WIP linux multiarch support

Changing bonnibel to respect the --arch flag to configure. This requires
some reworking of modules, mostly in the addition of the ModuleList
class instead of just a dict of modules.

.clang-format

@@ -1,21 +0,0 @@
----
-Language:        Cpp
-BasedOnStyle:    LLVM
-
-ColumnLimit:     100
-IndentWidth:     4
-TabWidth:        4
-UseTab:          Always
-
-AccessModifierOffset: -4
-
-AlignEscapedNewlinesLeft: true
-
-AllowShortIfStatementsOnASingleLine: true
-AllowShortLoopsOnASingleLine: true
-
-AlwaysBreakAfterReturnType: AllDefinitions
-AlwaysBreakBeforeMultilineStrings: true
-AlwaysBreakTemplateDeclarations: true
-
-BreakBeforeBraces: Linux

.github/workflows/sphinx_deploy.yml

@@ -0,0 +1,56 @@
+name: Deploy docs site with Sphinx
+on:
+  push:
+    branches: ["main"]
+
+  # Allows you to run this workflow manually from the Actions tab
+  workflow_dispatch:
+
+# Sets permissions of the GITHUB_TOKEN to allow deployment to GitHub Pages
+permissions:
+  contents: read
+  pages: write
+  id-token: write
+
+# Allow only one concurrent deployment, skipping runs queued between the run in-progress and latest queued.
+# However, do NOT cancel in-progress runs as we want to allow these production deployments to complete.
+concurrency:
+  group: "pages"
+  cancel-in-progress: false
+
+# Default to bash
+defaults:
+  run:
+    shell: bash
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          submodules: recursive
+      - name: Setup Pages
+        id: pages
+        uses: actions/configure-pages@v4
+      - name: Sphinx build
+        uses: jsix-os/sphinx-action@master
+        with:
+          docs-folder: "docs/"
+      - name: Upload artifact
+        uses: actions/upload-pages-artifact@v3
+        with:
+          path: ./docs/_build/html
+
+  # Deployment job
+  deploy:
+    environment:
+      name: github-pages
+      url: ${{ steps.deployment.outputs.page_url }}
+    runs-on: ubuntu-latest
+    needs: build
+    steps:
+      - name: Deploy to GitHub Pages
+        id: deployment
+        uses: actions/deploy-pages@v4

.gitignore

@@ -1,10 +1,19 @@
+.cache
 .lock*
 /build*
 *.bak
 tags
-jsix.log
+*.log
+*.out
 *.o
 *.a
 sysroot
 .gdb_history
 .peru
+__pycache__
+/venv
+compile_commands.json
+buddy_allocs.txt
+frame_allocs.txt
+heap_allocs.txt
+/docs/_build

.vscode/c_cpp_properties.json

@@ -0,0 +1,19 @@
+{
+    "configurations": [
+        {
+            "name": "Linux",
+            "includePath": [
+                "${workspaceFolder}/src/libraries/**",
+                "${workspaceFolder}/build/**",
+                "${workspaceFolder}/sysroot/include"
+            ],
+            "defines": [],
+            "compilerPath": "/usr/bin/clang",
+            "cStandard": "c17",
+            "cppStandard": "c++17",
+            "intelliSenseMode": "linux-clang-x64",
+            "compileCommands": "${workspaceFolder}/compile_commands.json"
+        }
+    ],
+    "version": 4
+}

.vscode/launch.json

@@ -0,0 +1,32 @@
+{
+    // Use IntelliSense to learn about possible attributes.
+    // Hover to view descriptions of existing attributes.
+    // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
+    "version": "0.2.0",
+    "configurations": [
+        {
+            "name": "QEMU Debug Server",
+            "type": "cppdbg",
+            "request": "launch",
+            "program": "${workspaceFolder}/build/jsix.elf",
+            "args": [],
+            "cwd": "${workspaceFolder}",
+            "logging": {
+                //"engineLogging": true,
+                "programOutput": true
+            },  
+            "stopAtConnect": true,
+            "stopAtEntry": false,
+
+            "setupCommands": [
+                {"text": "dashboard -enabled off", "ignoreFailures": true}
+            ],
+
+            "MIMode": "gdb",
+            "miDebuggerServerAddress": "localhost:1234",
+            "debugServerPath": "${workspaceFolder}/qemu.sh",
+            "debugServerArgs": "--debug --no-build",
+            "serverLaunchTimeout": 5000,
+        }
+    ]
+}

.vscode/tasks.json

@@ -1,28 +1,41 @@
 {
     // See https://go.microsoft.com/fwlink/?LinkId=733558
     // for the documentation about the tasks.json format
-    "version": "0.1.0",
+    "version": "2.0.0",
     "tasks": [
         {
-            "taskName": "make",
+            "label": "Ninja",
-            "command": "make.bat",
+            "type": "shell",
-            "isBuildCommand": true
+            "command": "source ${workspaceFolder}/venv/bin/activate.fish; ninja",
+            "detail": "Build the project",
+            "options": {
+                "cwd": "${workspaceFolder}/build"
+            },
+            "group": {
+                "kind": "build",
+                "isDefault": true
+            },
+            "problemMatcher": []
         },
         {
-            "taskName": "clean",
+            "label": "Run QEMU",
-            "command": "make.bat",
+            "command": "./qemu.sh",
-            "args": [ "clean" ],
+            "args": [ "--no-build", "--kvm" ],
-            "isShellCommand": true
+            "dependsOn": ["Ninja"],
+            "options": {
+                "cwd": "${workspaceFolder}"
+            },
         },
         {
-            "taskName": "qemu (windowed)",
+            "label": "clean",
-            "command": "qemu-win.bat",
+            "command": "${workspaceFolder}/venv/bin/ninja",
-            "showOutput": "never",
+            "options": {
-            "isTestCommand": true
+                "cwd": "${workspaceFolder}/build"
-        },
+            },
-        {
+            "args": [
-            "taskName": "qemu",
+                "-t",
-            "command": "qemu.bat"
+                "clean"
+            ]
         }
     ]
 }

NOTES.md

@@ -1,38 +0,0 @@
-# Design / WIP notes
-
-## TODO
-
-- Paging manager
-  - Copy-on-write pages
-  - Better page-allocation model?
-- Allow for more than one IOAPIC in ACPI module
-  - The objects get created, but GSI lookup only uses the one at index 0
-- mark kernel memory pages global
-- Serial out based on circular/bip biffer and interrupts, not spinning on
-  `write_ready()`
-- Split out more code into kutil for testing
-- AHCI / MSI interrupts on Vbox break?
-- FXSAVE to save XMM registers.
-  - optimization using #NM (0x7) to detect SSE usage
-- Clean up of process memory maps
-- Better stack tracer
-- Bootloader rewrite
-  - C++ and sharing library code for ELF, initrd, etc
-  - Parse initrd and pre-load certain ELF images, eg the process loader process?
-  - Do initial memory bootstrap?
-- Calling global ctors
-- Device Tree
-  - Actual serial driver
-  - Disk driver
-    - File system
-- Multiprocessing
-  - Fast syscalls using syscall/sysret
-
-### Build
-
-- Clean up build generator and its templates
-  - More robust objects to represent modules & targets to pass to templates
-  - Read project setup from a simple JSON/TOML/etc
-  - Better compartmentalizing when doing template inheritance
-- Move to LLD as sysroot linker
-

README.md

@@ -1,15 +1,19 @@
-# jsix: A toy OS kernel
+![jsix](assets/jsix.svg)
 
-**jsix** is the kernel for the hobby OS that I am currently building. It's
+# The jsix operating system
-far from finished, or even being usable. Instead, it's a sandbox for me to play
+
-with kernel-level code and explore architectures.
+**jsix** is a custom multi-core x64 operating system that I am building from
+scratch. It's far from finished, or even being usable (see the *Status and
+Roadmap* section, below) but all currently-planned major kernel features are
+now implemented to at least a passable level.
 
 The design goals of the project are:
 
 * Modernity - I'm not interested in designing for legacy systems, or running on
-  all hardware out there. My target is only 64 bit architecutres, and modern
+  all hardware out there. My target is only 64 bit architectures, and modern
   commodity hardware. Currently that means x64 systems with Nehalem or newer
-  CPUs and UEFI firmware. Eventually I'd like to work on an AArch64 port,
+  CPUs and UEFI firmware. (See [this list][cpu_features] for the currently
+  required CPU features.) Eventually I'd like to work on an AArch64 port,
   partly to force myself to factor out the architecture-dependent pieces of the
   code base.
 
@@ -17,13 +21,11 @@ The design goals of the project are:
   processes as possible, in the microkernel fashion. A sub-goal of this is to
   explore where the bottlenecks of such a microkernel are now, and whether
   eschewing legacy hardware will let me design a system that's less bogged down
-  by the traditional microkernel problems. Given that there are no processes
+  by the traditional microkernel problems.
-  yet, the kernel is monolithic by default.
 
 * Exploration - I'm really mostly doing this to have fun learning and exploring
-  modern OS development. Modular design may be tossed out (hopefully
+  modern OS development. Initial feature implementations may temporarily throw
-  temporarily) in some places to allow me to play around with the related
+  away modular design to allow for exploration of the related hardware.
-  hardware.
 
 A note on the name: This kernel was originally named Popcorn, but I have since
 discovered that the Popcorn Linux project is also developing a kernel with that
@@ -31,48 +33,149 @@ name, started around the same time as this project. So I've renamed this kernel
 jsix (Always styled _jsix_ or `j6`, never capitalized) as an homage to L4, xv6,
 and my wonderful wife.
 
+[cpu_features]: https://github.com/justinian/jsix/blob/master/src/libraries/cpu/include/cpu/features.inc
+
+## Status and Roadmap
+
+The following major feature areas are targets for jsix development:
+
+#### UEFI boot loader
+
+_Done._ The bootloader loads the kernel and initial userspace programs, and
+sets up necessary kernel arguments about the memory map and EFI GOP
+framebuffer. Possible future ideas:
+
+- take over more init-time functions from the kernel
+- rewrite it in Zig
+
+#### Memory
+
+_Virtual memory: Sufficient._ The kernel manages virtual memory with a number
+of kinds of `vm_area` objects representing mapped areas, which can belong to
+one or more `vm_space` objects which represent a whole virtual memory space.
+(Each process has a `vm_space`, and so does the kernel itself.)
+
+Remaining to do:
+
+- TLB shootdowns
+- Page swapping
+- Large / huge page support
+
+_Physical page allocation: Sufficient._ The current physical page allocator
+implementation uses a group of blocks representing up-to-1GiB areas of usable
+memory as defined by the bootloader. Each block has a three-level bitmap
+denoting free/used pages.
+
+Future work:
+
+- Align blocks so that their first page is 1GiB-aligned, making finding free
+  large/huge pages easier.
+
+#### Multitasking
+
+_Sufficient._ The global scheduler object keeps separate ready/blocked lists
+per core. Cores periodically attempt to balance load via work stealing.
+
+User-space tasks are able to create threads as well as other processes.
+
+#### API
+
+_Syscalls: Sufficient._ User-space tasks are able to make syscalls to the
+kernel via fast SYSCALL/SYSRET instructions. Syscalls made via `libj6` look to
+both the callee and the caller like standard SysV ABI function calls. The
+implementations are wrapped in generated wrapper functions which validate the
+request, check capabilities, and find the appropriate kernel objects or handles
+before calling the implementation functions.
+
+_IPC: Working, needs optimization._ The current IPC primitives are:
+
+- _Mailboxes_: endpoints for asynchronously-delivered small messages. Currently
+  these messages are double-copied - once from the caller into a kernel buffer,
+  and once from the kernel to the receiver. This works and is not a major cause
+  of slowdown, but will need to be optimized in the future.
+- _Channels_: endpoints for asynchronous uni-directional streams of bytes.
+  Currently these also suffer from a double-copy problem, and should probably
+  be replaced eventually by userspace shared memory communication.
+- _Events_: objects that can be signalled to send asynchronous notifications to
+  waiting threads.
+
+#### Hardware Support
+
+- Framebuffer driver: _In progress._ Currently on machines with a video
+  device accessible by UEFI, jsix starts a user-space framebuffer driver that
+  only prints out kernel logs.
+- Serial driver: _In progress._ The current UART currently only exposes COM1
+  as an output channel, but no input or other serial ports are exposed.
+- USB driver: _To do_
+- AHCI (SATA) driver: _To do_
+
 ## Building
 
-jsix uses the [Ninja][] build tool, and generates the build files for it with a
+jsix uses the [Ninja][] build tool, and generates the build files for it with
-custom tool called [Bonnibel][]. Bonnibel can be installed with [Cargo][], or
+the `configure` script. The build also relies on a custom toolchain sysroot, which can be
-downloaded as a prebuilt binary from its Github repository.
+downloaded or built using the scripts in [jsix-os/toolchain][].
 
-[Ninja]:    https://ninja-build.org
+[Ninja]:             https://ninja-build.org
-[Bonnibel]: https://github.com/justinian/bonnibel
+[jsix-os/toolchain]: https://github.com/jsix-os/toolchain
-[Cargo]:    https://crates.io/crates/bonnibel
 
-Requrirements:
+Other build dependencies:
 
-* bonnibel
+* [clang][]: the C/C++ compiler
-* ninja
+* [nasm][]: the assembler
-* clang
+* [lld][]: the linker
-* nasm
+* [mtools][]: for creating the FAT image
-* mtools
-* curl for downloading the toolchain
 
-### Setting up the cross toolchain
+[clang]:    https://clang.llvm.org
+[nasm]:     https://www.nasm.us
+[lld]:      https://lld.llvm.org
+[mtools]:   https://www.gnu.org/software/mtools/
 
-Running `pb sync` will download and unpack the toolchain into `sysroot`. 
+The `configure` script has some Python dependencies - these can be installed via
+`pip`, though doing so in a python virtual environment is recommended.
+Installing via `pip` will also install `ninja`.
 
-#### Compiling the toolchain yourself
+A Debian 11 (Bullseye) system can be configured with the necessary build
+dependencies by running the following commands from the jsix repository root:
 
-If you have `clang` and `curl` installed, runing the `scripts/build_sysroot.sh`
+```bash
-script will download and build a LLVM toolchain configured for building jsix
+sudo apt install clang lld nasm mtools python3-pip python3-venv
-host binaries.
+python3 -m venv ./venv
+source venv/bin/activate
+pip install -r requirements.txt
+peru sync
+```
+
+### Setting up the sysroot
+
+Build or download the toolchain sysroot as mentioned above with
+[jsix-os/toolchain][], and symlink the built toolchain directory as `sysroot`
+at the root of this project.
+
+```bash
+# Example if both the toolchain and this project are cloned under ~/src
+ln -s ~/src/toolchain/toolchains/llvm-13 ~/src/jsix/sysroot
+```
 
 ### Building and running jsix
 
-Once the toolchain has been set up, running Bonnibel's `pb init` command will
+Once the toolchain has been set up, running the `./configure` script (see
-set up the build configuration, and `pb build` will actually run the build.  If
+`./configure --help` for available options) will set up the build configuration,
-you have `qemu-system-x86_64` installed, the `qemu.sh` script will to run jsix
+and `ninja -C build` (or wherever you put the build directory) will actually run
-in QEMU `-nographic` mode.
+the build. If you have `qemu-system-x86_64` installed, the `qemu.sh` script will
+to run jsix in QEMU `-nographic` mode.
 
-I personally run this either from a real debian amd64 testing/buster machine or
+I personally run this either from a real debian amd64 bullseye machine or
-a windows WSL debian testing/buster installation. The following should be
+a windows WSL debian bullseye installation. Your mileage may vary with other
-enough to set up such a system to build the kernel:
+setups and distros.
 
