)
+                                .map_err(|()| VmRunError::InvalidOperand)?,
+                        ),
+                    )),
+                    FTI64 => handler!(self, |OpsRRB(tg, reg, mode)| self.write_reg(
+                        tg,
+                        crate::float::f64toint(
+                            self.read_reg(reg).cast::<f64>(),
+                            RoundingMode::try_from(mode)
+                                .map_err(|()| VmRunError::InvalidOperand)?,
+                        ),
+                    )),
+                    FC32T64 => handler!(self, |OpsRR(tg, reg)| self
+                        .write_reg(tg, self.read_reg(reg).cast::<f32>() as f64)),
+                    FC64T32 => handler!(self, |OpsRRB(tg, reg, mode)| self.write_reg(
+                        tg,
+                        crate::float::conv64to32(
+                            self.read_reg(reg).cast(),
+                            RoundingMode::try_from(mode)
+                                .map_err(|()| VmRunError::InvalidOperand)?,
+                        ),
+                    )),
+                    LRA16 => handler!(self, |OpsRRP(tg, reg, imm)| self.write_reg(
+                        tg,
+                        (self.pc + self.read_reg(reg).cast::<u64>() + imm + 3_u16).get(),
+                    )),
+                    LDR16 => handler!(self, |OpsRRPH(dst, base, off, count)| self.load(
+                        dst,
+                        base,
+                        self.pcrel(off, 3).get(),
+                        count
+                    )?),
+                    STR16 => handler!(self, |OpsRRPH(dst, base, off, count)| self.store(
+                        dst,
+                        base,
+                        self.pcrel(off, 3).get(),
+                        count
+                    )?),
+                    JMP16 => {
+                        let OpsP(off) = self.decode();
+                        self.pc = self.pcrel(off, 1);
+                    }
+                    op => return Err(VmRunError::InvalidOpcode(op)),
+                }
+            }
+
+            if TIMER_QUOTIENT != 0 {
+                self.timer = self.timer.wrapping_add(1);
+                if self.timer % TIMER_QUOTIENT == 0 {
+                    return Ok(VmRunOk::Timer);
+                }
+            }
+        }
+    }
+
+    /// Bump instruction pointer
+    #[inline(always)]
+    fn bump_pc<T: Copy>(&mut self) {
+        self.pc = self
+            .pc
+            .wrapping_add(core::mem::size_of::<T>())
+            .wrapping_add(1);
+    }
+
+    /// Decode instruction operands
+    #[inline(always)]
+    unsafe fn decode<T: Copy>(&mut self) -> T {
+        unsafe { self.memory.prog_read::<T>(self.pc + 1_u64) }
+    }
+
+    /// Load
+    #[inline(always)]
+    unsafe fn load(
+        &mut self,
+        dst: u8,
+        base: u8,
+        offset: u64,
+        count: u16,
+    ) -> Result<(), VmRunError> {
+        let n: u8 = match dst {
+            0 => 1,
+            _ => 0,
+        };
+
+        unsafe {
+            self.memory.load(
+                self.ldst_addr_uber(dst, base, offset, count, n)?,
+                self.registers
+                    .as_mut_ptr()
+                    .add(usize::from(dst) + usize::from(n))
+                    .cast(),
+                usize::from(count).saturating_sub(n.into()),
+            )
+        }?;
+
+        Ok(())
+    }
+
+    /// Store
+    #[inline(always)]
+    unsafe fn store(
+        &mut self,
+        dst: u8,
+        base: u8,
+        offset: u64,
+        count: u16,
+    ) -> Result<(), VmRunError> {
+        unsafe {
+            self.memory.store(
+                self.ldst_addr_uber(dst, base, offset, count, 0)?,
+                self.registers.as_ptr().add(usize::from(dst)).cast(),
+                count.into(),
+            )
+        }?;
+        Ok(())
+    }
+
+    /// Three-way comparsion
+    #[inline(always)]
+    unsafe fn cmp<T: ValueVariant + Ord>(&mut self, to: u8, reg: u8, val: T) {
+        self.write_reg(to, self.read_reg(reg).cast::<T>().cmp(&val) as i64);
+    }
+
+    /// Perform binary operating over two registers
+    #[inline(always)]
+    unsafe fn binary_op<T: ValueVariant>(&mut self, op: impl Fn(T, T) -> T) {
+        let OpsRRR(tg, a0, a1) = unsafe { self.decode() };
+        self.write_reg(