-    sudo apt install qemu-system-x86 nasm clang-6.0 mtools curl
+### Running the test suite
-    sudo update-alternatives /usr/bin/clang clang /usr/bin/clang-6.0 1000
-    sudo update-alternatives /usr/bin/clang++ clang++ /usr/bin/clang++-6.0 1000
-	curl -L -o pb https://github.com/justinian/bonnibel/releases/download/2.0.0/pb_linux_amd64 && chmod a+x pb
 
+jsix now has the `test_runner` userspace program that runs various automated
+tests. It is not included in the default build, but if you use the `test.yml`
+manifest it will be built, and can be run with the `test.sh` script or the
+`qemu.sh` script.
+
+```bash
+./configure --manifest=assets/manifests/test.yml
+if ./test.sh; then echo "All tests passed!"; else echo "Failed."; fi
+```

assets/build/amd64/config.debug.yaml

@@ -0,0 +1,9 @@
+---
+ccflags: [
+  "-g3",
+  "-ggdb",
+]
+
+ldflags: [
+  "-g",
+]

assets/build/amd64/config.release.yaml

@@ -0,0 +1,3 @@
+ccflags: [
+  "-O3",
+]

assets/build/amd64/global.yaml

@@ -0,0 +1,39 @@
+---
+cc: "${source_root}/sysroot/bin/clang"
+cxx: "${source_root}/sysroot/bin/clang++"
+ld: "${source_root}/sysroot/bin/ld.lld"
+ar: ar
+nasm: nasm
+objcopy: objcopy
+
+ccflags: [
+  "-I${source_root}/src/include",
+  "-fcolor-diagnostics",
+  "-U__STDCPP_THREADS__",
+  "-D_LIBCPP_HAS_NO_THREADS",
+  "-D__jsix_config=${build_config}",
+  "-D__jsix_config_${build_config}",
+  "-DVERSION_MAJOR=${version_major}",
+  "-DVERSION_MINOR=${version_minor}",
+  "-DVERSION_PATCH=${version_patch}",
+  "-DVERSION_GITSHA=0x${version_sha}",
+  '-DGIT_VERSION=\"${version_major}.${version_minor}.${version_patch}+${version_sha}\"',
+  '-DGIT_VERSION_WIDE=L\"${version_major}.${version_minor}.${version_patch}+${version_sha}\"',
+
+  "-Wformat=2", "-Winit-self", "-Winline", "-Wmissing-format-attribute",
+  "-Wmissing-include-dirs", "-Wswitch", "-Wundef", "-Wdisabled-optimization",
+  "-Wpointer-arith", "-Wno-attributes", "-Wno-sign-compare", "-Wno-multichar",
+  "-Wno-div-by-zero", "-Wno-endif-labels", "-Wno-pragmas", "-Wno-format-extra-args",
+  "-Wno-unused-result", "-Wno-deprecated-declarations", "-Wno-unused-function",
+  "-Wno-address-of-packed-member", "-Wno-invalid-offsetof", "-Wno-format-nonliteral",
+  "-Werror" ]
+
+asflags: [
+  "-DVERSION_MAJOR=${version_major}",
+  "-DVERSION_MINOR=${version_minor}",
+  "-DVERSION_PATCH=${version_patch}",
+  "-DVERSION_GITSHA=0x${version_sha}",
+  "-I${source_root}/src/include" ]
+
+cflags: [ "-std=c11" ]
+cxxflags: [ "-std=c++17" ]

assets/build/amd64/rules.ninja

@@ -0,0 +1,91 @@
+rule compile.c
+  command = $cc -MMD -MF $out.d $cflags $ccflags -o $out -c $in
+  description = Compiling [$target]:$name
+  depfile = $out.d
+  deps = gcc
+
+rule dump_c_defs
+  command = echo | $cc $ccflags $cflags -dM -E - > $out
+  description = Dumping C defines for $target
+
+rule dump_c_run
+  command = echo '#!/bin/bash' > $out; echo '$cc $ccflags $cflags $$*' >> $
+      $out; chmod a+x $out
+  description = Dumping C arguments for $target
+
+rule compile.cpp
+  command = $cxx -MMD -MF $out.d $cxxflags $ccflags -o $out -c $in
+  description = Compiling [$target]:$name
+  depfile = $out.d
+  deps = gcc
+
+rule dump_cpp_defs
+  command = echo | $cxx -x c++ $ccflags $cxxflags -dM -E - > $out
+  description = Dumping C++ defines for $target
+
+rule dump_cpp_run
+  command = echo '#!/bin/bash' > $out; echo '$cxx $ccflags $cxxflags $$*' $
+      >> $out; chmod a+x $out
+  description = Dumping C++ arguments for $target
+
+rule compile.s
+  command = $nasm -o $out -felf64 -MD $out.d $asflags $in
+  description = Assembling [$target]:$name
+  depfile = $out.d
+  deps = gcc
+
+rule parse.cog
+  command = cog -o $out -d -D target=$target $cogflags $in
+  description = Parsing [$target]:$name
+
+rule exe
+  command = $ld $ldflags -o $out $in $libs
+  description = Linking exe [$target]:$name
+
+rule driver
+  command = $ld $ldflags -o $out $in $libs
+  description = Linking driver [$target]:$name
+
+rule lib
+  command = $ld -shared -soname $soname $ldflags -o $out $in $libs
+  description = Linking [$target]:$name
+
+rule lib_static
+  command = $ar qcs $out $in
+  description = Archiving [$target]:$name
+
+rule cp
+  command = cp $in $out
+  description = Copying [$target]:$name
+
+rule dump
+  command = objdump -DSC -M intel $in > $out
+  description = Dumping decompiled $name
+
+rule makest
+    description = Making symbol table
+    command = nm -n -S --demangle $in | ${source_root}/scripts/build_symbol_table.py $out
+
+rule makeinitrd
+    description = Creating $name
+    command = ${source_root}/scripts/mkj6romfs.py -c $format $in $out
+
+rule makefat
+    description = Creating $name
+    command = $
+        cp $in $out; $
+        mcopy -s -D o -i $out@@1M ${build_root}/fatroot/* ::/
+
+rule strip
+    description = Stripping $name
+    command = $
+        cp $in $out; $
+        objcopy --only-keep-debug $out $debug; $
+        strip --discard-all -g $out; $
+        objcopy --add-gnu-debuglink=$debug $out
+
+rule touch
+    command = touch $out
+
+rule compdb
+    command = ninja -t compdb > $out

assets/build/amd64/target.boot.yaml

@@ -0,0 +1,26 @@
+---
+ld: clang++
+
+ccflags: [
+  "-nostdlib",
+  "-nodefaultlibs",
+  "-fno-builtin",
+
+  "-I${source_root}/external",
+  "--target=x86_64-unknown-windows",
+  "-ffreestanding",
+  "-mno-red-zone",
+  "-fshort-wchar",
+  "-fno-omit-frame-pointer",
+]
+
+cxxflags: [ "-fno-exceptions", "-fno-rtti" ]
+
+ldflags: [
+  "--target=x86_64-unknown-windows",
+  "-nostdlib",
+  "-Wl,-entry:efi_main",
+  "-Wl,-subsystem:efi_application",
+  "-fuse-ld=lld-link",
+]
+

assets/build/amd64/target.init.yaml

@@ -0,0 +1,40 @@
+---
+ccflags: [
+  "--target=x86_64-jsix-elf",
+  "-fno-omit-frame-pointer",
+  "-fno-stack-protector",
+
+  "-fvisibility=hidden",
+  "-fvisibility-inlines-hidden",
+
+  "-D__ELF__",
+  "-D__jsix__",
+  "-U__linux",
+  "-U__linux__",
+
+  "-DMSPACES",
+
+  "--sysroot='${source_root}/sysroot'"
+]
+
+
+cxxflags: [
+  "-fno-exceptions",
+  "-fno-rtti",
+  "-isystem", "${source_root}/sysroot/include/c++/v1",
+]
+
+ldflags: [
+  "-Bstatic",
+  "-m", "elf_x86_64",
+  "--sysroot='${source_root}/sysroot'",
+  "--no-eh-frame-hdr",
+  "-L", "${source_root}/sysroot/lib",
+  "-z", "separate-code",
+  "-lc++", "-lc++abi", "-lunwind",
+  "--no-dependent-libraries",
+]
+
+libs: [
+  "${target_dir}/crt0.o",
+]

assets/build/amd64/target.kernel.yaml

@@ -0,0 +1,52 @@
+---
+asflags: [ "-I${source_root}/src/kernel/" ]
+
+ccflags: [
+  "--target=x86_64-jsix-elf",
+  "-fno-stack-protector",
+
+  "-I${source_root}/external",
+
+  "-nostdinc",
+  "-nostdlib",
+  "-ffreestanding",
+  "-nodefaultlibs",
+  "-fno-builtin",
+  "-fno-plt",
+
+  "-mno-sse",
+  "-fno-omit-frame-pointer",
+  "-mno-red-zone",
+  "-mcmodel=kernel",
+  "-fvisibility=hidden",
+  "-fvisibility-inlines-hidden",
+
+  "-D__ELF__",
+  "-D__jsix__",
+  "-D__j6kernel",
+  "-U__linux",
+  "-U__linux__",
+  "-DPRINTF_ALIAS_STANDARD_FUNCTION_NAMES=1",
+  "-DPRINTF_INCLUDE_CONFIG_H=1",
+
+  "--sysroot='${source_root}/sysroot'"
+]
+
+
+cflags: [ '-nostdinc' ]
+
+cxxflags: [
+  "-fno-exceptions",
+  "-fno-rtti",
+  "-nostdinc",
+]
+
+ldflags: [
+  "-m", "elf_x86_64",
+  "-nostdlib",
+  "-Bstatic",
+  "--no-eh-frame-hdr",
+  "-z", "norelro",
+  "-z", "separate-code"
+]
+

assets/build/amd64/target.user.exe.yaml

@@ -0,0 +1,15 @@
+---
+
+ccflags: [
+  "-fpie"
+]
+
+ldflags: [
+  "-pie",
+  "--dynamic-linker", "/jsix/lib/ld.so",
+  "--push-state", "--as-needed", "-Bstatic", "-lc++", "-lc++abi", "-lunwind", "--pop-state",
+]
+
+libs: [
+  "${target_dir}/crt0.o",
+]

assets/build/amd64/target.user.shared.yaml

@@ -0,0 +1,7 @@
+---
+ccflags: [
+]
+
+ldflags: [
+  "-shared",
+]

assets/build/amd64/target.user.yaml

@@ -0,0 +1,34 @@
+---
+asflags: []
+
+ccflags: [
+  "--target=x86_64-jsix-elf",
+  "-fno-omit-frame-pointer",
+  "-fno-stack-protector",
+
+  "-fvisibility=hidden",
+  "-fvisibility-inlines-hidden",
+
+  "-D__ELF__",
+  "-D__jsix__",
+  "-U__linux",
+  "-U__linux__",
+
+  "--sysroot='${source_root}/sysroot'",
+  "-fpic",
+]
+
+cxxflags: [
+  "-fno-exceptions",
+  "-fno-rtti",
+  "-isystem", "${source_root}/sysroot/include/c++/v1",
+  ]
+
+ldflags: [
+  "-m", "elf_x86_64",
+  "--sysroot='${source_root}/sysroot'",
+  "--no-eh-frame-hdr",
+  "-L", "${source_root}/sysroot/lib",
+  "-z", "separate-code",
+  "--no-dependent-libraries",
+]

assets/build/linux/config.debug.yaml

@@ -0,0 +1,9 @@
+---
+ccflags: [
+  "-g3",
+  "-ggdb",
+]
+
+ldflags: [
+  "-g",
+]

assets/build/linux/config.release.yaml

@@ -0,0 +1,3 @@
+ccflags: [
+  "-O3",
+]

assets/build/linux/global.yaml

@@ -0,0 +1,39 @@
+---
+cc: "clang-16"
+cxx: "clang++-16"
+ld: "ld.lld-16"
+ar: ar
+nasm: nasm
+objcopy: objcopy
+
+ccflags: [
+  "-I${source_root}/src/include",
+  "-I${source_root}/sysroot/include/c++/v1",
+  "-fcolor-diagnostics",
+  "-U__STDCPP_THREADS__",
+  "-D__jsix_config=${build_config}",
+  "-D__jsix_config_${build_config}",
+  "-DVERSION_MAJOR=${version_major}",
+  "-DVERSION_MINOR=${version_minor}",
+  "-DVERSION_PATCH=${version_patch}",
+  "-DVERSION_GITSHA=0x${version_sha}",
+  '-DGIT_VERSION=\"${version_major}.${version_minor}.${version_patch}+${version_sha}\"',
+  '-DGIT_VERSION_WIDE=L\"${version_major}.${version_minor}.${version_patch}+${version_sha}\"',
+
+  "-Wformat=2", "-Winit-self", "-Winline", "-Wmissing-format-attribute",
+  "-Wmissing-include-dirs", "-Wswitch", "-Wundef", "-Wdisabled-optimization",
+  "-Wpointer-arith", "-Wno-attributes", "-Wno-sign-compare", "-Wno-multichar",
+  "-Wno-div-by-zero", "-Wno-endif-labels", "-Wno-pragmas", "-Wno-format-extra-args",
+  "-Wno-unused-result", "-Wno-deprecated-declarations", "-Wno-unused-function",
+  "-Wno-address-of-packed-member", "-Wno-invalid-offsetof", "-Wno-format-nonliteral",
+  "-Werror" ]
+
+asflags: [
+  "-DVERSION_MAJOR=${version_major}",
+  "-DVERSION_MINOR=${version_minor}",
+  "-DVERSION_PATCH=${version_patch}",
+  "-DVERSION_GITSHA=0x${version_sha}",
+  "-I${source_root}/src/include" ]
+
+cflags: [ "-std=c11" ]
+cxxflags: [ "-std=c++17" ]

assets/build/linux/rules.ninja

@@ -0,0 +1,91 @@
+rule compile.c
+  command = $cc -MMD -MF $out.d $cflags $ccflags -o $out -c $in
+  description = Compiling [$target]:$name
+  depfile = $out.d
+  deps = gcc
+
+rule dump_c_defs
+  command = echo | $cc $ccflags $cflags -dM -E - > $out
+  description = Dumping C defines for $target
+
+rule dump_c_run
+  command = echo '#!/bin/bash' > $out; echo '$cc $ccflags $cflags $$*' >> $
+      $out; chmod a+x $out
+  description = Dumping C arguments for $target
+
+rule compile.cpp
+  command = $cxx -MMD -MF $out.d $cxxflags $ccflags -o $out -c $in
+  description = Compiling [$target]:$name
+  depfile = $out.d
+  deps = gcc
+
+rule dump_cpp_defs
+  command = echo | $cxx -x c++ $ccflags $cxxflags -dM -E - > $out
+  description = Dumping C++ defines for $target
+
+rule dump_cpp_run
+  command = echo '#!/bin/bash' > $out; echo '$cxx $ccflags $cxxflags $$*' $
+      >> $out; chmod a+x $out
+  description = Dumping C++ arguments for $target
+
+rule compile.s
+  command = $nasm -o $out -felf64 -MD $out.d $asflags $in
+  description = Assembling [$target]:$name
+  depfile = $out.d
+  deps = gcc
+
+rule parse.cog
+  command = cog -o $out -d -D target=$target $cogflags $in
+  description = Parsing [$target]:$name
+
+rule exe
+  command = $ld $ldflags -o $out $in $libs
+  description = Linking exe [$target]:$name
+
+rule driver
+  command = $ld $ldflags -o $out $in $libs
+  description = Linking driver [$target]:$name
+
+rule lib
+  command = $ld -shared -soname $soname $ldflags -o $out $in $libs
+  description = Linking [$target]:$name
+
+rule lib_static
+  command = $ar qcs $out $in
+  description = Archiving [$target]:$name
+
+rule cp
+  command = cp $in $out
+  description = Copying [$target]:$name
+
+rule dump
+  command = objdump -DSC -M intel $in > $out
+  description = Dumping decompiled $name
+
+rule makest
+    description = Making symbol table
+    command = nm -n -S --demangle $in | ${source_root}/scripts/build_symbol_table.py $out
+
+rule makeinitrd
+    description = Creating $name
+    command = ${source_root}/scripts/mkj6romfs.py -c $format $in $out
+
+rule makefat
+    description = Creating $name
+    command = $
+        cp $in $out; $
+        mcopy -s -D o -i $out@@1M ${build_root}/fatroot/* ::/
+
+rule strip
+    description = Stripping $name
+    command = $
+        cp $in $out; $
+        objcopy --only-keep-debug $out $debug; $
+        strip --discard-all -g $out; $
+        objcopy --add-gnu-debuglink=$debug $out
+
+rule touch
+    command = touch $out
+
+rule compdb
+    command = ninja -t compdb > $out

assets/build/linux/target.init.yaml

@@ -0,0 +1,39 @@
+---
+ccflags: [
+  "--target=x86_64-jsix-elf",
+  "-fno-omit-frame-pointer",
+  "-fno-stack-protector",
+
+  "-fvisibility=hidden",
+  "-fvisibility-inlines-hidden",
+
+  "-D__ELF__",
+  "-D__jsix__",
+  "-U__linux",
+  "-U__linux__",
+
+  "-DMSPACES",
+
+  "--sysroot='${source_root}/sysroot'"
+]
+
+
+cxxflags: [
+  "-fno-exceptions",
+  "-fno-rtti",
+]
+
+ldflags: [
+  "-Bstatic",
+  "-m", "elf_x86_64",
+  "--sysroot='${source_root}/sysroot'",
+  "--no-eh-frame-hdr",
+  "-L", "${source_root}/sysroot/lib",
+  "-z", "separate-code",
+  "-lc++", "-lc++abi", "-lunwind",
+  "--no-dependent-libraries",
+]
+
+libs: [
+  "${target_dir}/crt0.o",
+]

assets/build/linux/target.user.exe.yaml

@@ -0,0 +1,16 @@
+---
+
+ccflags: [
+  "-fpie"
+]
+
+ldflags: [
+  "-pie",
+  "-rpath", "${target_dir}",
+  "--dynamic-linker", "/lib64/ld-linux-x86-64.so.2",
+  "--push-state", "--as-needed", "-Bstatic", "-lc++", "-lc++abi", "-lunwind", "--pop-state",
+]
+
+libs: [
+  "${target_dir}/crt0.o",
+]

assets/build/linux/target.user.shared.yaml

@@ -0,0 +1,7 @@
+---
+ccflags: [
+]
+
+ldflags: [
+  "-shared",
+]

assets/build/linux/target.user.yaml

@@ -0,0 +1,33 @@
+---
+asflags: []
+
+ccflags: [
+  "--target=x86_64-jsix-elf",
+  "-fno-omit-frame-pointer",
+  "-fno-stack-protector",
+
+  "-fvisibility=hidden",
+  "-fvisibility-inlines-hidden",
+
+  "-D__ELF__",
+  "-D__jsix__",
+  "-U__linux",
+  "-U__linux__",
+
+  "--sysroot='${source_root}/sysroot'",
+  "-fpic",
+]
+
+cxxflags: [
+  "-fno-exceptions",
+  "-fno-rtti",
+  ]
+
+ldflags: [
+  "-m", "elf_x86_64",
+  "--sysroot='${source_root}/sysroot'",
+  "--no-eh-frame-hdr",
+  "-L", "${source_root}/sysroot/lib",
+  "-z", "separate-code",
+  "--no-dependent-libraries",
+]

assets/debugging/jsix.elf-gdb.py

@@ -1,4 +1,14 @@
 import gdb
+import gdb.printing
+
+import sys
+sys.path.append('./scripts')
+
+import re
+
+from collections import namedtuple
+Capability = namedtuple("Capability", ["id", "parent", "refcount", "caps", "type", "koid"])
+LogEntry = namedtuple("LogHeader", ["id", "bytes", "severity", "area", "message"])
 
 class PrintStackCommand(gdb.Command):
     def __init__(self):
@@ -10,6 +20,7 @@ class PrintStackCommand(gdb.Command):
         base = "$rsp"
         if len(args) > 0:
             base = args[0]
+        base = int(gdb.parse_and_eval(base))
 
         depth = 22
         if len(args) > 1:
@@ -18,14 +29,35 @@ class PrintStackCommand(gdb.Command):
         for i in range(depth-1, -1, -1):
             try:
                 offset = i * 8
-                base_addr = gdb.parse_and_eval(base)
+                value = gdb.parse_and_eval(f"*(uint64_t*)({base:#x} + {offset:#x})")
-                value = gdb.parse_and_eval(f"*(uint64_t*)({base} + {offset})")
+                print("{:016x} (+{:04x}): {:016x}".format(base + offset, offset, int(value)))
-                print("{:016x} (+{:04x}): {:016x}".format(int(base_addr) + offset, offset, int(value)))
             except Exception as e:
                 print(e)
                 continue
 