+            tg,
+            op(self.read_reg(a0).cast::<T>(), self.read_reg(a1).cast::<T>()),
+        );
+        self.bump_pc::<OpsRRR>();
+    }
+
+    /// Perform binary operation over register and immediate
+    #[inline(always)]
+    unsafe fn binary_op_imm<T: ValueVariant>(&mut self, op: impl Fn(T, T) -> T) {
+        #[derive(Clone, Copy)]
+        #[repr(packed)]
+        struct OpsRRImm<I>(OpsRR, I);
+
+        let OpsRRImm::<T>(OpsRR(tg, reg), imm) = unsafe { self.decode() };
+        self.write_reg(tg, op(self.read_reg(reg).cast::<T>(), imm));
+        self.bump_pc::<OpsRRImm<T>>();
+    }
+
+    /// Perform binary operation over register and shift immediate
+    #[inline(always)]
+    unsafe fn binary_op_shift<T: ValueVariant>(&mut self, op: impl Fn(T, u32) -> T) {
+        let OpsRRR(tg, a0, a1) = unsafe { self.decode() };
+        self.write_reg(
+            tg,
+            op(
+                self.read_reg(a0).cast::<T>(),
+                self.read_reg(a1).cast::<u32>(),
+            ),
+        );
+        self.bump_pc::<OpsRRR>();
+    }
+
+    /// Perform binary operation over register and shift immediate
+    #[inline(always)]
+    unsafe fn binary_op_ims<T: ValueVariant>(&mut self, op: impl Fn(T, u32) -> T) {
+        let OpsRRB(tg, reg, imm) = unsafe { self.decode() };
+        self.write_reg(tg, op(self.read_reg(reg).cast::<T>(), imm.into()));
+        self.bump_pc::<OpsRRB>();
+    }
+
+    /// Fused division-remainder
+    #[inline(always)]
+    unsafe fn dir<T: ValueVariant + CheckedDivRem>(&mut self) {
+        handler!(self, |OpsRRRR(td, tr, a0, a1)| {
+            let a0 = self.read_reg(a0).cast::<T>();
+            let a1 = self.read_reg(a1).cast::<T>();
+
+            if let Some(div) = a0.checked_div(a1) {
+                self.write_reg(td, div);
+            } else {
+                self.write_reg(td, -1_i64);
+            }
+
+            if let Some(rem) = a0.checked_rem(a1) {
+                self.write_reg(tr, rem);
+            } else {
+                self.write_reg(tr, a0);
+            }
+        });
+    }
+
+    /// Fused multiply-add
+    #[inline(always)]
+    unsafe fn fma<T>(&mut self)
+    where
+        T: ValueVariant + core::ops::Mul<Output = T> + core::ops::Add<Output = T>,
+    {
+        handler!(self, |OpsRRRR(tg, a0, a1, a2)| {
+            let a0 = self.read_reg(a0).cast::<T>();
+            let a1 = self.read_reg(a1).cast::<T>();
+            let a2 = self.read_reg(a2).cast::<T>();
+            self.write_reg(tg, a0 * a1 + a2)
+        });
+    }
+
+    /// Float comparsion
+    #[inline(always)]
+    unsafe fn fcmp<T: PartialOrd + ValueVariant>(&mut self, nan: Ordering) {
+        handler!(self, |OpsRRR(tg, a0, a1)| {
+            let a0 = self.read_reg(a0).cast::<T>();
+            let a1 = self.read_reg(a1).cast::<T>();
+            self.write_reg(tg, (a0.partial_cmp(&a1).unwrap_or(nan) as i8 + 1) as u8)
+        });
+    }
+
+    /// Calculate pc-relative address
+    #[inline(always)]
+    fn pcrel(&self, offset: impl AddressOp, pos: u8) -> Address {
+        self.pc.wrapping_add(pos).wrapping_add(offset)
+    }
+
+    /// Jump at `PC + #3` if ordering on `#0 <=> #1` is equal to expected
+    #[inline(always)]
+    unsafe fn cond_jmp<T: ValueVariant + Ord>(&mut self, expected: Ordering) {
+        let OpsRRP(a0, a1, ja) = unsafe { self.decode() };
+        if self
+            .read_reg(a0)
+            .cast::<T>()
+            .cmp(&self.read_reg(a1).cast::<T>())
+            == expected
+        {
+            self.pc = self.pcrel(ja, 3);
+        } else {
+            self.bump_pc::<OpsRRP>();
+        }
+    }
+
+    /// Load / Store Address check-computation überfunction
+    #[inline(always)]
+    unsafe fn ldst_addr_uber(
+        &self,
+        dst: u8,
+        base: u8,
+        offset: u64,
+        size: u16,
+        adder: u8,
+    ) -> Result<Address, VmRunError> {
+        let reg = dst.checked_add(adder).ok_or(VmRunError::RegOutOfBounds)?;
+
+        if usize::from(reg) * 8 + usize::from(size) > 2048 {
+            Err(VmRunError::RegOutOfBounds)
+        } else {
+            self.read_reg(base)
+                .cast::<u64>()
+                .checked_add(offset)
+                .and_then(|x| x.checked_add(adder.into()))
+                .ok_or(VmRunError::AddrOutOfBounds)
+                .map(Address::new)
+        }
+    }
+}