 
+def stack_walk(frame, depth):
+    for i in range(depth-1, -1, -1):
+        ret = gdb.parse_and_eval(f"*(uint64_t*)({frame:#x} + 0x8)")
+
+        name = ""
+        try:
+            block = gdb.block_for_pc(int(ret))
+            if block:
+                name = block.function or ""
+        except RuntimeError:
+            pass
+
+        try:
+            print("{:016x}: {:016x} {}".format(int(frame), int(ret), name))
+            frame = int(gdb.parse_and_eval(f"*(uint64_t*)({frame:#x})"))
+        except gdb.MemoryError:
+            return
+
+        if frame == 0 or ret == 0:
+            return
+
+
 class PrintBacktraceCommand(gdb.Command):
     def __init__(self):
         super().__init__("j6bt", gdb.COMMAND_DATA)
@@ -33,31 +65,477 @@ class PrintBacktraceCommand(gdb.Command):
     def invoke(self, arg, from_tty):
         args = gdb.string_to_argv(arg)
 
-        depth = 30
-        if len(args) > 0:
-            depth = int(args[0])
-
         frame = "$rbp"
+        if len(args) > 0:
+            frame = args[0]
+
+        frame = int(gdb.parse_and_eval(f"{frame}"))
+
+        depth = 30
         if len(args) > 1:
-            frame = args[1]
+            depth = int(gdb.parse_and_eval(args[1]))
 
-        for i in range(depth-1, -1, -1):
+        stack_walk(frame, depth)
-            ret = gdb.parse_and_eval(f"*(uint64_t*)({frame} + 8)")
-            frame = gdb.parse_and_eval(f"*(uint64_t*)({frame})")
 
-            name = ""
-            block = gdb.block_for_pc(int(ret))
-            if block:
-                name = block.function or ""
 
-            print("{:016x} {}".format(int(ret), name))
+class TableWalkCommand(gdb.Command):
+    def __init__(self):
+        super().__init__("j6tw", gdb.COMMAND_DATA)
 
-            if frame == 0 or ret == 0:
+    def invoke(self, arg, from_tty):
-                return
+        args = gdb.string_to_argv(arg)
+        if len(args) < 2:
+            raise Exception("Must be: j6tw <pml4> <addr>")
 
+        pml4 = int(gdb.parse_and_eval(args[0]))
+        addr = int(gdb.parse_and_eval(args[1]))
+
+        indices = [
+            (addr >> 39) & 0x1ff,
+            (addr >> 30) & 0x1ff,
+            (addr >> 21) & 0x1ff,
+            (addr >> 12) & 0x1ff,
+            ]
+
+        names = ["PML4", "PDP", "PD", "PT"]
+
+        table_flags = [
+            (0x0001, "present"),
+            (0x0002, "write"),
+            (0x0004, "user"),
+            (0x0008, "pwt"),
+            (0x0010, "pcd"),
+            (0x0020, "accessed"),
+            (0x0040, "dirty"),
+            (0x0080, "largepage"),
+            (0x0100, "global"),
+            (0x1080, "pat"),
+            ((1<<63), "xd"),
+        ]
+
+        page_flags = [
+            (0x0001, "present"),
+            (0x0002, "write"),
+            (0x0004, "user"),
+            (0x0008, "pwt"),
+            (0x0010, "pcd"),
+            (0x0020, "accessed"),
+            (0x0040, "dirty"),
+            (0x0080, "pat"),
+            (0x0100, "global"),
+            ((1<<63), "xd"),
+        ]
+
+        flagsets = [table_flags, table_flags, table_flags, page_flags]
+
+        table = pml4
+        entry = 0
+        for i in range(len(indices)):
+            entry = int(gdb.parse_and_eval(f'((uint64_t*)0x{table:x})[0x{indices[i]:x}]'))
+            flagset = flagsets[i]
+            flag_names = " | ".join([f[1] for f in flagset if (entry & f[0]) == f[0]])
+
+            print(f"{names[i]:>4}: {table:016x}")
+            print(f"      index: {indices[i]:3} {entry:016x}")
+            print(f"      flags: {flag_names}")
+
+            if (entry & 1) == 0 or (i < 3 and (entry & 0x80)):
+                break
+
+            table = (entry & 0x7ffffffffffffe00) | 0xffffc00000000000
+
+
+class GetThreadsCommand(gdb.Command):
+    FLAGS = {
+        "ready":    0x01,
+        "loading":  0x02,
+        "exited":   0x04,
+        "constant": 0x80,
+    }
+
+    def __init__(self):
+        super().__init__("j6threads", gdb.COMMAND_DATA)
+
+    def get_flags(self, bitset):
+        flags = []
+        for k, v in GetThreadsCommand.FLAGS.items():
+            if bitset & v:
+                flags.append(k)
+        return " ".join(flags)
+
+    def print_thread(self, addr):
+        if addr == 0:
+            print("  <no thread>\n")
+            return
+
+        tcb = f"((TCB*){addr:#x})"
+        thread = f"({tcb}->thread)"
+        stack = int(gdb.parse_and_eval(f"{tcb}->kernel_stack"))
+        rsp = int(gdb.parse_and_eval(f"{tcb}->rsp"))
+        pri = int(gdb.parse_and_eval(f"{tcb}->priority"))
+        flags = int(gdb.parse_and_eval(f"{thread}->m_state"))
+        koid = int(gdb.parse_and_eval(f"{thread}->m_obj_id"))
+        proc = int(gdb.parse_and_eval(f"{thread}->m_parent.m_obj_id"))
+
+        creator = int(gdb.parse_and_eval(f"{thread}->m_creator"))
+        if creator != 0:
+            creator_koid = int(gdb.parse_and_eval(f"{thread}->m_creator->m_obj_id"))
+            creator = f"{creator_koid:x}"
+        else:
+            creator = "<no thread>"
+
+        print(f"  Thread {proc:x}:{koid:x}")
+        print(f"         creator: {creator}")
+        print(f"        priority: {pri}")
+        print(f"           flags: {self.get_flags(flags)}")
+        print(f"          kstack: {stack:#x}")
+        print(f"             rsp: {rsp:#x}")
+
+        if stack == 0: return 0
+        return int(gdb.parse_and_eval(f"{tcb}->rsp"))
+
+
+    def print_thread_list(self, addr, name):
+        if addr == 0:
+            return
+
+        print(f"=== {name} ===")
+
+        while addr != 0:
+            rsp = self.print_thread(addr)
+            if rsp != 0:
+                print("------------------------------------")
+                stack_walk(rsp + 5*8, 5)
+
+            addr = int(gdb.parse_and_eval(f"((tcb_node*){addr:#x})->m_next"))
+            print()
+
+    def print_cpudata(self, index):
+        cpu = f"(g_cpu_data[{index}])"
+        tss = f"{cpu}->tss"
+        tss_rsp0 = int(gdb.parse_and_eval(f"{tss}->m_rsp[0]"))
+        tss_ist = [int(gdb.parse_and_eval(f"{tss}->m_ist[{i}]")) for i in range(8)]
+        print(f"        tss rsp0: {tss_rsp0:#x}")
+        for i in range(1,8):
+            if tss_ist[i] == 0: continue
+            print(f"        tss ist{i}: {tss_ist[i]:#x}")
+
+    def invoke(self, arg, from_tty):
+        args = gdb.string_to_argv(arg)
+        if len(args) > 1:
+            raise RuntimeError("Usage: j6threads [cpu]")
+
+        ncpus = int(gdb.parse_and_eval("g_num_cpus"))
+        cpus = list(range(ncpus))
+        if len(args) == 1:
+            cpus = [int(args[0])]
+
+        for cpu in cpus:
+            runlist = f"scheduler::s_instance->m_run_queues.m_elements[{cpu:#x}]"
+
+            print(f"=== CPU {cpu:2}: CURRENT ===")
+            current = int(gdb.parse_and_eval(f"{runlist}.current"))
+            self.print_thread(current)
+            self.print_cpudata(cpu)
+
+            previous = int(gdb.parse_and_eval(f"{runlist}.prev"))
+            print(f"            prev: {previous:x}")
+            print()
+
+            for pri in range(8):
+                ready = int(gdb.parse_and_eval(f"{runlist}.ready[{pri:#x}].m_head"))
+                self.print_thread_list(ready, f"CPU {cpu:2}: PRIORITY {pri}")
+
+            blocked = int(gdb.parse_and_eval(f"{runlist}.blocked.m_head"))
+            self.print_thread_list(blocked, f"CPU {cpu:2}: BLOCKED")
+            print()
+
+
+class PrintProfilesCommand(gdb.Command):
+    def __init__(self):
+        super().__init__("j6prof", gdb.COMMAND_DATA)
+
+    def invoke(self, arg, from_tty):
+        args = gdb.string_to_argv(arg)
+        if len(args) != 1:
+            raise RuntimeError("Usage: j6prof <profiler class>")
+
+        profclass = args[0]
+        root_type = f"profile_class<{profclass}>"
+
+        try:
+            baseclass = gdb.lookup_type(root_type)
+        except Exception as e:
+            print(e)
+            return
+
+        results = {}
+        max_len = 0
+        count = gdb.parse_and_eval(f"{profclass}::count")
+        for i in range(count):
+            name = gdb.parse_and_eval(f"{root_type}::function_names[{i:#x}]")
+            if name == 0: continue
+
+            call_counts = gdb.parse_and_eval(f"{root_type}::call_counts[{i:#x}]")
+            call_durations = gdb.parse_and_eval(f"{root_type}::call_durations[{i:#x}]")
+            results[name.string()] = float(call_durations) / float(call_counts)
+            max_len = max(max_len, len(name.string()))
+
+        for name, avg in results.items():
+            print(f"{name:>{max_len}}: {avg:15.3f}")
+
+
+class DumpLogCommand(gdb.Command):
+    level_names = ["", "fatal", "error", "warn", "info", "verbose", "spam"]
+
+    def __init__(self):
+        super().__init__("j6log", gdb.COMMAND_DATA)
+
+        from memory import Layout
+        layout = Layout("definitions/memory_layout.yaml")
+        for region in layout.regions:
+            if region.name == "logs":
+                self.base_addr = region.start
+                break
+
+        self.areas = []
+        area_re = re.compile(r"LOG\(\s*(\w+).*")
+        with open("src/libraries/j6/include/j6/tables/log_areas.inc", 'r') as areas_inc:
+            for line in areas_inc:
+                m = area_re.match(line)
+                if m:
+                    self.areas.append(m.group(1))
+
+    def get_entry(self, addr):
+        addr = int(addr)
+        size = int(gdb.parse_and_eval(f"((j6_log_entry*){addr:#x})->bytes"))
+        mlen = size - 8
+
+        return LogEntry(
+                int(gdb.parse_and_eval(f"((j6_log_entry*){addr:#x})->id")),
+                size,
+                int(gdb.parse_and_eval(f"((j6_log_entry*){addr:#x})->severity")),
+                int(gdb.parse_and_eval(f"((j6_log_entry*){addr:#x})->area")),
+                gdb.parse_and_eval(f"((j6_log_entry*){addr:#x})->message").string(length=mlen))
+
+    def invoke(self, arg, from_tty):
+        start = gdb.parse_and_eval("g_logger.m_start & (g_logger.m_buffer.count - 1)")
+        end = gdb.parse_and_eval("g_logger.m_end & (g_logger.m_buffer.count - 1)")
+        if end < start:
+            end += gdb.parse_and_eval("g_logger.m_buffer.count")
+
+        print(f"Logs are {start} -> {end}")
+
+        addr = self.base_addr + start
+        end += self.base_addr
+        while addr < end:
+            entry = self.get_entry(addr)
+            if entry.bytes < 8:
+                print(f"Bad log header size: {entry.bytes}")
+                break
+            addr += entry.bytes
+            area = "??"
+            if entry.area < len(self.areas):
+                area = self.areas[entry.area]
+            level = self.level_names[entry.severity]
+            print(f"{area:>7}:{level:7}  {entry.message}")
+
+
+class ShowCurrentProcessCommand(gdb.Command):
+    def __init__(self):
+        super().__init__("j6current", gdb.COMMAND_DATA)
+
+    def invoke(self, arg, from_tty):
+        def get_obj_and_id(name):
+            obj = int(gdb.parse_and_eval(f"((cpu_data*)$gs_base)->{name}"))
+            oid = -1
+            if obj != 0:
+                oid = int(gdb.parse_and_eval(f"((obj::kobject*){obj:#x})->m_obj_id"))
+            return obj, oid
+
+        process, pid = get_obj_and_id("process")
+        thread, tid = get_obj_and_id("thread")
+        print(f"{pid:02x}/{tid:02x}  [ {process:x} / {thread:x} ]")
+
+
+class CapTablePrinter:
+    def __init__(self, val):
+        node_map = val["m_caps"]
+        self.nodes = node_map["m_nodes"]
+        self.count = int(node_map["m_count"])
+        self.capacity = int(node_map["m_capacity"])
+
+    class _iterator:
+        def __init__(self, nodes, capacity):
+            self.nodes = []
+
+            for i in range(capacity):
+                node = nodes[i]
+
+                node_id = int(node["id"])
+                if node_id == 0:
+                    continue
+
+                self.nodes.append(Capability(
+                    id = node_id,
+                    parent = int(node["parent"]),
+                    refcount = int(node["holders"]),
+                    caps = int(node["caps"]),
+                    type = str(node["type"])[14:],
+                    koid = node['object']['m_obj_id']))
+
+            self.nodes.sort(key=lambda n: n.id, reverse=True)
+
+        def __iter__(self):
+            return self
+
+        def __next__(self):
+            if not self.nodes:
+                raise StopIteration
+
+            node = self.nodes.pop()
+
+            desc = f'p:{node.parent:016x}  refs:{node.refcount:2}  caps:{node.caps:04x}  {node.type:14} {node.koid}'
+            return (f"{node.id:016x}", desc)
+
+    def to_string(self):
+        return f"Cap table with {self.count}/{self.capacity} nodes.\n"
+
+    def children(self):
+        return self._iterator(self.nodes, self.capacity)
+
+
+class VectorPrinter:
+    def __init__(self, vector):
+        self.name = vector.type.tag
+        self.count = vector["m_size"]
+        self.array = vector["m_elements"]
+
+    class _iterator:
+        def __init__(self, array, count, deref):
+            self.array = array
+            self.count = count
+            self.deref = deref
+            self.index = 0
+
+        def __iter__(self):
+            return self
+
+        def __next__(self):
+            if self.index >= self.count:
+                raise StopIteration
+
+            item = self.array[self.index]
+            if self.deref:
+                item = item.dereference()
+
+            result = (f"{self.index:3}", item)
+            self.index += 1
+            return result
+
+    def to_string(self):
+        return f"{self.name} [{self.count}]"
+
+    def children(self):
+        deref = self.count > 0 and self.array[0].type.code == gdb.TYPE_CODE_PTR
+        return self._iterator(self.array, int(self.count), deref)
+
+
+class HandleSetPrinter:
+    def __init__(self, nodeset):
+        self.node_map = nodeset['m_map']
+        self.count = self.node_map['m_count']
+        self.capacity = self.node_map['m_capacity']
+
+    class _iterator:
+        def __init__(self, nodes, capacity):
+            self.items = []
+            self.index = 0
+            for i in range(capacity):
+                item = nodes[i]
+                if int(item) != 0:
+                    self.items.append(int(item))
+            self.items.sort()
+
+        def __iter__(self):
+            return self
+
+        def __next__(self):
+            if self.index >= len(self.items):
+                raise StopIteration
+            result = (f"{self.index}", f"{self.items[self.index]:016x}")
+            self.index += 1
+            return result
+
+    def to_string(self):
+        return f"node_set[{self.count} / {self.capacity}]:"
+
+    def children(self):
+        return self._iterator(self.node_map['m_nodes'], self.capacity)
+
+
+class LinkedListPrinter:
+    def __init__(self, llist):
+        self.name = llist.type.tag
+        self.head = llist['m_head']
+        self.tail = llist['m_tail']
+
+        self.items = []
+        current = self.head
+        while current:
+            item = current.dereference()
+            self.items.append((str(len(self.items)), item))
+            current = item['m_next']
+
+    class _iterator:
+        def __iter__(self):
+            return self
+
+        def __next__(self):
+            raise StopIteration
+
+    def to_string(self):
+        return f"{self.name}[{len(self.items)}]"
+
+    def children(self):
+        return self.items
+
+
+class IsRunning(gdb.Function):
+    def __init__(self):
+        super(IsRunning, self).__init__("is_running")
+
+    def invoke(self):
+        inferior = gdb.selected_inferior()
+        return \
+            inferior is not None and \
+            inferior.is_valid() and \
+            len(inferior.threads()) > 0
+
+
+def build_pretty_printers():
+    pp = gdb.printing.RegexpCollectionPrettyPrinter("jsix")
+    pp.add_printer("cap table", '^cap_table$', CapTablePrinter)
+    pp.add_printer("handle set", '^util::node_set<unsigned long, 0, heap_allocated>$', HandleSetPrinter)
+    pp.add_printer("vector", '^util::vector<.*>$', VectorPrinter)
+    pp.add_printer("linked list", '^util::linked_list<.*>$', LinkedListPrinter)
+    return pp
+
+gdb.printing.register_pretty_printer(
+    gdb.current_objfile(),
+    build_pretty_printers())
 
 PrintStackCommand()
 PrintBacktraceCommand()
+TableWalkCommand()
+GetThreadsCommand()
+PrintProfilesCommand()
+DumpLogCommand()
+ShowCurrentProcessCommand()
+IsRunning()
 