hbvm_aos_on_linux/Cargo.toml

@@ -0,0 +1,9 @@
+[package]
+name = "hbvm_aos_on_linux"
+version = "0.1.0"
+edition = "2021"
+default-run = "hbvm_aos_on_linux"
+
+[dependencies]
+hbvm.path = "../hbvm"
+nix = { version = "0.27", features = ["mman", "signal"] }

hbvm_aos_on_linux/readme.md

@@ -0,0 +1,3 @@
+As close to the AbleOS runtime as possible
+
+useful for me to spec out things on my laptop

hbvm_aos_on_linux/src/main.rs

@@ -0,0 +1,96 @@
+//! Holey Bytes Experimental Runtime
+mod mem;
+
+use {
+    hbvm::{mem::Address, Vm, VmRunOk},
+    nix::sys::mman::{mmap, MapFlags, ProtFlags},
+    std::{env::args, fs::File, num::NonZeroUsize, process::exit},
+};
+
+fn main() -> Result<(), Box<dyn std::error::Error>> {
+    eprintln!("== HB×RT (Holey Bytes Linux Runtime) v0.1 ==");
+    eprintln!("[W] Currently supporting only flat images");
+
+    let Some(image_path) = args().nth(1) else {
+        eprintln!("[E] Missing image path");
+        exit(1);
+    };
+
+    // Load program
+    eprintln!("[I] Loading image from \"{image_path}\"");
+    let file = File::open(image_path)?;
+    let ptr = unsafe {
+        mmap(
+            None,
+            NonZeroUsize::new(file.metadata()?.len() as usize).ok_or("File is empty")?,
+            ProtFlags::PROT_READ,
+            MapFlags::MAP_PRIVATE,
+            Some(&file),
+            0,
+        )?
+    };
+
+    eprintln!("[I] Image loaded at {ptr:p}");
+
+    // Execute program
+    let mut vm = unsafe { Vm::<_, 0>::new(mem::HostMemory, Address::new(ptr as u64)) };
+
+    // Memory access fault handling
+    unsafe {
+        use nix::sys::signal;
+
+        extern "C" fn action(
+            _: std::ffi::c_int,
+            info: *mut nix::libc::siginfo_t,
+            _: *mut std::ffi::c_void,
+        ) {
+            unsafe {
+                eprintln!("[E] Memory access fault at {:p}", (*info).si_addr());
+            }
+        }
+
+        signal::sigaction(
+            signal::Signal::SIGSEGV,
+            &nix::sys::signal::SigAction::new(
+                signal::SigHandler::SigAction(action),
+                signal::SaFlags::SA_NODEFER,
+                nix::sys::signalfd::SigSet::empty(),
+            ),
+        )?;
+    }
+
+    let stat = loop {
+        match vm.run() {
+            Ok(VmRunOk::Breakpoint) => eprintln!(
+                "[I] Hit breakpoint\nIP: {}\n== Registers ==\n{:?}",
+                vm.pc, vm.registers
+            ),
+            Ok(VmRunOk::Timer) => (),
+            Ok(VmRunOk::Ecall) => {
+
+                //     unsafe {
+                //         std::arch::asm!(
+                //             "syscall",
+                //             inlateout("rax") vm.registers[1].0,
+                //             in("rdi") vm.registers[2].0,
+                //             in("rsi") vm.registers[3].0,
+                //             in("rdx") vm.registers[4].0,
+                //             in("r10") vm.registers[5].0,
+                //             in("r8")  vm.registers[6].0,
+                //             in("r9")  vm.registers[7].0,
+                //         )
+                //     }
+            }
+            Ok(VmRunOk::End) => break Ok(()),
+            Err(e) => break Err(e),
+        }
+    };
+
+    eprintln!("\n== Registers ==\n{:?}", vm.registers);
+    if let Err(e) = stat {
+        eprintln!("\n[E] Runtime error: {e:?}");
+        exit(2);
+    }
+
+    Ok(())
+}

hbvm_aos_on_linux/src/mem.rs

@@ -0,0 +1,31 @@
+use hbvm::mem::{Address, LoadError, Memory, StoreError};
+
+pub struct HostMemory;
+impl Memory for HostMemory {
+    #[inline]
+    unsafe fn load(
+        &mut self,
+        addr: Address,
+        target: *mut u8,
+        count: usize,
+    ) -> Result<(), LoadError> {
+        unsafe { core::ptr::copy(addr.get() as *const u8, target, count) }
+        Ok(())
+    }
+
+    #[inline]
+    unsafe fn store(
+        &mut self,
+        addr: Address,
+        source: *const u8,
+        count: usize,
+    ) -> Result<(), StoreError> {
+        unsafe { core::ptr::copy(source, addr.get() as *mut u8, count) }
+        Ok(())
+    }
+
+    #[inline]
+    unsafe fn prog_read<T: Copy>(&mut self, addr: Address) -> T {
+        core::ptr::read(addr.get() as *const T)
+    }
+}

hbxrt/Cargo.toml

@@ -0,0 +1,9 @@
+[package]
+name        = "hbxrt"
+version     = "0.1.0"
+edition     = "2021"
+default-run = "hbxrt"
+
+[dependencies]
+hbvm.path = "../hbvm"
+memmap2   = "0.9"

hbxrt/src/main.rs

@@ -0,0 +1,94 @@
+//! Holey Bytes Experimental Runtime
+
+#![deny(unsafe_op_in_unsafe_fn)]
+
+mod mem;
+
+use {
+    hbvm::{mem::Address, Vm, VmRunOk},
+    memmap2::Mmap,
+    std::{env::args, fs::File, process::exit},
+};
+
+fn main() -> Result<(), Box<dyn std::error::Error>> {
+    eprintln!("== HB×RT (Holey Bytes Experimental Runtime) v0.1 ==");
+    eprintln!("[W] Currently supporting only flat images");
+
+    if !hbvm::FL_ARCH_SPECIFIC_SUPPORTED {
+        eprintln!(
+            "\
+                [W] Architecture not fully supported!\n    \
+                    FTI32, FTI64 will yield {:#x}\n    \
+                    FC64T32 will yield NaN\
+            ",
+            i64::MAX,
+        )
+    }
+
+    let mut args = args().skip(1);
+    let Some(image_path) = args.next() else {
+        eprintln!("[E] Missing image path");
+        exit(1);
+    };
+
+    let dsls = args.next().as_deref() == Some("-L");
+    if cfg!(not(target_os = "linux")) && dsls {
+        eprintln!("[E] Unsupported platform for Direct Linux syscall mode");
+        exit(1);
+    }
+
+    if dsls {
+        eprintln!("[I] Direct Linux syscall mode activated")
+    }
+
+    // Allocate stack
+    let mut stack = unsafe { mem::alloc_stack() };
+    eprintln!("[I] Stack allocated at {:p}", stack.as_ptr());
+
+    // Load program
+    eprintln!("[I] Loading image from \"{image_path}\"");
+    let file_handle = File::open(image_path)?;
+    let mmap = unsafe { Mmap::map(&file_handle) }?;
+
+    eprintln!("[I] Image loaded at {:p}", mmap.as_ptr());
+
+    let mut vm = unsafe { Vm::<_, 0>::new(mem::HostMemory, Address::new(mmap.as_ptr() as u64)) };
+    vm.write_reg(254, stack.as_mut_ptr() as u64);
+
+    // Execute program
+    let stat = loop {
+        match vm.run() {
+            Ok(VmRunOk::Breakpoint) => eprintln!(
+                "[I] Hit breakpoint\nIP: {}\n== Registers ==\n{:?}",
+                vm.pc, vm.registers
+            ),
+            Ok(VmRunOk::Timer) => (),
+            Ok(VmRunOk::Ecall) if dsls => unsafe {
+                std::arch::asm!(
+                    "syscall",
+                    inlateout("rax") vm.registers[1].0,
+                    in("rdi") vm.registers[2].0,
+                    in("rsi") vm.registers[3].0,
+                    in("rdx") vm.registers[4].0,
+                    in("r10") vm.registers[5].0,
+                    in("r8")  vm.registers[6].0,
+                    in("r9")  vm.registers[7].0,
+                )
+            },
+            Ok(VmRunOk::Ecall) => {
+                eprintln!("[E] General environment calls not supported");
+                exit(1);
+            }
+            Ok(VmRunOk::End) => break Ok(()),
+            Err(e) => break Err(e),
+        }
+    };
+
+    eprintln!("\n== Registers ==\n{:?}", vm.registers);
+    if let Err(e) = stat {
+        eprintln!("\n[E] Runtime error: {e:?}");
+        exit(2);
+    }
+
+    Ok(())
+}

hbxrt/src/mem.rs

@@ -0,0 +1,47 @@
+use std::alloc::Layout;
+
+use hbvm::mem::{Address, LoadError, Memory, StoreError};
+
+pub struct HostMemory;
+impl Memory for HostMemory {
+    #[inline]
+    unsafe fn load(
+        &mut self,
+        addr: Address,
+        target: *mut u8,
+        count: usize,
+    ) -> Result<(), LoadError> {
+        unsafe { core::ptr::copy(addr.get() as *const u8, target, count) }
+        Ok(())
+    }
+
+    #[inline]
+    unsafe fn store(
+        &mut self,
+        addr: Address,
+        source: *const u8,
+        count: usize,
+    ) -> Result<(), StoreError> {
+        unsafe { core::ptr::copy(source, addr.get() as *mut u8, count) }
+        Ok(())
+    }
+
+    #[inline]
+    unsafe fn prog_read<T: Copy>(&mut self, addr: Address) -> T {
+        unsafe { core::ptr::read(addr.get() as *const T) }
+    }
+}
+
+const STACK_SIZE: usize = 2; // MiB
+type Stack = [u8; 1024 * 1024 * STACK_SIZE];
+
+/// Allocate stack of size [`STACK_SIZE`] MiB
+pub unsafe fn alloc_stack() -> Box<Stack> {
+    let layout = Layout::new::<Stack>();
+    let ptr = unsafe { std::alloc::alloc(layout) };
+    if ptr.is_null() {
+        std::alloc::handle_alloc_error(layout);
+    }
+
+    unsafe { Box::from_raw(ptr.cast()) }
+}

rustfmt.toml

@@ -1,3 +1,4 @@
 hex_literal_case = "Upper"
 imports_granularity = "One"
-struct_field_align_threshold = 5
+struct_field_align_threshold = 8
+enum_discrim_align_threshold = 8