-gdb.execute("target remote :1234")
 gdb.execute("display/i $rip")
+gdb.execute("define hook-quit\nif $is_running()\n  kill\nend\nend")
+if not gdb.selected_inferior().was_attached:
+    gdb.execute("add-symbol-file build/panic.serial.elf")
+    gdb.execute("target remote :1234")

assets/grammars/definitions.g

@@ -0,0 +1,49 @@
+start: import_statement* (object|interface)+
+
+import_statement: "import" PATH
+
+object: description? "object" name options? super? "{" uid cname? capabilities? method* "}"
+
+interface: description? "interface" name options? "{" uid interface_param* "}"
+
+?interface_param: expose | function
+
+expose: "expose" type
+
+uid: "uid" UID
+
+cname: "cname" IDENTIFIER
+
+capabilities: "capabilities" "[" IDENTIFIER+ "]"
+
+super: ":" name
+
+function: description? "function" name options? ("{" param* "}")?
+
+method: description? "method" name options? ("{" param* "}")?
+
+param: "param" name type options? description?
+
+?type: PRIMITIVE | object_name | struct_name
+
+object_name: "ref" name
+
+struct_name: "struct" name
+
+id: NUMBER
+name: IDENTIFIER
+options: "[" ( OPTION | IDENTIFIER )+ "]"
+description: COMMENT+
+
+PRIMITIVE: INT_TYPE "*"? | "size" | "string" | "buffer" | "address"
+INT_TYPE: /u?int(8|16|32|64)?/
+NUMBER: /(0x)?[0-9a-fA-F]+/
+UID: /[0-9a-fA-F]{16}/
+OPTION.2: IDENTIFIER ":" IDENTIFIER
+COMMENT: /#.*/
+PATH: /"[^"]*"/
+
+%import common.LETTER
+%import common.CNAME -> IDENTIFIER
+%import common.WS
+%ignore WS

assets/initrd.toml

@@ -1,28 +0,0 @@
-# This is the manifest for the initial ramdisk, read by the `makerd` tool.
-# The contents should be a table array of files to add to the ramdistk:
-#
-#  [[files]]
-#  dest = "foo.bar"             # Name of the file in the ramdisk
-#  source = "build/foo/foo.bar" # Location of the file from the project root
-#  executable = true            # Optional, default false. Whether this is an
-#                               # initial application for the kernel to execute
-#                               # on startup
-
-[[files]]
-dest = "screenfont.psf"
-source = "../assets/fonts/tamsyn8x16r.psf"
-
-[[files]]
-dest = "symbol_table.dat"
-source = "symbol_table.dat"
-symbols = true
-
-[[files]]
-dest = "nulldrv1"
-source = "user/nulldrv"
-executable = true
-
-[[files]]
-dest = "nulldrv2"
-source = "user/nulldrv"
-executable = true

assets/jsix.svg

@@ -0,0 +1 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?><!-- Generator: Gravit.io --><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" style="isolation:isolate" viewBox="176.562 356.069 211.11 113" width="211.11pt" height="113pt"><g><g><rect x="176.562" y="356.069" width="211.11" height="113" transform="matrix(1,0,0,1,0,0)" fill="rgb(255,255,255)"/><g><path d=" M 212.981 372.36 L 219.564 376.16 L 226.147 379.961 L 226.147 387.563 L 226.147 395.164 L 219.564 398.965 L 212.981 402.766 L 206.398 398.965 L 199.815 395.164 L 199.815 387.563 L 199.815 379.961 L 206.398 376.16 L 212.981 372.36 L 212.981 372.36 L 212.981 372.36 Z  M 256.292 397.366 L 262.875 401.166 L 269.458 404.967 L 269.458 412.569 L 269.458 420.17 L 262.875 423.971 L 256.292 427.772 L 249.709 423.971 L 243.126 420.17 L 243.126 412.569 L 243.126 404.967 L 249.709 401.166 L 256.292 397.366 L 256.292 397.366 Z  M 183.622 387.283 L 205.52 374.64 L 227.418 361.997 L 249.316 374.64 L 271.214 387.283 L 271.214 412.569 L 271.214 437.854 L 249.316 450.497 L 227.418 463.14 L 205.52 450.497 L 183.622 437.854 L 183.622 412.569 L 183.622 387.283 L 183.622 387.283 L 183.622 387.283 Z  M 241.855 372.36 L 248.438 376.16 L 255.021 379.961 L 255.021 387.563 L 255.021 395.164 L 248.438 398.965 L 241.855 402.766 L 235.272 398.965 L 228.689 395.164 L 228.689 387.563 L 228.689 379.961 L 235.272 376.16 L 241.855 372.36 Z " fill-rule="evenodd" fill="rgb(49,79,128)"/><path d=" M 298.642 379.579 L 291.621 379.579 L 291.621 372.558 L 298.642 372.558 L 298.642 379.579 Z  M 285.214 446.718 L 285.214 441.452 L 287.32 441.452 L 287.32 441.452 Q 289.339 441.452 290.524 440.092 L 290.524 440.092 L 290.524 440.092 Q 291.708 438.731 291.708 436.625 L 291.708 436.625 L 291.708 387.039 L 298.729 387.039 L 298.729 436.011 L 298.729 436.011 Q 298.729 440.925 295.921 443.822 L 295.921 443.822 L 295.921 443.822 Q 293.113 446.718 288.286 446.718 L 288.286 446.718 L 285.214 446.718 Z  M 306.628 432.676 L 306.628 427.41 L 314.088 427.41 L 314.088 427.41 Q 317.862 427.41 319.573 425.347 L 319.573 425.347 L 319.573 425.347 Q 321.285 423.285 321.285 419.95 L 321.285 419.95 L 321.285 419.95 Q 321.285 417.317 319.705 415.474 L 319.705 415.474 L 319.705 415.474 Q 318.125 413.631 314.966 411.174 L 314.966 411.174 L 314.966 411.174 Q 312.245 408.98 310.621 407.356 L 310.621 407.356 L 310.621 407.356 Q 308.998 405.732 307.813 403.319 L 307.813 403.319 L 307.813 403.319 Q 306.628 400.905 306.628 397.746 L 306.628 397.746 L 306.628 397.746 Q 306.628 393.095 309.744 390.067 L 309.744 390.067 L 309.744 390.067 Q 312.859 387.039 318.125 387.039 L 318.125 387.039 L 325.76 387.039 L 325.76 392.305 L 319.441 392.305 L 319.441 392.305 Q 313.21 392.305 313.21 398.185 L 313.21 398.185 L 313.21 398.185 Q 313.21 400.467 314.615 402.134 L 314.615 402.134 L 314.615 402.134 Q 316.019 403.802 319.003 406.083 L 319.003 406.083 L 319.003 406.083 Q 321.723 408.19 323.479 409.901 L 323.479 409.901 L 323.479 409.901 Q 325.234 411.613 326.463 414.202 L 326.463 414.202 L 326.463 414.202 Q 327.691 416.791 327.691 420.301 L 327.691 420.301 L 327.691 420.301 Q 327.691 426.532 324.4 429.604 L 324.4 429.604 L 324.4 429.604 Q 321.109 432.676 315.141 432.676 L 315.141 432.676 L 306.628 432.676 Z  M 342.611 379.579 L 335.59 379.579 L 335.59 372.558 L 342.611 372.558 L 342.611 379.579 Z  M 342.611 432.676 L 335.59 432.676 L 335.59 387.039 L 342.611 387.039 L 342.611 432.676 Z  M 356.126 432.676 L 348.754 432.676 L 361.392 409.77 L 349.632 387.039 L 356.39 387.039 L 364.639 403.187 L 372.977 387.039 L 379.735 387.039 L 367.974 409.77 L 380.612 432.676 L 373.24 432.676 L 364.639 416.001 L 356.126 432.676 Z " fill="rgb(49,79,128)"/></g></g></g></svg>

assets/manifests/default.yaml

@@ -0,0 +1,16 @@
+---
+location: jsix
+init: srv.init
+initrd:
+  name: initrd.dat
+  format: zstd
+panic:
+  - panic.serial
+services:
+  - srv.logger
+  - testapp
+drivers:
+  - drv.uart
+  - drv.uefi_fb
+libs:
+  - ld.so

assets/manifests/minimal.yaml

@@ -0,0 +1,14 @@
+---
+location: jsix
+init: srv.init
+initrd:
+  name: initrd.dat
+  format: zstd
+panic:
+  - panic.serial
+services:
+  - srv.logger
+drivers:
+  - drv.uart
+libs:
+  - ld.so

assets/manifests/shell.yaml

@@ -0,0 +1,14 @@
+---
+location: jsix
+init: srv.init
+initrd:
+  name: initrd.dat
+  format: zstd
+panic:
+  - panic.serial
+services:
+  - 6s
+drivers:
+  - drv.uart
+libs:
+  - ld.so

assets/manifests/test.yaml

@@ -0,0 +1,11 @@
+---
+location: jsix
+init: srv.init
+initrd:
+  name: initrd.dat
+  format: zstd
+flags: ["test"]
+panic:
+  - panic.serial
+services:
+  - test_runner

configure

@@ -0,0 +1,82 @@
+#!/usr/bin/env python3
+
+def generate(output, config, arch, manifest):
+    from os import makedirs
+    from glob import iglob
+    from pathlib import Path
+    from bonnibel.module import Module, ModuleList
+    from bonnibel.project import Project
+
+    root = Path(__file__).parent.resolve()
+    project = Project(root)
+
+    output = root / output
+    manifest = root / manifest
+
+    sources = [
+        str(root / "src/**/*.module"),
+        str(root / "external/*.module"),
+    ]
+
+    modules = ModuleList(arch)
+    for source in sources:
+        for modfile in iglob(source, recursive=True):
+            modfile = Path(modfile)
+            path = modfile.parent
+
+            def module_init(name, **kwargs):
+                if not "root" in kwargs:
+                    kwargs["root"] = path
+                m = Module(name, modfile, **kwargs)
+                modules.add(m)
+                return m
+
+            glo = {
+                "module": module_init,
+                "source_root": root,
+                "build_root": output,
+                "module_root": path,
+                "config": config,
+                "arch": arch,
+            }
+            code = compile(open(modfile, 'r').read(), modfile, "exec")
+
+            loc = {}
+            exec(code, glo, loc)
+
+    makedirs(output.resolve(), exist_ok=True)
+    project.generate(root, output, modules, config, arch, manifest)
+    modules.generate(output)
+
+if __name__ == "__main__":
+    import sys
+    from pathlib import Path
+    sys.path.insert(0, str(Path(__file__).parent / "scripts"))
+
+    from argparse import ArgumentParser
+    from bonnibel import BonnibelError
+
+    default_arch = "amd64"
+
+    p = ArgumentParser(description="Generate jsix build files")
+    p.add_argument("--manifest", "-m", metavar="FILE", default="assets/manifests/default.yaml",
+            help="File to use as the system manifest")
+    p.add_argument("--conf", "-c", metavar="NAME", default="debug",
+            help="Configuration to build (eg, 'debug' or 'release')")
+    p.add_argument("--arch", "-a", metavar="NAME", default=default_arch,
+            help="Architecture to build (eg, 'amd64' or 'linux')")
+    p.add_argument("--verbose", "-v", action='count', default=0,
+            help="More verbose log output")
+    p.add_argument("output", metavar="DIR", default=None, nargs='?',
+            help="Where to create the build root")
+
+    args = p.parse_args()
+
+    output = args.output or f"build.{args.arch}"
+    try:
+        generate(output, args.conf, args.arch, args.manifest)
+
+    except BonnibelError as be:
+        import sys
+        print(f"Error: {be}", file=sys.stderr)
+        sys.exit(1)

definitions/memory_layout.yaml

@@ -0,0 +1,38 @@
+---
+- name: linear
+  desc: Linearly-mapped physical memory
+  size: 64T
+  shared: true
+
+- name: bitmap
+  desc: Used/free page tracking bitmap
+  size: 1T
+  shared: true
+
+- name: heapmap
+  desc: Kernel heap accounting structures
+  size: 32G
+
+- name: heap
+  desc: Kernel heap
+  size: 32G
+
+- name: capsmap
+  desc: Capabilities accounting structures
+  size: 32G
+
+- name: caps
+  desc: Capabilities
+  size: 32G
+
+- name: stacks
+  desc: Kernel thread stacks
+  size: 64G
+
+- name: buffers
+  desc: Kernel buffers
+  size: 64G
+
+- name: logs
+  desc: Kernel logs circular buffer
+  size: 2G

definitions/objects/event.def

@@ -0,0 +1,25 @@
+# An ``event`` is a simple synchronization object. It contains up to 64 signals
+# that threads can wait for and signal in parallel.
+
+object event : object {
+    uid f441e03da5516b1a
+
+    capabilities [
+        signal
+        wait
+    ]
+
+    method create [constructor]
+
+    # Signal events on this object
+    method signal [cap:signal] {
+        param signals uint64        # A bitset of which events to signal
+    }
+
+    # Wait for signaled events on this object
+    method wait [cap:wait] {
+        param signals uint64 [out]  # A bitset of which events were signaled
+        param timeout uint64        # Wait timeout in nanoseconds
+    }
+}
+

definitions/objects/mailbox.def

@@ -0,0 +1,41 @@
+# Mailboxes are objects that enable synchronous IPC via arbitrary
+# message-passing of tagged data and/or handles. Not as efficient
+# as shared memory channels, but more flexible.
+
+object mailbox : object {
+    uid 99934ad04ece1e07
+
+    capabilities [
+        send
+        receive
+        close
+    ]
+
+    method create [constructor]
+    method close [destructor cap:close]
+
+    # Send a message to the reciever, and block until a response is
+    # sent. Note that getting this response does not require the
+    # receive capability.
+    method call [cap:send] {
+        param tag uint64 [inout]
+        param data buffer [optional inout]
+        param data_size size  # number of total bytes in data buffer
+        param handles ref object [optional inout handle list]
+        param handles_size size  # total size of handles buffer
+    }
+
+    # Respond to a message sent using call, and wait for another
+    # message to arrive. Note that this does not require the send
+    # capability. A reply tag of 0 skips the reply and goes directly
+    # to waiting for a new message.
+    method respond [cap:receive] {
+        param tag uint64 [inout]
+        param data buffer [optional inout]
+        param data_size size  # number of total bytes in data buffer
+        param handles ref object [optional inout handle list]
+        param handles_size size  # total size of handles buffer
+        param reply_tag uint64 [inout]
+        param flags uint64
+    }
+}

definitions/objects/object.def

@@ -0,0 +1,16 @@
+# All kernel-exposed objects inherit from the base ``object`` type, so the
+# ``object`` syscalls can be used with any object's handle.
+
+object object [virtual] {
+    uid 667f61fb2cd57bb4
+    cname kobject
+
+    capabilities [
+        clone
+    ]
+
+    # Get the internal kernel object id of an object
+    method koid {
+        param koid uint64 [out]
+    }
+}

definitions/objects/process.def

@@ -0,0 +1,32 @@
+import "objects/object.def"
+
+# A ``process`` object represents a process running on the system, and allows
+# control over the threads, handles, and virtual memory space of that process.
+
+object process : object {
+    uid 0c69ee0b7502ba31
+
+    capabilities [
+        kill
+        create_thread
+    ]
+
+    # Create a new empty process
+    method create [constructor] {
+        param name string
+    }
+
+    # Stop all threads and exit the given process
+    method kill [destructor cap:kill]
+
+    # Stop all threads and exit the current process
+    method exit [static noreturn] {
+        param result int64    # The result to retrun to the parent process
+    }
+
+    # Give the given process a handle that points to the same
+    # object as the specified handle.
+    method give_handle {
+        param target ref object [handle]          # A handle in the caller process to send
+    }
+}

definitions/objects/system.def

@@ -0,0 +1,42 @@
+# The singular ``system`` object represents a handle to kernel functionality
+# needed by drivers and other priviledged services.
+
+object system : object {
+    uid fa72506a2cf71a30
+
+    capabilities [
+        get_log
+        bind_irq
+        map_phys
+        change_iopl
+    ]
+
+    # Get the next log line from the kernel log
+    method get_log [cap:get_log] {
+        param seen uint64                 # Last seen log id
+        param buffer buffer [out zero_ok] # Buffer for the log message data structure
+    }
+
+    # Ask the kernel to send this process messages whenever
+    # the given IRQ fires
+    method bind_irq [cap:bind_irq] {
+        param dest ref event     # Event object that will receive messages
+        param irq uint           # IRQ number to bind
+        param signal uint        # Signal number on the event to bind to
+    }
+
+    # Create a VMA and map an area of physical memory into it,
+    # also mapping that VMA into the current process
+    method map_phys [cap:map_phys] {
+        param area ref vma [out] # Receives a handle to the VMA created
+        param phys address       # The physical address of the area
+        param size size          # Size of the area, in bytes
+        param flags uint32       # Flags to apply to the created VMA
+    }
+
+    # Request the kernel change the IOPL for this process. The only values
+    # that make sense are 0 and 3.
+    method request_iopl [cap:change_iopl] {
+        param iopl uint             # The IOPL to set for this process
+    }
+}

definitions/objects/thread.def

@@ -0,0 +1,30 @@
+# A ``thread`` object represents a thread of execution within a process running
+# on the system. The actual thread does not need to be currently running to
+# hold a handle to it.
+
+object thread : object {
+    uid 11f23e593d5761bd
+
+    capabilities [
+        kill
+        join
+    ]
+
+    method create [constructor] {
+        param process ref process [optional cap:create_thread]
+        param stack_top address
+        param entrypoint address
+        param arg0 uint64
+        param arg1 uint64
+    }
+
+    method kill [destructor cap:kill]
+
+    method join [cap:join]
+
+    method exit [static]
+
+    method sleep [static] {
+        param duration uint64
+    }
+}