spec.md

@@ -0,0 +1,505 @@
+# HoleyBytes ISA Specification
+
+# Bytecode format
+- Image format is not specified, though ELF is recommended
+- All numbers are encoded little-endian
+- There is 256 registers, they are represented by a byte
+- Immediate values are 8, 16, 32 or 64 bit
+
+## Instruction encoding
+- Instruction operands are packed (no alignment)
+- [opcode, operand 0, operand 1, …]
+
+## Instruction parameter types
+- `R`: Register (8 bits)
+- Relative program-counter offset immediates:
+    - `O`: 32 bit (Si32)
+    - `P`: 16 bit (Si16)
+- Immediates:
+    - `B`: Byte, 8 bit (Xi8)
+    - `H`: Half-word, 16 bit (Xi16)
+    - `W`: Word, 32 bit (Xi32)
+    - `D`: Double-word, 64 bit (Xi64)
+- `A`: Absolute address immediate, 64 bit (Ui64)
+
+## Types
+- Si*n*: Signed integer of size *n* bits (Si8, Si16, Si32, Si64)
+- Ui*n*: Unsigned integer of size *n* bits (Ui8, Ui16, Ui32, Ui64)
+- Xi*n*: Sign-agnostic integer of size *n* bits (Xi8, Xi16, Xi32, Xi64)
+- Fl*n*: Floating point number of size *n* bits (Fl32, Fl64)
+
+# Behaviour
+- There is only one type of register, a general-purpose one.
+  Used for both integers and floats.
+- Integer operations are wrapping, including signed numbers
+    - Bitshifts are truncating
+- Two's complement
+- Floats as specified by IEEE 754
+- Execution model is implementation defined as long all observable
+  effects are performed in correct order
+
+## Relative addressing
+Relative addresses are computed from address of the first byte
+of offset in the code. Not from the beginning of current or following instruction.
+
+## Zero register
+- Register 0
+- Cannot be clobbered
+    - Write is no-op
+- Load always yields 0
+
+## Rounding modes
+| Rounding mode            | Value |
+|:-------------------------|:------|
+| To nearest, ties to even | 0b00  |
+| Towards 0 (truncate)     | 0b01  |
+| Towards +∞ (up)          | 0b10  |
+| Towards -∞ (down)        | 0b11  |
+
+- Remaining values in the byte traps with invalid operand exception
+
+# Memory
+- Memory implementation is implementation-defined
+- Zero address (`0x0`) is considered invalid
+
+# Traps
+- Environment call
+- Environment breakpoint
+
+Program counter goes to the following instruction
+
+## Exceptions
+- Memory access fault
+- Invalid operand
+- Unknown opcode
+
+Program counter stays on the currently executed instruction
+
+# Instructions
+- `#n`: register in parameter *n*
+- `$n`: for immediate in parameter *n*
+- `#P ← V`: Set register P to value V
+- `[x]`: Address x
+- `XY`: X bytes from location Y
+- `pc`: Program counter
+- `<XYZ>`: Placeholder
+- `Type(X)`: Cast
+
+## Program execution control
+- Type `N`
+
+| Opcode | Mnemonic | Action                                      |
+|:-------|:---------|:--------------------------------------------|
+| 0x00   | UN       | Throw unreachable code exception            |
+| 0x01   | TX       | Terminate execution (eg. on end of program) |
+| 0x02   | NOP      | Do nothing                                  |
+
+## Binary register-register ops
+- Type `RRR`
+- Action: `#0 ← #1 <OP> #2`
+
+## Addition (`+`)
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x03   | ADD8     | Xi8  |
+| 0x04   | ADD16    | Xi16 |
+| 0x05   | ADD32    | Xi32 |
+| 0x06   | ADD64    | Xi64 |
+
+## Subtraction (`-`)
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x07   | SUB8     | Xi8  |
+| 0x08   | SUB16    | Xi16 |
+| 0x09   | SUB32    | Xi32 |
+| 0x0A   | SUB64    | Xi64 |
+
+## Multiplication (`*`)
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x0B   | MUL8     | Xi8  |
+| 0x0C   | MUL16    | Xi16 |
+| 0x0D   | MUL32    | Xi32 |
+| 0x0E   | MUL64    | Xi64 |
+
+## Bitwise ops (type: Xi64)
+| Opcode | Mnemonic | Operation           |
+|:-------|:---------|:--------------------|
+| 0x0F   | AND      | Conjunction (&)     |
+| 0x10   | OR       | Disjunction (\|)    |
+| 0x11   | XOR      | Non-equivalence (^) |
+
+## Unsigned left bitshift (`<<`)
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x12   | SLU8     | Ui8  |
+| 0x13   | SLU16    | Ui16 |
+| 0x14   | SLU32    | Ui32 |
+| 0x15   | SLU64    | Ui64 |
+
+## Unsigned right bitshift (`>>`)
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x16   | SRU8     | Ui8  |
+| 0x17   | SRU16    | Ui16 |
+| 0x18   | SRU32    | Ui32 |
+| 0x19   | SRU64    | Ui64 |
+
+## Signed right bitshift (`>>`)
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x1A   | SRS8     | Si8  |
+| 0x1B   | SRS16    | Si16 |
+| 0x1C   | SRS32    | Si32 |
+| 0x1D   | SRS64    | Si64 |
+
+## Comparsion
+- Compares two numbers, saves result to register
+- Operation: `#0 ← #1 <=> #2`
+
+| Ordering | Number |
+|:---------|:-------|
+| <        | -1     |
+| =        | 0      |
+| >        | 1      |
+
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x1E   | CMPU     | Ui64 |
+| 0x1F   | CMPS     | Si64 |
+
+# Merged divide-remainder
+- Type `RRRR`
+- Operation:
+    - `#0 ← #2 / #3`
+    - `#1 ← #2 % #3`
+
+- If dividing by zero:
+    - `#0 ← Ui64(-1)`
+    - `#1 ← #2`
+
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x20   | DIRU8    | Ui8  |
+| 0x21   | DIRU16   | Ui16 |
+| 0x22   | DIRU32   | Ui32 |
+| 0x23   | DIRU64   | Ui64 |
+| 0x24   | DIRS8    | Si8  |
+| 0x25   | DIRS16   | Si16 |
+| 0x26   | DIRS32   | Si32 |
+| 0x27   | DIRS64   | Si64 |
+
+# Unary register operations (type: Xi64)
+- Type: `RR`
+- Operation: `#0 ← <OP> #1`
+
+| Opcode | Mnemonic | Operation                |
+|:-------|:---------|:-------------------------|
+| 0x28   | NEG      | Bitwise complement (`~`) |
+| 0x29   | NOT      | Logical negation (`!`)   |
+
+## Sign extensions
+- Operation: `#0 ← Si64(#1)`
+
+| Opcode | Mnemonic | Source type |
+|:-------|:---------|:------------|
+| 0x2A   | SXT8     | Si8         |
+| 0x2B   | SXT16    | Si16        |
+| 0x2C   | SXT32    | Si32        |
+
+# Binary register-immediate operations
+- Type: `RR<IMM>`
+- Operation: `#0 ← #1 <OP> $2`
+
+## Addition (`+`)
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x2D   | ADDI8    | Xi8  |
+| 0x2E   | ADDI16   | Xi16 |
+| 0x2F   | ADDI32   | Xi32 |
+| 0x30   | ADDI64   | Xi64 |
+
+## Multiplication (`*`)
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x31   | MULI8    | Xi8  |
+| 0x32   | MULI16   | Xi16 |
+| 0x33   | MULI32   | Xi32 |
+| 0x34   | MULI64   | Xi64 |
+
+## Bitwise ops (type: Xi64)
+| Opcode | Mnemonic | Operation           |
+|:-------|:---------|:--------------------|
+| 0x35   | ANDI     | Conjunction (&)     |
+| 0x36   | ORI      | Disjunction (\|)    |
+| 0x37   | XORI     | Non-equivalence (^) |
+
+# Register-immediate bitshifts
+- Type: `RRB`
+- Operation: `#0 ← #1 <OP> $2`
+
+## Unsigned left bitshift (`<<`)
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x38   | SLUI8    | Ui8  |
+| 0x39   | SLUI16   | Ui16 |
+| 0x3A   | SLUI32   | Ui32 |
+| 0x3B   | SLUI64   | Ui64 |
+
+## Unsigned right bitshift (`>>`)
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x3C   | SRUI8    | Ui8  |
+| 0x3D   | SRUI16   | Ui16 |
+| 0x3E   | SRUI32   | Ui32 |
+| 0x3F   | SRUI64   | Ui64 |
+
+## Signed right bitshift (`>>`)
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x40   | SRSI8    | Si8  |
+| 0x41   | SRSI16   | Si16 |
+| 0x42   | SRSI32   | Si32 |
+| 0x43   | SRSI64   | Si64 |
+
+## Comparsion
+- Compares two numbers, saves result to register
+- Operation: `#0 ← #1 <=> $2`
+- Comparsion table same for register-register one
+
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x44   | CMPUI    | Ui64 |
+| 0x45   | CMPSI    | Si64 |
+
+# Register copies
+- Type: `RR`
+
+| Opcode | Mnemonic | Operation                        |
+|:-------|:---------|:---------------------------------|
+| 0x46   | CP       | Copy register value (`#0 ← #1`)  |
+| 0x47   | SWA      | Swap register values (`#0 ⇆ #1`) |
+
+# Load immediate
+- Load immediate value from code to register
+- Type: `R<IMM>`
+- Operation: `#0 ← $1`
+
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x48   | LI8      | Xi8  |
+| 0x49   | LI16     | Xi16 |
+| 0x4A   | Li32     | Xi32 |
+| 0x4B   | Li64     | Xi64 |
+
+# Load relative address
+- Compute value from program counter, register value and offset
+- Type: `RRO`
+- Operation: `#0 ← pc + #1 + $2`
+
+| Opcode | Mnemonic |
+|:-------|:---------|
+| 0x4C   | LRA      |
+
+# Memory access operations
+- Immediate `$3` specifies size
+- If size is greater than register size,
+    it overflows to adjecent register
+    (eg. copying 16 bytes to register `r1` copies first 8 bytes to it
+         and the remaining to `r2`)
+
+## Absolute addressing
+- Type: `RRAH`
+- Computes address from base register and absolute offset
+
+| Opcode | Mnemonic | Operation          |
+|:-------|:---------|:-------------------|
+| 0x4D   | LD       | `#0 ← $3[#1 + $2]` |
+| 0x4E   | ST       | `$3[#1 + $2] ← #0` |
+
+## Relative addressing
+- Type: `RROH`
+- Computes address from register and offset from program counter
+
+| Opcode | Mnemonic | Operation               |
+|:-------|:---------|:------------------------|
+| 0x4F   | LDR      | `#0 ← $3[pc + #1 + $2]` |
+| 0x50   | STR      | `$3[pc + #1 + $2] ← #0` |
+
+# Block memory copy
+- Type: `RRH`
+- Copies block of `$3` bytes from memory location on address on `#0` to `#1`
+
+| Opcode | Mnemonic | Operation         |
+|:-------|:---------|:------------------|
+| 0x51   | BMC      | `$3[#1] ← $3[x0]` |
+
+# Block register copy
+- Type: `RRB`
+- Copy block of `$3` registers starting with `#0` to `#1`
+- Copying over the 256 registers causes an exception
+
+| Opcode | Mnemonic | Operation     |
+|:-------|:---------|:--------------|
+| 0x52   | BRC      | `$3#1 ← $3#0` |
+
+# Relative jump
+- Type: `O`
+
+| Opcode | Mnemonic | Operation      |
+|:-------|:---------|:---------------|
+| 0x53   | JMP      | `pc ← pc + $0` |
+
+# Linking jump
+- Operation:
+    - Save address of following instruction to `#0`
+        - `#0 ← pc+<instruction size>`
+    - Jump to specified address 
+
+| Opcode | Mnemonic | Instruction type  | Address                  |
+|:-------|:---------|:------------------|:-------------------------|
+| 0x54   | JAL      | RRO (size = 6 B)  | Relative, `pc + #1 + $2` |
+| 0x55   | JALA     | RRA (size = 10 B) | Absolute, `#1 + $2`      |
+
+# Conditional jump
+- Perform comparsion, if operation met, jump to relative address
+- Type: `RRP`
+- Operation: `if #0 <CMP> #1 { pc ← pc + $2 }`
+
+| Opcode | Mnemonic | Condition          | Type |
+|:-------|:---------|:-------------------|:-----|
+| 0x56   | JEQ      | Equals (`=`)       | Xi64 |
+| 0x57   | JNE      | Not-equals (`≠`)   | Xi64 |
+| 0x58   | JLTU     | Less-than (`<`)    | Ui64 |
+| 0x59   | JGTU     | Greater-than (`>`) | Ui64 |
+| 0x5A   | JLTS     | Less-than (`<`)    | Si64 |
+| 0x5B   | JGTS     | Greater-than (`>`) | Si64 |
+
+# Environment traps
+- Traps to the environment
+- Type: `N`
+
+| Opcode | Mnemonic | Trap type        |
+|:-------|:---------|:-----------------|
+| 0x5C   | ECA      | Environment call |
+| 0x5D   | EBP      | Breakpoint       |
+
+# Floating point binary operations
+- Type: `RRR`
+- Operation: `#0 ← #1 <OP> #2`
+
+| Opcode | Mnemonic | Operation            | Type |
+|:-------|:---------|:---------------------|:-----|
+| 0x5E   | FADD32   | Addition (`+`)       | Fl32 |
+| 0x5F   | FADD64   | Addition (`+`)       | Fl64 |
+| 0x60   | FSUB32   | Subtraction (`-`)    | Fl32 |
+| 0x61   | FSUB64   | Subtraction (`-`)    | Fl64 |
+| 0x62   | FMUL32   | Multiplication (`*`) | Fl32 |
+| 0x63   | FMUL64   | Multiplication (`*`) | Fl64 |
+| 0x64   | FDIV32   | Division (`/`)       | Fl32 |
+| 0x65   | FDIV64   | Division (`/`)       | Fl64 |
+
+# Fused multiply-add
+- Type: `RRRR`
+- Operation: `#0 ← (#1 * #2) + #3`
+
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x66   | FMA32    | Fl32 |
+| 0x67   | FMA64    | Fl64 |
+
+# Comparsions
+- Type: `RRR`
+- Operation: `#0 ← #1 <=> #2`
+- Comparsion table same as for `CMPx`/`CMPxI`
+- NaN is less-than/greater-than depends on variant
+
+| Opcode | Mnemonic | Type | NaN is |
+|:-------|:---------|:-----|:-------|
+| 0x6A   | FCMPLT32 | Fl32 | <      |
+| 0x6B   | FCMPLT64 | Fl64 | <      |
+| 0x6C   | FCMPGT32 | Fl32 | >      |
+| 0x6D   | FCMPGT64 | Fl64 | >      |
+
+# Int to float
+- Type: `RR`
+- Converts from `Si64`
+- Operation: `#0 ← Fl<SIZE>(#1)`
+
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x6E   | ITF32    | Fl32 |
+| 0x6F   | ITF64    | Fl64 |
+
+# Float to int
+- Type: `RRB`
+- Operation: `#0 ← Si64(#1)`
+- Immediate `$2` specifies rounding mode
+
+| Opcode | Mnemonic | Type |
+|:-------|:---------|:-----|
+| 0x70   | FTI32    | Fl32 |
+| 0x71   | FTI64    | Fl64 |
+
+# Fl32 to Fl64
+- Type: `RR`
+- Operation: `#0 ← Fl64(#1)`
+
+| Opcode | Mnemonic |
+|:-------|:---------|
+| 0x72   | FC32T64  |
+
+# Fl64 to Fl32
+- Type: `RRB`
+- Operation: `#0 ← Fl32(#1)`
+- Immediate `$2` specified rounding mode
+
+| Opcode | Mnemonic |
+|:-------|:---------|
+| 0x73   | FC64T32  |
+
+# 16-bit relative address instruction variants
+
+| Opcode | Mnemonic | Type | Variant of |
+|:-------|:---------|:-----|:-----------|
+| 0x74   | LRA16    | RRP  | LRA        |
+| 0x75   | LDR16    | RRPH | LDR        |
+| 0x76   | STR16    | RRPH | STR        |
+| 0x77   | JMP16    | P    | JMP        |
+
+# psABI
+## C datatypes and alignment
+- One byte is 8 bits
+
+| C Type      | Description              | Byte sizes |
+|:------------|:-------------------------|:-----------|
+| char        | Character / byte         | 1          |
+| short       | Short integer            | 2          |
+| int         | Integer                  | 4          |
+| long        | Long integer             | 8          |
+| long long   | Long long integer        | 8          |
+| float       | Single-precision float   | 4          |
+| double      | Double-precision float   | 8          |
+| long double | Extended-precision float | 8          |
+
+- Bikeshedding note: `long double` is now 8 bytes as
+  the base ISA does not support `f128`. an extension
+  for that should be made.
+
+## Call convention
+- Registers r1 – r31 are caller saved
+- Registers r32 – r255 are callee saved
+
+| Register   | Description         | Saver  |
+|:-----------|:--------------------|:-------|
+| r0         | Hard-wired zero     | N/A    |
+| r1 - r2    | Return values       | Caller |
+| r2 - r11   | Function parameters | Caller |
+| r12 - r30  | General purpose     | Caller |
+| r31        | Return address      | Caller |
+| r32 - r253 | General purpose     | Callee |
+| r254       | Stack pointer       | Callee |
+| r255       | Thread pointer      | N/A    |
+
+- If return value is too big to fit r1, r2 is also used.
+- Values larger than two double-words are passed by reference