definitions/objects/vma.def

@@ -0,0 +1,41 @@
+import "objects/process.def"
+
+# A ``vma`` object represents a single virtual memory area, which may be shared
+# between several processes. A process having a handle to a ``vma`` does not
+# necessarily mean that it is mapped into that process' virtual memory space.
+
+object vma : object {
+    uid d6a12b63b3ed3937
+    cname vm_area
+
+    capabilities [
+        map
+        unmap
+        resize
+    ]
+
+    method create [constructor] {
+        param size size
+        param flags uint32
+    }
+
+    method create_map [constructor cap:map] {
+        param size size
+        param address address [inout]
+        param flags uint32
+    }
+
+    method map [cap:map] {
+        param process ref process [optional]
+        param address address [inout]
+        param flags uint32
+    }
+
+    method unmap [cap:unmap] {
+        param process ref process [optional]
+    }
+
+    method resize [cap:resize] {
+        param size size [inout]  # New size for the VMA, or 0 to query the current size without changing
+    }
+}

definitions/syscalls.def

@@ -0,0 +1,69 @@
+import "objects/object.def"
+
+import "objects/event.def"
+import "objects/mailbox.def"
+import "objects/process.def"
+import "objects/system.def"
+import "objects/thread.def"
+import "objects/vma.def"
+
+interface syscalls [syscall] {
+    uid 01d9b6a948961097
+
+    expose ref object
+
+    expose ref event
+    expose ref mailbox
+    expose ref process
+    expose ref system
+    expose ref thread
+    expose ref vma
+
+    # Simple no-op syscall for testing
+    function noop
+
+    # Write a message to the kernel log
+    function log {
+        param area uint8
+        param severity uint8
+        param message string
+    }
+
+    # Get a list of handles owned by this process. If the
+    # supplied list is not big enough, will set the size
+    # needed in `size` and return j6_err_insufficient
+    function handle_list {
+        param handles struct handle_descriptor [list inout zero_ok] # A list of handles to be filled
+    }
+
+    # Create a clone of an existing handle, possibly with
+    # some capabilities masked out.
+    function handle_clone {
+        param orig ref object [handle cap:clone] # The handle to clone
+        param clone ref object [out]             # The new handle
+        param mask uint32                        # The capability bitmask
+    }
+
+    # Close the handle to an object
+    function handle_close {
+        param hnd ref object [handle] # The handle to close
+    }
+
+    # Block waiting on a futex
+    function futex_wait [static] {
+        param address uint32*  # Address of the futex value
+        param current uint32   # Current value of the futex
+        param timeout uint64   # Wait timeout in nanoseconds
+    }
+
+    # Wake threads waiting on a futex
+    function futex_wake [static] {
+        param address uint32*  # Address of the futex value
+        param count uint64     # Number of threads to wake, or 0 for all
+    }
+
+    # Testing mode only: Have the kernel finish and exit QEMU with the given exit code
+    function test_finish [test] {
+        param exit_code uint32
+    }
+}

definitions/sysconf.yaml

@@ -0,0 +1,34 @@
+---
+address: 0x6a360000
+vars:
+  - name: version_major
+    section: kernel
+    type: uint8_t
+
+  - name: version_minor
+    section: kernel
+    type: uint8_t
+
+  - name: version_patch
+    section: kernel
+    type: uint16_t
+
+  - name: version_gitsha
+    section: kernel
+    type: uint32_t
+
+  - name: page_size
+    section: sys
+    type: size_t
+
+  - name: large_page_size
+    section: sys
+    type: size_t
+
+  - name: huge_page_size
+    section: sys
+    type: size_t
+
+  - name: num_cpus
+    section: sys
+    type: uint32_t

docs/Makefile

@@ -0,0 +1,23 @@
+# Minimal makefile for Sphinx documentation
+#
+
+# You can set these variables from the command line, and also
+# from the environment for the first two.
+SPHINXOPTS    ?=
+SPHINXBUILD   ?= sphinx-build
+SOURCEDIR     = .
+BUILDDIR      = _build
+ROOTDIR       = $(SOURCEDIR)/..
+
+# Put it first so that "make" without argument is like "make help".
+help:
+	@$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
+
+.PHONY: help Makefile
+
+# Catch-all target: route all unknown targets to Sphinx using the new
+# "make mode" option.  $(O) is meant as a shortcut for $(SPHINXOPTS).
+%: Makefile
+	PYTHONPATH=$(ROOTDIR)/scripts cog -r -D definitions_path=$(ROOTDIR)/definitions -c syscall_interface.rst
+	PYTHONPATH=$(ROOTDIR)/scripts cog -r -D definitions_path=$(ROOTDIR)/definitions -c kernel_memory.rst
+	@$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)

docs/_static/custom.css

@@ -0,0 +1,254 @@
+/* custom.css - jsix version */
+
+:root {
+    --background-color: #181820;
+    --link-color: #7070e0;
+    --link-hover-color: #9090ff;
+    --text-color: #3d3d3d;
+    --text-literal-color: #d26a98;
+}
+.wy-nav-side {
+  background: var(--background-color);
+}
+
+@media screen and (min-width: 1100px) {
+  .wy-nav-content-wrap {
+      background: var(--background-color);
+  }
+}
+
+a {
+  color: var(--link-color);
+}
+
+a:hover {
+  color: var(--link-hover-color);
+}
+
+a:visited {
+  color: var(--link-color);
+}
+
+.rst-content {
+  color: var(--text-color);
+}
+
+.rst-content code.literal {
+  color: var(--text-literal-color);
+}
+
+.rst-content tt.literal {
+  color: var(--text-literal-color);
+}
+
+.rst-content .note {
+  color: #003274;
+  background: #ccddf3;
+  padding: 1rem;
+  margin-bottom: 1rem;
+}
+
+.rst-content .note .admonition-title {
+  display: none;
+}
+
+.rst-content .warning {
+  color: #605000;
+  background: #fcf4cc;
+  padding: 1rem;
+  margin-bottom: 1rem;
+}
+
+.rst-content .warning .admonition-title {
+  display: none;
+}
+
+.rst-content .highlight {
+  background: #f5f5f5;
+}
+
+.wy-side-scroll {
+  background-color: var(--background-color);
+}
+
+.wy-side-nav-search {
+  background-color: var(--background-color);
+}
+
+.wy-side-nav-search input[type="text"] {
+  width: 100%;
+  border-radius: 0px;
+  padding: 6px 12px;
+  border-color: var(--background-color);
+}
+
+.wy-menu-vertical a {
+  font-size: 100%;
+  color: #d9d9d9;
+  padding-top: 0.6rem;
+  padding-bottom: 0.6rem;
+  background-color: inherit;
+}
+
+.wy-menu-vertical a:hover {
+  background-color: unset;
+  opacity: 1;
+}
+
+.wy-menu-vertical li.current > a {
+  background-color: var(--background-color);
+  color: white;
+}
+
+.wy-menu-vertical li.current > a span.toctree-expand {
+  display: block;
+  font-size: inherit;
+  line-height: inherit;
+  color: inherit;
+}
+
+.wy-menu-vertical li.current > a span.toctree-expand:before {
+  display: block;
+  font-size: inherit;
+  line-height: inherit;
+  color: inherit;
+}
+
+.wy-menu-vertical li.current > a span.toctree-expand:hover {
+  color: white;
+}
+
+.wy-menu-vertical li.current > a:hover {
+  background-color: var(--background-color);
+  color: white;
+}
+
+.wy-menu-vertical li.current > a:hover span.toctree-expand {
+  color: white;
+}
+
+.wy-menu-vertical .toctree-l1 {
+  opacity: 0.5;
+}
+
+.wy-menu-vertical .toctree-l1:hover {
+  opacity: 1;
+  background-color: inherit;
+}
+
+.wy-menu-vertical li.toctree-l1.current {
+  opacity: 1;
+  background-color: inherit;
+}
+
+.wy-menu-vertical li.toctree-l1.current > a {
+  border: 0px;
+}
+
+.wy-menu-vertical .toctree-l2:hover {
+  background-color: #566673;
+}
+
+.wy-menu-vertical li.toctree-l2.current > a {
+  background-color: #566673;
+  color: white;
+}
+
+.wy-menu-vertical li.toctree-l2.current li.toctree-l3 > a {
+  background-color: #e4e7ea;
+  color: #838383;
+}
+
+.wy-menu-vertical li.toctree-l2.current li.toctree-l3 > a:hover {
+  color: var(--text-color);
+}
+
+.wy-menu-vertical li.toctree-l2.current li.toctree-l3 > a:hover span.toctree-expand {
+  color: var(--text-color);
+}
+
+.wy-menu-vertical li.toctree-l2.current li.toctree-l3.current > a {
+  color: var(--text-color);
+}
+
+.wy-menu-vertical li.toctree-l2 a {
+  border: 0px;
+  background-color: #566673;
+  color: #d9d9d9;
+}
+
+.wy-menu-vertical li.toctree-l2 a span.toctree-expand {
+  display: block;
+  font-size: inherit;
+  line-height: inherit;
+  color: inherit;
+}
+
+.wy-menu-vertical li.toctree-l2 a span.toctree-expand:before {
+  display: block;
+  font-size: inherit;
+  line-height: inherit;
+  color: inherit;
+}
+
+.wy-menu-vertical li.toctree-l2 a span.toctree-expand:hover {
+  color: white;
+}
+
+.wy-menu-vertical li.toctree-l2 a:hover {
+  color: white;
+  background-color: #566673;
+}
+
+.wy-menu-vertical li.toctree-l2 a:hover span.toctree-expand {
+  color: white;
+}
+
+.wy-menu-vertical li.toctree-l3.current > a {
+  background-color: #e4e7ea;
+  color: #838383;
+}
+
+.wy-menu-vertical li.toctree-l3.current li.toctree-l4 > a {
+  background-color: #e4e7ea;
+  color: #838383;
+}
+
+.wy-menu-vertical li.toctree-l3.current li.toctree-l4.current > a {
+  color: var(--text-color);
+}
+
+.wy-nav-top {
+  background-color: var(--background-color);
+}
+
+.btn {
+  display: inline-block;
+  font-weight: 400;
+  line-height: 1.5;
+  color: var(--text-color);
+  text-align: center;
+  text-decoration: none;
+  vertical-align: middle;
+  cursor: pointer;
+  -webkit-user-select: none;
+  -ms-user-select: none;
+  user-select: none;
+  background-color: transparent;
+  border: 1px solid transparent;
+  padding: 0.375rem 0.75rem;
+  font-size: 1rem;
+  border-radius: 0;
+  transition: color 0.15s ease-in-out, background-color 0.15s ease-in-out, border-color 0.15s ease-in-out, box-shadow 0.15s ease-in-out;
+  box-shadow: unset;
+}
+
+.btn-neutral {
+  background: unset !important;
+  color: #838383 !important;
+}
+
+.btn-neutral:active {
+  padding: 0.375rem 0.75rem;
+  box-shadow: unset;
+}

docs/conf.py

@@ -0,0 +1,50 @@
+# Configuration file for the Sphinx documentation builder.
+#
+# For the full list of built-in configuration values, see the documentation:
+# https://www.sphinx-doc.org/en/master/usage/configuration.html
+
+# -- Project information -----------------------------------------------------
+# https://www.sphinx-doc.org/en/master/usage/configuration.html#project-information
+
+project = 'jsix'
+copyright = '2024, Justin C. Miller'
+author = 'Justin C. Miller'
+release = '0.8'
+
+# -- General configuration ---------------------------------------------------
+# https://www.sphinx-doc.org/en/master/usage/configuration.html#general-configuration
+
+extensions = ['sphinx.ext.todo']
+primary_domain = 'cpp'
+todo_include_todos = True
+
+templates_path = ['_templates']
+exclude_patterns = ['_build', 'Thumbs.db', '.DS_Store']
+
+
+
+# -- Options for HTML output -------------------------------------------------
+# https://www.sphinx-doc.org/en/master/usage/configuration.html#options-for-html-output
+
+html_theme = 'renku'
+html_title = 'jsix'
+html_logo = 'jsix_transparent.svg'
+html_static_path = ['_static']
+html_css_files = ['custom.css']
+html_theme_options = {
+    "description": "The jsix description",
+    "github_repo": "https://github.com/justinian/jsix",
+    "logo_only": True,
+    "footer_icons": [
+        {
+            "name": "GitHub",
+            "url": "https://github.com/justinian/jsix",
+            "html": """
+                <svg stroke="currentColor" fill="currentColor" stroke-width="0" viewBox="0 0 16 16">
+                    <path fill-rule="evenodd" d="M8 0C3.58 0 0 3.58 0 8c0 3.54 2.29 6.53 5.47 7.59.4.07.55-.17.55-.38 0-.19-.01-.82-.01-1.49-2.01.37-2.53-.49-2.69-.94-.09-.23-.48-.94-.82-1.13-.28-.15-.68-.52-.01-.53.63-.01 1.08.58 1.23.82.72 1.21 1.87.87 2.33.66.07-.52.28-.87.51-1.07-1.78-.2-3.64-.89-3.64-3.95 0-.87.31-1.59.82-2.15-.08-.2-.36-1.02.08-2.12 0 0 .67-.21 2.2.82.64-.18 1.32-.27 2-.27.68 0 1.36.09 2 .27 1.53-1.04 2.2-.82 2.2-.82.44 1.1.16 1.92.08 2.12.51.56.82 1.27.82 2.15 0 3.07-1.87 3.75-3.65 3.95.29.25.54.73.54 1.48 0 1.07-.01 1.93-.01 2.2 0 .21.15.46.55.38A8.013 8.013 0 0 0 16 8c0-4.42-3.58-8-8-8z"></path>
+                </svg>
+            """,
+            "class": "",
+        },
+    ],
+}

docs/index.rst

@@ -0,0 +1,81 @@
+.. jsix documentation master file
+.. |amd64| replace:: :abbr:`amd64 (aka x86_64)`
+
+The jsix Operating System
+=========================
+Introduction
+------------
+
+**jsix** is a custom multi-core x64 operating system being built from scratch,
+supporting modern [#]_ Intel or AMD CPUs, and UEFI firmware. It was initially
+created out of a desire to explore UEFI and to explore what's possible with a
+microkernel architecture on modern 64-bit architectures.
+
+Most of jsix is written in C++ (C++17, using `LLVM <https://llvm.org>`_), but
+you'll also find some assembly (in `NASM <https://nasm.us>`_ syntax) and Python
+for development tooling.
+
+jsix can be found `on GitHub <https://github.com/justinian/jsix>`_, and is
+released under the terms of the `MPL 2.0 <https://mozilla.org/MPL/2.0/>`_.
+
+.. admonition:: A note on the name
+
+    This kernel was originally named Popcorn, but I have since discovered that
+    the Popcorn Linux project is also developing a kernel with that name,
+    started around the same time as this project. So I've renamed this kernel
+    jsix as an homage to L4, xv6, and my wonderful wife.
+
+    The name jsix is always styled *jsix* or ``j6``, never capitalized.
+
+.. [#] jsix aims to support amd64 (x86_64) CPUs released in the last 10 years.
+
+Current Features
+----------------
+
+The jsix kernel is quite far along now, but the userland systems are still lacking.
+
+- Platform: |amd64|
+- UEFI bootloader
+- Multi-core & multi-tasking microkernel
+
+  - Work-stealing SMP scheduler
+  - Pluggable panic handler modules
+
+- Capability-style object-oriented syscall API
+
+  - Custom IDL for specifying and documenting syscalls
+
+- Virtual memory based on sharable Virtual Memory Area objects (VMAs)
+- Kernel API library (libj6), also provides features built on kernel primitives:
+
+  - Channels (async stream IPC) built on shared memory and futexes
+  - Ring buffers via doubly-mapped pages
+
+- Custom libc
+- Runtime dynamic linker
+- Init service
+
+  - Built-in VFS service for the initrd
+  - ELF loader
+  - Service-lookup protocol service
+
+- Userland UART driver
+- Userland UEFI framebuffer driver
+- Userland kernel log output service
+- Userland unit test runner
+- Build configuration system (bonnibel)
+
+
+.. toctree::
+   :maxdepth: 1
+   :caption: Site Contents:
+
+   syscall_interface
+   kernel_memory
+   process_initialization
+
+
+* :ref:`genindex`
+* :ref:`search`
+
+

docs/jsix_transparent.svg

@@ -0,0 +1,12 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Generator: Gravit.io -->
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" style="isolation:isolate" viewBox="176.562 356.069 211.11 113" width="211.11pt" height="113pt">
+<g>
+    <g>
+        <g>
+            <path d=" M 212.981 372.36 L 219.564 376.16 L 226.147 379.961 L 226.147 387.563 L 226.147 395.164 L 219.564 398.965 L 212.981 402.766 L 206.398 398.965 L 199.815 395.164 L 199.815 387.563 L 199.815 379.961 L 206.398 376.16 L 212.981 372.36 L 212.981 372.36 L 212.981 372.36 Z  M 256.292 397.366 L 262.875 401.166 L 269.458 404.967 L 269.458 412.569 L 269.458 420.17 L 262.875 423.971 L 256.292 427.772 L 249.709 423.971 L 243.126 420.17 L 243.126 412.569 L 243.126 404.967 L 249.709 401.166 L 256.292 397.366 L 256.292 397.366 Z  M 183.622 387.283 L 205.52 374.64 L 227.418 361.997 L 249.316 374.64 L 271.214 387.283 L 271.214 412.569 L 271.214 437.854 L 249.316 450.497 L 227.418 463.14 L 205.52 450.497 L 183.622 437.854 L 183.622 412.569 L 183.622 387.283 L 183.622 387.283 L 183.622 387.283 Z  M 241.855 372.36 L 248.438 376.16 L 255.021 379.961 L 255.021 387.563 L 255.021 395.164 L 248.438 398.965 L 241.855 402.766 L 235.272 398.965 L 228.689 395.164 L 228.689 387.563 L 228.689 379.961 L 235.272 376.16 L 241.855 372.36 Z " fill-rule="evenodd" fill="rgb(49,79,128)"/>
+            <path d=" M 298.642 379.579 L 291.621 379.579 L 291.621 372.558 L 298.642 372.558 L 298.642 379.579 Z  M 285.214 446.718 L 285.214 441.452 L 287.32 441.452 L 287.32 441.452 Q 289.339 441.452 290.524 440.092 L 290.524 440.092 L 290.524 440.092 Q 291.708 438.731 291.708 436.625 L 291.708 436.625 L 291.708 387.039 L 298.729 387.039 L 298.729 436.011 L 298.729 436.011 Q 298.729 440.925 295.921 443.822 L 295.921 443.822 L 295.921 443.822 Q 293.113 446.718 288.286 446.718 L 288.286 446.718 L 285.214 446.718 Z  M 306.628 432.676 L 306.628 427.41 L 314.088 427.41 L 314.088 427.41 Q 317.862 427.41 319.573 425.347 L 319.573 425.347 L 319.573 425.347 Q 321.285 423.285 321.285 419.95 L 321.285 419.95 L 321.285 419.95 Q 321.285 417.317 319.705 415.474 L 319.705 415.474 L 319.705 415.474 Q 318.125 413.631 314.966 411.174 L 314.966 411.174 L 314.966 411.174 Q 312.245 408.98 310.621 407.356 L 310.621 407.356 L 310.621 407.356 Q 308.998 405.732 307.813 403.319 L 307.813 403.319 L 307.813 403.319 Q 306.628 400.905 306.628 397.746 L 306.628 397.746 L 306.628 397.746 Q 306.628 393.095 309.744 390.067 L 309.744 390.067 L 309.744 390.067 Q 312.859 387.039 318.125 387.039 L 318.125 387.039 L 325.76 387.039 L 325.76 392.305 L 319.441 392.305 L 319.441 392.305 Q 313.21 392.305 313.21 398.185 L 313.21 398.185 L 313.21 398.185 Q 313.21 400.467 314.615 402.134 L 314.615 402.134 L 314.615 402.134 Q 316.019 403.802 319.003 406.083 L 319.003 406.083 L 319.003 406.083 Q 321.723 408.19 323.479 409.901 L 323.479 409.901 L 323.479 409.901 Q 325.234 411.613 326.463 414.202 L 326.463 414.202 L 326.463 414.202 Q 327.691 416.791 327.691 420.301 L 327.691 420.301 L 327.691 420.301 Q 327.691 426.532 324.4 429.604 L 324.4 429.604 L 324.4 429.604 Q 321.109 432.676 315.141 432.676 L 315.141 432.676 L 306.628 432.676 Z  M 342.611 379.579 L 335.59 379.579 L 335.59 372.558 L 342.611 372.558 L 342.611 379.579 Z  M 342.611 432.676 L 335.59 432.676 L 335.59 387.039 L 342.611 387.039 L 342.611 432.676 Z  M 356.126 432.676 L 348.754 432.676 L 361.392 409.77 L 349.632 387.039 L 356.39 387.039 L 364.639 403.187 L 372.977 387.039 L 379.735 387.039 L 367.974 409.77 L 380.612 432.676 L 373.24 432.676 L 364.639 416.001 L 356.126 432.676 Z " fill="rgb(49,79,128)"/>
+        </g>
+    </g>
+</g>
+</svg>

docs/kernel_memory.rst

@@ -0,0 +1,178 @@
+.. jsix syscall interface.
+.. Automatically updated from the definition files using cog!
+
+.. [[[cog code generation
+.. from os.path import join
+.. from memory import Layout, unit
+..
+.. layout = Layout(join(definitions_path, "memory_layout.yaml"))
+.. l = max([len(r.name) for r in layout.regions])
+.. ]]]
+.. [[[end]]] (checksum: d41d8cd98f00b204e9800998ecf8427e)
+
+Kernel memory
+=============
+
+While jsix probably should eventually use KASLR to randomize its memory layout,
+currently the layout is mostly fixed. (Kernel code locations are not consistent
+but aren't explicitly randomized.)
+
+.. [[[cog code generation
+.. line_size = 128 * 1024**3  # Each line represents up to 32 GiB
+.. max_lines = 32
+.. totals = sum([r.size for r in layout.regions])
+.. remain = unit((128 * 1024**4) - totals)
+..
+.. def split(val):
+..    return f"0x {val >> 48:04x} {(val >> 32) & 0xffff:04x} {(val >> 16) & 0xffff:04x} {val & 0xffff:04x}"
+..
+.. cog.outl()
+.. cog.outl(f"+-+-----------------------------+----------+---------------------------------------+")
+.. cog.outl(f"| | Address                     | Size     | Use                                   |")
+.. cog.outl(f"+=+=============================+==========+=======================================+")
+..
+.. for region in layout.regions:
+..     cog.outl(f"| | ``{split(region.start)}``  | {unit(region.size):>8} | {region.desc:37} |")
+..     lines = min(max_lines, region.size // line_size)
+..     for i in range(1, lines):
+..         cog.outl(f"+-+                             |          |                                       |")
+..         cog.outl(f"| |                             |          |                                       |")
+..     cog.outl(f"+-+-----------------------------+----------+---------------------------------------+")
+..
+.. cog.outl(f"| |  ...                        |          |                                       |")
+.. cog.outl(f"+-+-----------------------------+----------+---------------------------------------+")
+.. cog.outl(f"| |  ``0x ffff 0000 0000 0000`` |          | Kernel code / headers                 |")
+.. cog.outl(f"+-+-----------------------------+----------+---------------------------------------+")
+.. cog.outl("")
+.. cog.outl("")
+.. cog.outl(f"Un-reserved virtual memory address space in the higher half: {remain}")
+.. cog.outl("")
+..
+.. ]]]
+
++-+-----------------------------+----------+---------------------------------------+
+| | Address                     | Size     | Use                                   |
++=+=============================+==========+=======================================+
+| | ``0x ffff c000 0000 0000``  |   64 TiB | Linearly-mapped physical memory       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+-----------------------------+----------+---------------------------------------+
+| | ``0x ffff bf00 0000 0000``  |    1 TiB | Used/free page tracking bitmap        |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+-----------------------------+----------+---------------------------------------+
+| | ``0x ffff be00 0000 0000``  |    1 TiB | Per-page state tracking structures    |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+                             |          |                                       |
+| |                             |          |                                       |
++-+-----------------------------+----------+---------------------------------------+
+| | ``0x ffff bdf8 0000 0000``  |   32 GiB | Kernel heap accounting structures     |
++-+-----------------------------+----------+---------------------------------------+
+| | ``0x ffff bdf0 0000 0000``  |   32 GiB | Kernel heap                           |
++-+-----------------------------+----------+---------------------------------------+
+| | ``0x ffff bde8 0000 0000``  |   32 GiB | Capabilities accounting structures    |
++-+-----------------------------+----------+---------------------------------------+
+| | ``0x ffff bde0 0000 0000``  |   32 GiB | Capabilities                          |
++-+-----------------------------+----------+---------------------------------------+
+| | ``0x ffff bdd0 0000 0000``  |   64 GiB | Kernel thread stacks                  |
++-+-----------------------------+----------+---------------------------------------+
+| | ``0x ffff bdc0 0000 0000``  |   64 GiB | Kernel buffers                        |
++-+-----------------------------+----------+---------------------------------------+
+| | ``0x ffff bdbf 8000 0000``  |    2 GiB | Kernel logs circular buffer           |
++-+-----------------------------+----------+---------------------------------------+
+| |  ...                        |          |                                       |
++-+-----------------------------+----------+---------------------------------------+
+| |  ``0x ffff 0000 0000 0000`` |          | Kernel code / headers                 |
++-+-----------------------------+----------+---------------------------------------+
+
+
+Un-reserved virtual memory address space in the higher half: 61 TiB
+
+.. [[[end]]] (checksum: 8c336cc8151beba1a79c8d3b653f1109)
+
+* :ref:`genindex`
+* :ref:`search`
+

docs/modd.conf

@@ -0,0 +1,7 @@
+** !_build/** ../definitions/**.def {
+    prep: rm -rf _build; make html
+}
+
+_build/html/** {
+    daemon: devd -m _build/html
+}

docs/process_initialization.rst

@@ -0,0 +1,42 @@
+.. jsix process initialization in userspace
+
+Process Initialization
+======================
+
+jsix follows the `System V ABI`_ on the ``amd64`` architecture. All arguments
+needed for program initialization are passed to the program's initial thread on
+the stack.
+
+Note that jsix adds a number of additional auxiliary vector entry types for
+passing jsix-specific data to a program. The jsix-specific auxiliary vector type
+codes (what the ABI document refers to as ``a_type``) start from ``0xf000``. See
+the header file ``<j6/init.h>`` for more detail.
+
+.. _System V ABI: https://gitlab.com/x86-psABIs/x86-64-ABI
+
+The initial stack frame
+-----------------------
+
+==============  ====================  ============  =======
+Address         Value                 Bytes         Notes
+==============  ====================  ============  =======
+``top``                                             Stack top (out of stack bounds)
+``top`` - 16    0                     16            Stack sentinel
+\               ``envp`` string data  ?
+\               ``argv`` string data  ?
+\               ...                   ?             Possible padding
+\               0, 0 (``AT_NULL``)    16            Aux vector sentinel
+\               Aux vectors           16 * `m`      ``AT_NULL``-terminated array of Aux vectors
+\               0                     8             Environment sentinel
+\                ``envp``             8 * `n`       0-terminated array of environment
+                                                    string pointers
+\               0                     8             Args sentinel
+\               ``argv``              8 * ``argc``  Pointers to argument strings
+``rsp``         ``argc``              8             Number of elements in argv
+==============  ====================  ============  =======
+
+
+* :ref:`genindex`
+* :ref:`search`
+
+

docs/requirements.txt

@@ -0,0 +1,5 @@
+cogapp >= 3
+pyyaml >= 5.4
+lark == 0.12.0
+sphinx
+renku-sphinx-theme

docs/syscall_interface.rst

@@ -0,0 +1,444 @@
+.. jsix syscall interface.
+.. Automatically updated from the definition files using cog!
+
+.. [[[cog code generation
+.. from textwrap import indent
+.. from definitions.context import Context
+.. 
+.. ctx = Context(definitions_path)
+.. ctx.parse("syscalls.def")
+.. syscalls = ctx.interfaces["syscalls"]
+.. 
+.. def caplist(caps):
+..   return ', '.join([f"``{c}``" for c in caps])
+.. ]]]
+.. [[[end]]] (checksum: d41d8cd98f00b204e9800998ecf8427e)
+
+Syscall interface
+=================
+
+The jsix kernel's syscall design is based around object handles. Object handles
+are also a collection of capabilities, encoding certain rights over the object
+they reference.
+
+Very few syscalls in jsix can be made without some handle, and most of them are
+requests to the kernel to create a given kind of object. This is analogous to
+methods on an object in an object-oriented programming language.
+
+.. [[[cog code generation
+.. cog.outl()
+.. for obj in syscalls.exposes:
+..     cog.outl(f"``{obj.name}`` syscalls")
+..     cog.outl(f"-------------------------")
+..     desc = obj.desc or "Undocumented"
+..     cog.outl(desc)
+..     cog.outl()
+..     cog.outl(f":capabilites:  {caplist(obj.caps)}")
+..     cog.outl()
+..     for method in obj.methods:
+..       args = []
+..       if method.constructor:
+..         args.append("j6_handle_t *self")
+..       elif not method.static:
+..         args.append("j6_handle_t self")
+..
+..       for param in method.params:
+..         for type, suffix in param.type.c_names(param.options):
+..           args.append(f"{type} {param.name}{suffix}")
+..
+..       cog.outl(f".. cpp:function:: j6_result_t j6_{obj.name}_{method.name} ({', '.join(args)})")
+..       cog.outl()
+..       desc = method.desc or "Undocumented"
+..       cog.outl(indent(desc, "   "))
+..       cog.outl()
+..       if "cap" in method.options:
+..         cog.outl(f"   :capabilities: {caplist(method.options['cap'])}")
+..         cog.outl()
+..       if method.constructor:
+..         cog.outl(f"   :param self: *[out]* Handle to the new {obj.name} object")
+..       elif not method.static:
+..         cog.outl(f"   :param self: Handle to the {obj.name} object")
+..       for param in method.params:
+..         opts = param.options and f"*[{', '.join(param.options)}]*" or ""
+..         desc = param.desc or 'Undocumented'
+..         cog.outl(f"   :param {param.name}: {opts} {desc}")
+..       cog.outl()
+.. ]]]
+
+``object`` syscalls
+-------------------------
+All kernel-exposed objects inherit from the base ``object`` type, so the
+``object`` syscalls can be used with any object's handle.
+
+:capabilites:  ``clone``
+
+.. cpp:function:: j6_result_t j6_object_koid (j6_handle_t self, uint64_t * koid)
+
+   Get the internal kernel object id of an object
+
+   :param self: Handle to the object object
+   :param koid: *[out]* Undocumented
+
+``event`` syscalls
+-------------------------
+An ``event`` is a simple synchronization object. It contains up to 64 signals
+that threads can wait for and signal in parallel.
+
+:capabilites:  ``signal``, ``wait``
+
+.. cpp:function:: j6_result_t j6_event_create (j6_handle_t *self)
+
+   Undocumented
+
+   :param self: *[out]* Handle to the new event object
+
+.. cpp:function:: j6_result_t j6_event_signal (j6_handle_t self, uint64_t signals)
+
+   Signal events on this object
+
+   :capabilities: ``signal``
+
+   :param self: Handle to the event object
+   :param signals:  A bitset of which events to signal
+
+.. cpp:function:: j6_result_t j6_event_wait (j6_handle_t self, uint64_t * signals, uint64_t timeout)
+
+   Wait for signaled events on this object
+
+   :capabilities: ``wait``
+
+   :param self: Handle to the event object
+   :param signals: *[out]* A bitset of which events were signaled
+   :param timeout:  Wait timeout in nanoseconds
+
+``mailbox`` syscalls
+-------------------------
+Mailboxes are objects that enable synchronous IPC via arbitrary
+message-passing of tagged data and/or handles. Not as efficient
+as shared memory channels, but more flexible.
+
+:capabilites:  ``send``, ``receive``, ``close``
+
+.. cpp:function:: j6_result_t j6_mailbox_create (j6_handle_t *self)
+
+   Undocumented
+
+   :param self: *[out]* Handle to the new mailbox object
+
+.. cpp:function:: j6_result_t j6_mailbox_close (j6_handle_t self)
+
+   Undocumented
+
+   :capabilities: ``close``
+
+   :param self: Handle to the mailbox object
+
+.. cpp:function:: j6_result_t j6_mailbox_call (j6_handle_t self, uint64_t * tag, void * data, size_t * data_len, size_t data_size, j6_handle_t * handles, size_t * handles_count, size_t handles_size)
+
+   Send a message to the reciever, and block until a response is
+   sent. Note that getting this response does not require the
+   receive capability.
+
+   :capabilities: ``send``
+
+   :param self: Handle to the mailbox object
+   :param tag: *[inout]* Undocumented
+   :param data: *[optional, inout]* Undocumented
+   :param data_size:  number of total bytes in data buffer
+   :param handles: *[optional, inout, handle, list]* Undocumented
+   :param handles_size:  total size of handles buffer
+
+.. cpp:function:: j6_result_t j6_mailbox_respond (j6_handle_t self, uint64_t * tag, void * data, size_t * data_len, size_t data_size, j6_handle_t * handles, size_t * handles_count, size_t handles_size, uint64_t * reply_tag, uint64_t flags)
+
+   Respond to a message sent using call, and wait for another
+   message to arrive. Note that this does not require the send
+   capability. A reply tag of 0 skips the reply and goes directly
+   to waiting for a new message.
+
+   :capabilities: ``receive``
+
+   :param self: Handle to the mailbox object
+   :param tag: *[inout]* Undocumented
+   :param data: *[optional, inout]* Undocumented
+   :param data_size:  number of total bytes in data buffer
+   :param handles: *[optional, inout, handle, list]* Undocumented
+   :param handles_size:  total size of handles buffer
+   :param reply_tag: *[inout]* Undocumented
+   :param flags:  Undocumented
+
+``process`` syscalls
+-------------------------
+A ``process`` object represents a process running on the system, and allows
+control over the threads, handles, and virtual memory space of that process.
+
+:capabilites:  ``kill``, ``create_thread``
+
+.. cpp:function:: j6_result_t j6_process_create (j6_handle_t *self)
+
+   Create a new empty process
+
+   :param self: *[out]* Handle to the new process object
+
+.. cpp:function:: j6_result_t j6_process_kill (j6_handle_t self)
+
+   Stop all threads and exit the given process
+
+   :capabilities: ``kill``
+
+   :param self: Handle to the process object
+
+.. cpp:function:: j6_result_t j6_process_exit (int64_t result)
+
+   Stop all threads and exit the current process
+
+   :param result:  The result to retrun to the parent process
+
+.. cpp:function:: j6_result_t j6_process_give_handle (j6_handle_t self, j6_handle_t target)
+
+   Give the given process a handle that points to the same
+   object as the specified handle.
+
+   :param self: Handle to the process object
+   :param target: *[handle]* A handle in the caller process to send
+
+``system`` syscalls
+-------------------------
+The singular ``system`` object represents a handle to kernel functionality
+needed by drivers and other priviledged services.
+
+:capabilites:  ``get_log``, ``bind_irq``, ``map_phys``, ``change_iopl``
+
+.. cpp:function:: j6_result_t j6_system_get_log (j6_handle_t self, uint64_t seen, void * buffer, size_t * buffer_len)
+
+   Get the next log line from the kernel log
+
+   :capabilities: ``get_log``
+
+   :param self: Handle to the system object
+   :param seen:  Last seen log id
+   :param buffer: *[out, zero_ok]* Buffer for the log message data structure
+
+.. cpp:function:: j6_result_t j6_system_bind_irq (j6_handle_t self, j6_handle_t dest, unsigned irq, unsigned signal)
+
+   Ask the kernel to send this process messages whenever
+   the given IRQ fires
+
+   :capabilities: ``bind_irq``
+
+   :param self: Handle to the system object
+   :param dest:  Event object that will receive messages
+   :param irq:  IRQ number to bind
+   :param signal:  Signal number on the event to bind to
+
+.. cpp:function:: j6_result_t j6_system_map_phys (j6_handle_t self, j6_handle_t * area, uintptr_t phys, size_t size, uint32_t flags)
+
+   Create a VMA and map an area of physical memory into it,
+   also mapping that VMA into the current process
+
+   :capabilities: ``map_phys``
+
+   :param self: Handle to the system object
+   :param area: *[out]* Receives a handle to the VMA created
+   :param phys:  The physical address of the area
+   :param size:  Size of the area, in bytes
+   :param flags:  Flags to apply to the created VMA
+
+.. cpp:function:: j6_result_t j6_system_request_iopl (j6_handle_t self, unsigned iopl)
+
+   Request the kernel change the IOPL for this process. The only values
+   that make sense are 0 and 3.
+
+   :capabilities: ``change_iopl``
+
+   :param self: Handle to the system object
+   :param iopl:  The IOPL to set for this process
+
+``thread`` syscalls
+-------------------------
+A ``thread`` object represents a thread of execution within a process running
+on the system. The actual thread does not need to be currently running to
+hold a handle to it.
+
+:capabilites:  ``kill``, ``join``
+
+.. cpp:function:: j6_result_t j6_thread_create (j6_handle_t *self, j6_handle_t process, uintptr_t stack_top, uintptr_t entrypoint, uint64_t arg0, uint64_t arg1)
+
+   Undocumented
+
+   :param self: *[out]* Handle to the new thread object
+   :param process: *[optional, cap]* Undocumented
+   :param stack_top:  Undocumented
+   :param entrypoint:  Undocumented
+   :param arg0:  Undocumented
+   :param arg1:  Undocumented
+
+.. cpp:function:: j6_result_t j6_thread_kill (j6_handle_t self)
+
+   Undocumented
+
+   :capabilities: ``kill``
+
+   :param self: Handle to the thread object
+
+.. cpp:function:: j6_result_t j6_thread_join (j6_handle_t self)
+
+   Undocumented
+
+   :capabilities: ``join``
+
+   :param self: Handle to the thread object
+
+.. cpp:function:: j6_result_t j6_thread_exit ()
+
+   Undocumented
+
+
+.. cpp:function:: j6_result_t j6_thread_sleep (uint64_t duration)
+
+   Undocumented
+
+   :param duration:  Undocumented
+
+``vma`` syscalls
+-------------------------
+A ``vma`` object represents a single virtual memory area, which may be shared
+between several processes. A process having a handle to a ``vma`` does not
+necessarily mean that it is mapped into that process' virtual memory space.
+
+:capabilites:  ``map``, ``unmap``, ``resize``
+
+.. cpp:function:: j6_result_t j6_vma_create (j6_handle_t *self, size_t size, uint32_t flags)
+
+   Undocumented
+
+   :param self: *[out]* Handle to the new vma object
+   :param size:  Undocumented
+   :param flags:  Undocumented
+
+.. cpp:function:: j6_result_t j6_vma_create_map (j6_handle_t *self, size_t size, uintptr_t * address, uint32_t flags)
+
+   Undocumented
+
+   :capabilities: ``map``
+
+   :param self: *[out]* Handle to the new vma object
+   :param size:  Undocumented
+   :param address: *[inout]* Undocumented
+   :param flags:  Undocumented
+
+.. cpp:function:: j6_result_t j6_vma_map (j6_handle_t self, j6_handle_t process, uintptr_t * address, uint32_t flags)
+
+   Undocumented
+
+   :capabilities: ``map``
+
+   :param self: Handle to the vma object
+   :param process: *[optional]* Undocumented
+   :param address: *[inout]* Undocumented
+   :param flags:  Undocumented
+
+.. cpp:function:: j6_result_t j6_vma_unmap (j6_handle_t self, j6_handle_t process)
+
+   Undocumented
+
+   :capabilities: ``unmap``
+
+   :param self: Handle to the vma object
+   :param process: *[optional]* Undocumented
+
+.. cpp:function:: j6_result_t j6_vma_resize (j6_handle_t self, size_t * size)
+
+   Undocumented
+
+   :capabilities: ``resize``
+
+   :param self: Handle to the vma object
+   :param size: *[inout]* New size for the VMA, or 0 to query the current size without changing
+
+.. [[[end]]] (checksum: cb17f54e443d1d3b85995870f3e8dbf2)
+
+Non-object syscalls
+-------------------
+
+The following are the system calls that aren't constructors for objects, and
+either do not require an object handle, or operate generically on handles.
+
+.. [[[cog code generation
+.. cog.outl()
+.. for func in syscalls.functions:
+..   args = []
+..   for param in func.params:
+..     for type, suffix in param.type.c_names(param.options):
+..       args.append(f"{type} {param.name}{suffix}")
+..
+..   cog.outl(f".. cpp:function:: j6_result_t j6_{func.name} ({', '.join(args)})")
+..   cog.outl()
+..   desc = func.desc or "Undocumented"
+..   cog.outl(indent(desc, "   "))
+..   cog.outl()
+..   for param in func.params:
+..     opts = param.options and f"*[{', '.join(param.options)}]*" or ""
+..     desc = param.desc or 'Undocumented'
+..     cog.outl(f"   :param {param.name}: {opts} {desc}")
+..   cog.outl()
+.. ]]]
+
+.. cpp:function:: j6_result_t j6_noop ()
+
+   Simple no-op syscall for testing
+
+
+.. cpp:function:: j6_result_t j6_log (uint8_t area, uint8_t severity, const char * message)
+
+   Write a message to the kernel log
+
+   :param area:  Undocumented
+   :param severity:  Undocumented
+   :param message:  Undocumented
+
+.. cpp:function:: j6_result_t j6_handle_list (struct j6_handle_descriptor * handles, size_t * handles_size)
+
+   Get a list of handles owned by this process. If the
+   supplied list is not big enough, will set the size
+   needed in `size` and return j6_err_insufficient
+
+   :param handles: *[list, inout, zero_ok]* A list of handles to be filled
+
+.. cpp:function:: j6_result_t j6_handle_clone (j6_handle_t orig, j6_handle_t * clone, uint32_t mask)
+
+   Create a clone of an existing handle, possibly with
+   some capabilities masked out.
+
+   :param orig: *[handle, cap]* The handle to clone
+   :param clone: *[out]* The new handle
+   :param mask:  The capability bitmask
+
+.. cpp:function:: j6_result_t j6_handle_close (j6_handle_t hnd)
+
+   Close the handle to an object
+
+   :param hnd: *[handle]* The handle to close
+
+.. cpp:function:: j6_result_t j6_futex_wait (const uint32_t * address, uint32_t current, uint64_t timeout)
+
+   Block waiting on a futex
+
+   :param address:  Address of the futex value
+   :param current:  Current value of the futex
+   :param timeout:  Wait timeout in nanoseconds
+
+.. cpp:function:: j6_result_t j6_futex_wake (const uint32_t * address, uint64_t count)
+
+   Wake threads waiting on a futex
+
+   :param address:  Address of the futex value
+   :param count:  Number of threads to wake, or 0 for all
+
+.. cpp:function:: j6_result_t j6_test_finish (uint32_t exit_code)
+
+   Testing mode only: Have the kernel finish and exit QEMU with the given exit code
+
+   :param exit_code:  Undocumented
+
+.. [[[end]]] (checksum: 0b9d051972abcbb6de408f411331785f)
+