xtask/Cargo.toml

@@ -0,0 +1,9 @@
+[package]
+name    = "xtask"
+version = "0.1.0"
+edition = "2021"
+
+[dependencies]
+argh       = "0.1"
+color-eyre = "0.6"
+once_cell  = "1.18"

xtask/src/fmt.rs

@@ -0,0 +1,93 @@
+use {
+    crate::{utils::IterExt, ROOT},
+    argh::FromArgs,
+    color_eyre::{eyre::eyre, Result},
+    std::{
+        fs::File,
+        io::{BufRead, BufReader, BufWriter, Seek, Write},
+    },
+};
+
+/// Format `instructions.in`
+#[derive(Debug, FromArgs, PartialEq)]
+#[argh(subcommand, name = "fmt")]
+pub struct Command {
+    /// renumber instructions in their definition order
+    #[argh(switch, short = 'r')]
+    renumber: bool,
+}
+
+pub fn command(args: Command) -> Result<()> {
+    let mut file = File::options()
+        .read(true)
+        .write(true)
+        .open(ROOT.join("hbbytecode/instructions.in"))?;
+
+    // Extract records
+    let reader = BufReader::new(&file);
+    let mut recs = vec![];
+    let mut lens = [0_usize; 4];
+
+    for rec in reader.split(b';').filter_map(|r| {
+        r.map(|ln| {
+            let s = String::from_utf8_lossy(&ln);
+            let s = s.trim_matches('\n');
+            if s.is_empty() {
+                return None;
+            }
+
+            s.split(',')
+                .map(|s| Box::<str>::from(s.trim()))
+                .collect_array::<4>()
+                .map(Ok::<_, ()>)
+        })
+        .transpose()
+    }) {
+        let rec = rec?.map_err(|_| eyre!("Invalid record format"))?;
+        for (current, next) in lens.iter_mut().zip(rec.iter()) {
+            *current = (*current).max(next.len());
+        }
+
+        recs.push(rec);
+    }
+
+    // Clear file!
+    file.set_len(0)?;
+    file.seek(std::io::SeekFrom::Start(0))?;
+
+    let mut writer = BufWriter::new(file);
+
+    let ord_opco_len = digit_count(recs.len()) as usize;
+    for (n, rec) in recs.iter().enumerate() {
+        // Write opcode number
+        if args.renumber {
+            let n = format!("{n:#04X}");
+            write!(writer, "{n}, {}", padding(ord_opco_len, &n))?;
+        } else {
+            write!(writer, "{}, {}", rec[0], padding(lens[0], &rec[0]))?;
+        }
+
+        // Write other fields
+        writeln!(
+            writer,
+            "{}, {}{},{} {}{};",
+            rec[1],
+            padding(lens[1], &rec[1]),
+            rec[2],
+            padding(lens[2], &rec[2]),
+            rec[3],
+            padding(lens[3], &rec[3]),
+        )?;
+    }
+
+    Ok(())
+}
+
+fn padding(req: usize, s: &str) -> Box<str> {
+    " ".repeat(req.saturating_sub(s.len())).into()
+}
+
+#[inline]
+fn digit_count(n: usize) -> u32 {
+    n.checked_ilog10().unwrap_or(0) + 1
+}

xtask/src/main.rs

@@ -0,0 +1,25 @@
+mod fmt;
+mod utils;
+
+use {argh::FromArgs, color_eyre::Result, once_cell::sync::Lazy, std::path::Path};
+
+static ROOT: Lazy<&Path> = Lazy::new(|| Path::new(env!("CARGO_MANIFEST_DIR")).parent().unwrap());
+
+/// xTask for Holey Bytes project
+#[derive(FromArgs)]
+struct Command {
+    #[argh(subcommand)]
+    subcom: Subcommands,
+}
+
+#[derive(FromArgs)]
+#[argh(subcommand)]
+enum Subcommands {
+    Format(fmt::Command),
+}
+
+fn main() -> Result<()> {
+    match argh::from_env::<Command>().subcom {
+        Subcommands::Format(com) => fmt::command(com),
+    }
+}

xtask/src/utils.rs

@@ -0,0 +1,19 @@
+use std::mem::MaybeUninit;
+
+pub trait IterExt: Iterator {
+    fn collect_array<const N: usize>(&mut self) -> Option<[Self::Item; N]>
+    where
+        Self: Sized,
+    {
+        let mut array: [MaybeUninit<Self::Item>; N] =
+            unsafe { MaybeUninit::uninit().assume_init() };
+
+        for item in &mut array {
+            item.write(self.next()?);
+        }
+
+        Some(array.map(|item| unsafe { item.assume_init() }))
+    }
+}
+
+impl<T: Iterator> IterExt for T {}