external/uefi/boot_services.h

@@ -14,13 +14,15 @@
 namespace uefi {
 namespace bs_impl {
 	using allocate_pages = status (*)(allocate_type, memory_type, size_t, void**);
+	using free_pages = status (*)(void*, size_t);
 	using get_memory_map = status (*)(size_t*, memory_descriptor*, size_t*, size_t*, uint32_t*);
 	using allocate_pool = status (*)(memory_type, uint64_t, void**);
+	using free_pool = status (*)(void*);
 	using handle_protocol = status (*)(handle, const guid*, void**);
 	using create_event = status (*)(evt, tpl, event_notify, void*, event*);
 	using exit_boot_services = status (*)(handle, size_t);
 	using locate_protocol = status (*)(const guid*, void*, void**);
-	using copy_mem = void (*)(void*, void*, size_t);
+	using copy_mem = void (*)(void*, const void*, size_t);
 	using set_mem = void (*)(void*, uint64_t, uint8_t);
 }
 
@@ -35,10 +37,10 @@ struct boot_services {
 
 	// Memory Services
 	bs_impl::allocate_pages allocate_pages;
-	void *free_pages;
+	bs_impl::free_pages free_pages;
 	bs_impl::get_memory_map get_memory_map;
 	bs_impl::allocate_pool allocate_pool;
-	void *free_pool;
+	bs_impl::free_pool free_pool;
 
 	// Event & Timer Services
 	bs_impl::create_event create_event;

external/uefi/networking.h

@@ -0,0 +1,390 @@
+#pragma once
+#ifndef _uefi_networking_h_
+#define _uefi_networking_h_
+
+// This Source Code Form is part of the j6-uefi-headers project and is subject
+// to the terms of the Mozilla Public License, v. 2.0. If a copy of the MPL was
+// not distributed with this file, You can obtain one at
+// http://mozilla.org/MPL/2.0/.
+
+#include <stdint.h>
+#include <uefi/types.h>
+
+namespace uefi {
+
+//
+// IPv4 definitions
+//
+struct ipv4_address
+{
+	uint8_t addr[4];
+};
+
+//
+// IPv6 definitions
+//
+struct ipv6_address
+{
+	uint8_t addr[16];
+};
+
+struct ip6_address_info
+{
+    ipv6_address address;
+    uint8_t prefix_length;
+};
+
+struct ip6_route_table
+{
+    ipv6_address gateway;
+    ipv6_address destination;
+    uint8_t prefix_length;
+};
+
+enum class ip6_neighbor_state : int {
+    incomplete,
+    reachable,
+    stale,
+    delay,
+    probe,
+}
+
+struct ip6_neighbor_cache
+{
+    ipv6_address neighbor;
+    mac_address link_address;
+    ip6_neighbor_state state;
+};
+
+enum class icmpv6_type : uint8_t
+{
+    dest_unreachable = 0x1,
+    packet_too_big = 0x2,
+    time_exceeded = 0x3,
+    parameter_problem = 0x4,
+    echo_request = 0x80,
+    echo_reply = 0x81,
+    listener_query = 0x82,
+    listener_report = 0x83,
+    listener_done = 0x84,
+    router_solicit = 0x85,
+    router_advertise = 0x86,
+    neighbor_solicit = 0x87,
+    neighbor_advertise = 0x88,
+    redirect = 0x89,
+    listener_report_2 = 0x8f,
+};
+
+enum class icmpv6_code : uint8_t
+{
+    // codes for icmpv6_type::dest_unreachable
+    no_route_to_dest = 0x0,
+    comm_prohibited = 0x1,
+    beyond_scope = 0x2,
+    addr_unreachable = 0x3,
+    port_unreachable = 0x4,
+    source_addr_failed = 0x5,
+    route_rejected = 0x6,
+
+    // codes for icmpv6_type::time_exceeded
+    timeout_hop_limit = 0x0,
+    timeout_reassemble = 0x1,
+
+    // codes for icmpv6_type::parameter_problem
+    erroneous_header = 0x0,
+    unrecognize_next_hdr = 0x1,
+    unrecognize_option = 0x2,
+};
+
+struct ip6_icmp_type
+{
+    icmpv6_type type;
+    icmpv6_code code;
+};
+
+struct ip6_config_data
+{
+    uint8_t default_protocol;
+    bool accept_any_protocol;
+    bool accept_icmp_errors;
+    bool accept_promiscuous;
+    ipv6_address destination_address;
+    ipv6_address station_address;
+    uint8_t traffic_class;
+    uint8_t hop_limit;
+    uint32_t flow_label;
+    uint32_t receive_timeout;
+    uint32_t transmit_timeout;
+};
+
+struct ip6_mode_data
+{
+    bool is_started;
+    uint32_t max_packet_size;
+    ip6_config_data config_data;
+    bool is_configured;
+    uint32_t address_count;
+    ip6_address_info * address_list;
+    uint32_t group_count;
+    ipv6_address * group_table;
+    uint32_t route_count;
+    ip6_route_table * route_table;
+    uint32_t neighbor_count;
+    ip6_neighbor_cache * neighbor_cache;
+    uint32_t prefix_count;
+    ip6_address_info * prefix_table;
+    uint32_t icmp_type_count;
+     * icmp_type_list;
+};
+
+struct ip6_header
+{
+    uint8_t traffic_class_h : 4;
+    uint8_t version : 4;
+    uint8_t flow_label_h : 4;
+    uint8_t traffic_class_l : 4;
+    uint16_t flow_label_l;
+    uint16_t payload_length;
+    uint8_t next_header;
+    uint8_t hop_limit;
+    ipv6_address source_address;
+    ipv6_address destination_address;
+} __attribute__ ((packed));
+
+struct ip6_fragment_data
+{
+    uint32_t fragment_length;
+    void *fragment_buffer;
+};
+
+struct ip6_override_data
+{
+    uint8_t protocol;
+    uint8_t hop_limit;
+    uint32_t flow_label;
+};
+
+struct ip6_receive_data
+{
+    time time_stamp;
+    event recycle_signal;
+    uint32_t header_length;
+    ip6_header *header;
+    uint32_t data_length;
+    uint32_t fragment_count;
+    ip6_fragment_data fragment_table[1];
+};
+
+struct ip6_transmit_data
+{
+    ipv6_address destination_address;
+    ip6_override_data *override_data;
+
+    uint32_t ext_hdrs_length;
+    void *ext_hdrs;
+    uint8_t next_header;
+    uint32_t data_length;
+    uint32_t fragment_count;
+    ip6_fragment_data fragment_table[1];
+};
+
+struct ip6_completion_token
+{
+    event event;
+    status status;
+    union {
+        ip6_receive_data *rx_data;
+        ip6_transmit_data *tx_data;
+    } packet;
+};
+
+enum class ip6_config_data_type : int
+{
+    interface_info,
+    alt_interface_id,
+    policy,
+    dup_addr_detect_transmits,
+    manual_address,
+    gateway,
+    dns_server,
+    maximum
+};
+
+struct ip6_config_interface_info
+{
+    wchar_t name[32];
+    uint8_t if_type;
+    uint32_t hw_address_size;
+    mac_address hw_address;
+    uint32_t address_info_count;
+    ip6_address_info *address_info;
+    uint32_t route_count;
+    ip6_route_table *route_table;
+};
+
+struct ip6_config_interface_id
+{
+    uint8_t id[8];
+};
+
+enum class ip6_config_policy : int
+{
+    manual,
+    automatic
+};
+
+struct ip6_config_dup_addr_detect_transmits
+{
+    uint32_t dup_addr_detect_transmits;
+};
+
+struct ip6_config_manual_address
+{
+    ipv6_address address;
+    bool is_anycast;
+    uint8_t prefix_length;
+};
+
+//
+// IP definitions
+//
+union ip_address
+{
+	uint8_t addr[4];
+	ipv4_address v4;
+	ipv6_address v6;
+};
+
+//
+// HTTP definitions
+//
+struct httpv4_access_point
+{
+	bool use_default_address;
+	ipv4_address local_address;
+	ipv4_address local_subnet;
+	uint16_t local_port;
+};
+
+struct httpv6_access_point
+{
+	ipv6_address local_address;
+	uint16_t local_port;
+};
+
+enum class http_version : int {
+	v10,
+	v11,
+	unsupported,
+};
+
+struct http_config_data
+{
+	http_version http_version;
+	uint32_t time_out_millisec;
+	bool local_address_is_ipv6;
+	union {
+		httpv4_access_point *ipv4_node;
+		httpv6_access_point *ipv6_node;
+	} access_point;
+};
+
+enum class http_method : int {
+	get,
+	post,
+	patch,
+	options,
+	connect,
+	head,
+	put,
+	delete_,
+	trace,
+};
+
+struct http_request_data
+{
+	http_method method;
+	wchar_t *url;
+};
+
+enum class http_status_code : int {
+	unsupported,
+	continue_,
+	switching_protocols,
+	ok,
+	created,
+	accepted,
+	non_authoritative_information,
+	no_content,
+	reset_content,
+	partial_content,
+	multiple_choices,
+	moved_permanently,
+	found,
+	see_other,
+	not_modified,
+	use_proxy,
+	temporary_redirect,
+	bad_request,
+	unauthorized,
+	payment_required,
+	forbidden,
+	not_found,
+	method_not_allowed,
+	not_acceptable,
+	proxy_authentication_required,
+	request_time_out,
+	conflict,
+	gone,
+	length_required,
+	precondition_failed,
+	request_entity_too_large,
+	request_uri_too_large,
+	unsupported_media_type,
+	requested_range_not_satisfied,
+	expectation_failed,
+	internal_server_error,
+	not_implemented,
+	bad_gateway,
+	service_unavailable,
+	gateway_timeout,
+	http_version_not_supported,
+	permanent_redirect,  // I hate your decisions, uefi.
+};
+
+struct http_response_data
+{
+	http_status_code status_code;
+};
+
+struct http_header
+{
+	char *field_name;
+	char *field_value;
+};
+
+struct http_message
+{
+	union {
+		http_request_data *request;
+		http_response_data *response;
+	} data;
+
+	size_t header_count;
+	http_header *headers;
+
+	size_t body_length;
+	void *body;
+};
+
+struct http_token
+{
+	event event;
+	status status;
+	http_message *message;
+};
+
+
+} // namespace uefi
+
+#endif

external/uefi/protos/device_path.h

@@ -16,6 +16,7 @@ struct device_path
 {
     static constexpr uefi::guid guid{ 0x09576e91,0x6d3f,0x11d2,{0x8e,0x39,0x00,0xa0,0xc9,0x69,0x72,0x3b} };
 
+
     uint8_t type;
     uint8_t sub_type;
     uint16_t length;
@@ -27,4 +28,4 @@ public:
 } // namespace protos
 } // namespace uefi
 
-#endif // _uefi_protos_device_path_h_
+#endif // _uefi_protos_device_path_h_

external/uefi/protos/file.h

@@ -14,7 +14,7 @@ struct file;
 
 struct file
 {
-    
+
 
     inline uefi::status open(file ** new_handle, const wchar_t * file_name, file_mode open_mode, file_attr attributes) {
         return _open(this, new_handle, file_name, open_mode, attributes);
@@ -123,4 +123,4 @@ public:
 } // namespace protos
 } // namespace uefi
 
-#endif // _uefi_protos_file_h_
+#endif // _uefi_protos_file_h_

external/uefi/protos/file_info.h

@@ -16,6 +16,7 @@ struct file_info
 {
     static constexpr uefi::guid guid{ 0x09576e92,0x6d3f,0x11d2,{0x8e,0x39,0x00,0xa0,0xc9,0x69,0x72,0x3b} };
 
+
     uint64_t size;
     uint64_t file_size;
     uint64_t physical_size;
@@ -32,4 +33,4 @@ public:
 } // namespace protos
 } // namespace uefi
 
-#endif // _uefi_protos_file_info_h_
+#endif // _uefi_protos_file_info_h_

external/uefi/protos/graphics_output.h

@@ -17,6 +17,7 @@ struct graphics_output
 {
     static constexpr uefi::guid guid{ 0x9042a9de,0x23dc,0x4a38,{0x96,0xfb,0x7a,0xde,0xd0,0x80,0x51,0x6a} };
 
+
     inline uefi::status query_mode(uint32_t mode_number, uint64_t * size_of_info, uefi::graphics_output_mode_info ** info) {
         return _query_mode(this, mode_number, size_of_info, info);
     }
@@ -47,4 +48,4 @@ public:
 } // namespace protos
 } // namespace uefi
 
-#endif // _uefi_protos_graphics_output_h_
+#endif // _uefi_protos_graphics_output_h_

external/uefi/protos/http.h

@@ -0,0 +1,72 @@
+#pragma once
+#ifndef _uefi_protos_http_h_
+#define _uefi_protos_http_h_
+
+// This file was auto generated by the j6-uefi-headers project. Please see
+// https://github.com/justinian/j6-uefi-headers for more information.
+
+#include <uefi/guid.h>
+#include <uefi/types.h>
+#include <uefi/networking.h>
+
+namespace uefi {
+namespace protos {
+struct http;
+
+struct http
+{
+    static constexpr uefi::guid guid{ 0x7a59b29b,0x910b,0x4171,{0x82,0x42,0xa8,0x5a,0x0d,0xf2,0x5b,0x5b} };
+    static constexpr uefi::guid service_binding{ 0xbdc8e6af,0xd9bc,0x4379,{0xa7,0x2a,0xe0,0xc4,0xe7,0x5d,0xae,0x1c} };
+
+
+    inline uefi::status get_mode_data(uefi::http_config_data * http_config_data) {
+        return _get_mode_data(this, http_config_data);
+    }
+
+    inline uefi::status configure(uefi::http_config_data * http_config_data) {
+        return _configure(this, http_config_data);
+    }
+
+    inline uefi::status request(uefi::http_token * token) {
+        return _request(this, token);
+    }
+
+    inline uefi::status cancel(uefi::http_token * token) {
+        return _cancel(this, token);
+    }
+
+    inline uefi::status response(uefi::http_token * token) {
+        return _response(this, token);
+    }
+
+    inline uefi::status poll() {
+        return _poll(this);
+    }
+
+
+protected:
+    using _get_mode_data_def = uefi::status (*)(uefi::protos::http *, uefi::http_config_data *);
+    _get_mode_data_def _get_mode_data;
+
+    using _configure_def = uefi::status (*)(uefi::protos::http *, uefi::http_config_data *);
+    _configure_def _configure;
+
+    using _request_def = uefi::status (*)(uefi::protos::http *, uefi::http_token *);
+    _request_def _request;
+
+    using _cancel_def = uefi::status (*)(uefi::protos::http *, uefi::http_token *);
+    _cancel_def _cancel;
+
+    using _response_def = uefi::status (*)(uefi::protos::http *, uefi::http_token *);
+    _response_def _response;
+
+    using _poll_def = uefi::status (*)(uefi::protos::http *);
+    _poll_def _poll;
+
+public:
+};
+
+} // namespace protos
+} // namespace uefi
+
+#endif // _uefi_protos_http_h_

external/uefi/protos/ip6.h

@@ -0,0 +1,93 @@
+#pragma once
+#ifndef _uefi_protos_ip6_h_
+#define _uefi_protos_ip6_h_
+
+// This file was auto generated by the j6-uefi-headers project. Please see
+// https://github.com/justinian/j6-uefi-headers for more information.
+
+#include <uefi/guid.h>
+#include <uefi/types.h>
+#include <uefi/networking.h>
+
+namespace uefi {
+namespace protos {
+struct ip6;
+
+struct ip6
+{
+    static constexpr uefi::guid guid{ 0x2c8759d5,0x5c2d,0x66ef,{0x92,0x5f,0xb6,0x6c,0x10,0x19,0x57,0xe2} };
+    static constexpr uefi::guid service_binding{ 0xec835dd3,0xfe0f,0x617b,{0xa6,0x21,0xb3,0x50,0xc3,0xe1,0x33,0x88} };
+
+
+    inline uefi::status get_mode_data(uefi::ip6_mode_data * ip6_mode_data, uefi::managed_network_config_data * mnp_config_data, uefi::simple_network_mode * snp_config_data) {
+        return _get_mode_data(this, ip6_mode_data, mnp_config_data, snp_config_data);
+    }
+
+    inline uefi::status configure(uefi::ip6_config_data * ip6_config_data) {
+        return _configure(this, ip6_config_data);
+    }
+
+    inline uefi::status groups(bool join_flag, uefi::ipv6_address * group_address) {
+        return _groups(this, join_flag, group_address);
+    }
+
+    inline uefi::status routes(bool delete_route, uefi::ipv6_address * destination, uint8_t prefix_length, uefi::ipv6_address * gateway_address) {
+        return _routes(this, delete_route, destination, prefix_length, gateway_address);
+    }
+
+    inline uefi::status neighbors(bool delete_flag, uefi::ipv6_address * target_ip6_address, uefi::mac_address * target_link_address, uint32_t timeout, bool override) {
+        return _neighbors(this, delete_flag, target_ip6_address, target_link_address, timeout, override);
+    }
+
+    inline uefi::status transmit(uefi::ip6_completion_token * token) {
+        return _transmit(this, token);
+    }
+
+    inline uefi::status receive(uefi::ip6_completion_token * token) {
+        return _receive(this, token);
+    }
+
+    inline uefi::status cancel(uefi::ip6_completion_token * token) {
+        return _cancel(this, token);
+    }
+
+    inline uefi::status poll() {
+        return _poll(this);
+    }
+
+
+protected:
+    using _get_mode_data_def = uefi::status (*)(uefi::protos::ip6 *, uefi::ip6_mode_data *, uefi::managed_network_config_data *, uefi::simple_network_mode *);
+    _get_mode_data_def _get_mode_data;
+
+    using _configure_def = uefi::status (*)(uefi::protos::ip6 *, uefi::ip6_config_data *);
+    _configure_def _configure;
+
+    using _groups_def = uefi::status (*)(uefi::protos::ip6 *, bool, uefi::ipv6_address *);
+    _groups_def _groups;
+
+    using _routes_def = uefi::status (*)(uefi::protos::ip6 *, bool, uefi::ipv6_address *, uint8_t, uefi::ipv6_address *);
+    _routes_def _routes;
+
+    using _neighbors_def = uefi::status (*)(uefi::protos::ip6 *, bool, uefi::ipv6_address *, uefi::mac_address *, uint32_t, bool);
+    _neighbors_def _neighbors;
+
+    using _transmit_def = uefi::status (*)(uefi::protos::ip6 *, uefi::ip6_completion_token *);
+    _transmit_def _transmit;
+
+    using _receive_def = uefi::status (*)(uefi::protos::ip6 *, uefi::ip6_completion_token *);
+    _receive_def _receive;
+
+    using _cancel_def = uefi::status (*)(uefi::protos::ip6 *, uefi::ip6_completion_token *);
+    _cancel_def _cancel;
+
+    using _poll_def = uefi::status (*)(uefi::protos::ip6 *);
+    _poll_def _poll;
+
+public:
+};
+
+} // namespace protos
+} // namespace uefi
+
+#endif // _uefi_protos_ip6_h_

external/uefi/protos/ip6_config.h

@@ -0,0 +1,57 @@
+#pragma once
+#ifndef _uefi_protos_ip6_config_h_
+#define _uefi_protos_ip6_config_h_
+
+// This file was auto generated by the j6-uefi-headers project. Please see
+// https://github.com/justinian/j6-uefi-headers for more information.
+
+#include <uefi/guid.h>
+#include <uefi/types.h>
+#include <uefi/networking.h>
+
+namespace uefi {
+namespace protos {
+struct ip6_config;
+
+struct ip6_config
+{
+    static constexpr uefi::guid guid{ 0x937fe521,0x95ae,0x4d1a,{0x89,0x29,0x48,0xbc,0xd9,0x0a,0xd3,0x1a} };
+
+
+    inline uefi::status set_data(uefi::ip6_config_data_type data_type, size_t data_size, void * data) {
+        return _set_data(this, data_type, data_size, data);
+    }
+
+    inline uefi::status get_data(uefi::ip6_config_data_type data_type, size_t data_size, void * data) {
+        return _get_data(this, data_type, data_size, data);
+    }
+
+    inline uefi::status register_data_notify(uefi::ip6_config_data_type data_type, uefi::event event) {
+        return _register_data_notify(this, data_type, event);
+    }
+
+    inline uefi::status unregister_data_notify(uefi::ip6_config_data_type data_type, uefi::event event) {
+        return _unregister_data_notify(this, data_type, event);
+    }
+
+
+protected:
+    using _set_data_def = uefi::status (*)(uefi::protos::ip6_config *, uefi::ip6_config_data_type, size_t, void *);
+    _set_data_def _set_data;
+
+    using _get_data_def = uefi::status (*)(uefi::protos::ip6_config *, uefi::ip6_config_data_type, size_t, void *);
+    _get_data_def _get_data;
+
+    using _register_data_notify_def = uefi::status (*)(uefi::protos::ip6_config *, uefi::ip6_config_data_type, uefi::event);
+    _register_data_notify_def _register_data_notify;
+
+    using _unregister_data_notify_def = uefi::status (*)(uefi::protos::ip6_config *, uefi::ip6_config_data_type, uefi::event);
+    _unregister_data_notify_def _unregister_data_notify;
+
+public:
+};
+
+} // namespace protos
+} // namespace uefi
+
+#endif // _uefi_protos_ip6_config_h_

external/uefi/protos/load_file.h

@@ -0,0 +1,36 @@
+#pragma once
+#ifndef _uefi_protos_load_file_h_
+#define _uefi_protos_load_file_h_
+
+// This file was auto generated by the j6-uefi-headers project. Please see
+// https://github.com/justinian/j6-uefi-headers for more information.
+
+#include <uefi/guid.h>
+#include <uefi/types.h>
+#include <uefi/protos/device_path.h>
+
+namespace uefi {
+namespace protos {
+struct load_file;
+
+struct load_file
+{
+    static constexpr uefi::guid guid{ {0x56ec3091,0x954c,0x11d2,{0x8e,0x3f,0x00,0xa0,0xc9,0x69,0x72,0x3b} };
+
+
+    inline uefi::status load_file(uefi::protos::device_path * file_path, bool boot_policy, size_t * buffer_size, void * buffer) {
+        return _load_file(this, file_path, boot_policy, buffer_size, buffer);
+    }
+
+
+protected:
+    using _load_file_def = uefi::status (*)(uefi::protos::load_file *, uefi::protos::device_path *, bool, size_t *, void *);
+    _load_file_def _load_file;
+
+public:
+};
+
+} // namespace protos
+} // namespace uefi
+
+#endif // _uefi_protos_load_file_h_

external/uefi/protos/loaded_image.h

@@ -18,6 +18,7 @@ struct loaded_image
 {
     static constexpr uefi::guid guid{ 0x5b1b31a1,0x9562,0x11d2,{0x8e,0x3f,0x00,0xa0,0xc9,0x69,0x72,0x3b} };
 
+
     inline uefi::status unload(uefi::handle image_handle) {
         return _unload(image_handle);
     }
@@ -45,4 +46,4 @@ public:
 } // namespace protos
 } // namespace uefi
 
-#endif // _uefi_protos_loaded_image_h_
+#endif // _uefi_protos_loaded_image_h_

external/uefi/protos/service_binding.h

@@ -0,0 +1,41 @@
+#pragma once
+#ifndef _uefi_protos_service_binding_h_
+#define _uefi_protos_service_binding_h_
+
+// This file was auto generated by the j6-uefi-headers project. Please see
+// https://github.com/justinian/j6-uefi-headers for more information.
+
+#include <uefi/guid.h>
+#include <uefi/types.h>
+
+namespace uefi {
+namespace protos {
+struct service_binding;
+
+struct service_binding
+{
+
+
+    inline uefi::status create_child(uefi::handle * child_handle) {
+        return _create_child(this, child_handle);
+    }
+
+    inline uefi::status destroy_child(uefi::handle child_handle) {
+        return _destroy_child(this, child_handle);
+    }
+
+
+protected:
+    using _create_child_def = uefi::status (*)(uefi::protos::service_binding *, uefi::handle *);
+    _create_child_def _create_child;
+
+    using _destroy_child_def = uefi::status (*)(uefi::protos::service_binding *, uefi::handle);
+    _destroy_child_def _destroy_child;
+
+public:
+};
+
+} // namespace protos
+} // namespace uefi
+
+#endif // _uefi_protos_service_binding_h_

external/uefi/protos/simple_file_system.h

@@ -17,6 +17,7 @@ struct simple_file_system
 {
     static constexpr uefi::guid guid{ 0x0964e5b22,0x6459,0x11d2,{0x8e,0x39,0x00,0xa0,0xc9,0x69,0x72,0x3b} };
 
+
     inline uefi::status open_volume(uefi::protos::file ** root) {
         return _open_volume(this, root);
     }
@@ -33,4 +34,4 @@ public:
 } // namespace protos
 } // namespace uefi
 
-#endif // _uefi_protos_simple_file_system_h_
+#endif // _uefi_protos_simple_file_system_h_

external/uefi/protos/simple_text_output.h

@@ -17,6 +17,7 @@ struct simple_text_output
 {
     static constexpr uefi::guid guid{ 0x387477c2,0x69c7,0x11d2,{0x8e,0x39,0x00,0xa0,0xc9,0x69,0x72,0x3b} };
 
+
     inline uefi::status reset(bool extended_verification) {
         return _reset(this, extended_verification);
     }
@@ -89,4 +90,4 @@ public:
 } // namespace protos
 } // namespace uefi
 
-#endif // _uefi_protos_simple_text_output_h_
+#endif // _uefi_protos_simple_text_output_h_

external/zstd.module

@@ -0,0 +1,33 @@
+# vim: ft=python
+
+zstd = module("zstd",
+    root = "${source_root}/external/zstd",
+    kind = "lib",
+    copy_headers = True,
+    deps = [ "libc" ],
+    sources = [
+        "decompress/zstd_decompress.c",
+        "decompress/zstd_ddict.c",
+        "decompress/huf_decompress.c",
+        "decompress/zstd_decompress_block.c",
+        "common/threading.c",
+        "common/fse_decompress.c",
+        "common/debug.c",
+        "common/xxhash.c",
+        "common/pool.c",
+        "common/error_private.c",
+        "common/entropy_common.c",
+        "common/zstd_common.c",
+    ],
+    public_headers = [
+        "zdict.h",
+        "zstd.h",
+        "zstd_errors.h",
+    ])
+
+zstd.variables['ccflags'] = [
+    "${ccflags}",
+    "-DXXH_NAMESPACE=ZSTD_",
+    "-DDEBUGLEVEL=0",
+    "-DZSTD_DISABLE_ASM",
+]

external/zstd/.gitignore

@@ -0,0 +1,3 @@
+# make install artefact
+libzstd.pc
+libzstd-nomt

external/zstd/BUCK

@@ -0,0 +1,232 @@
+cxx_library(
+    name='zstd',
+    header_namespace='',
+    exported_headers=['zstd.h'],
+    visibility=['PUBLIC'],
+    deps=[
+        ':common',
+        ':compress',
+        ':decompress',
+        ':deprecated',
+    ],
+)
+
+cxx_library(
+    name='compress',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    exported_headers=subdir_glob([
+        ('compress', 'zstd*.h'),
+    ]),
+    srcs=glob(['compress/zstd*.c', 'compress/hist.c']),
+    deps=[':common'],
+)
+
+cxx_library(
+    name='decompress',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    headers=subdir_glob([
+        ('decompress', '*_impl.h'),
+    ]),
+    srcs=glob(['decompress/zstd*.c']),
+    deps=[
+        ':common',
+        ':legacy',
+    ],
+)
+
+cxx_library(
+    name='deprecated',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    exported_headers=subdir_glob([
+        ('deprecated', '*.h'),
+    ]),
+    srcs=glob(['deprecated/*.c']),
+    deps=[':common'],
+)
+
+cxx_library(
+    name='legacy',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    exported_headers=subdir_glob([
+        ('legacy', '*.h'),
+    ]),
+    srcs=glob(['legacy/*.c']),
+    deps=[':common'],
+    exported_preprocessor_flags=[
+        '-DZSTD_LEGACY_SUPPORT=4',
+    ],
+)
+
+cxx_library(
+    name='zdict',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    exported_headers=['zdict.h'],
+    headers=subdir_glob([
+        ('dictBuilder', 'divsufsort.h'),
+        ('dictBuilder', 'cover.h'),
+    ]),
+    srcs=glob(['dictBuilder/*.c']),
+    deps=[':common'],
+)
+
+cxx_library(
+    name='compiler',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    exported_headers=subdir_glob([
+        ('common', 'compiler.h'),
+    ]),
+)
+
+cxx_library(
+    name='cpu',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    exported_headers=subdir_glob([
+        ('common', 'cpu.h'),
+    ]),
+)
+
+cxx_library(
+    name='bitstream',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    exported_headers=subdir_glob([
+        ('common', 'bitstream.h'),
+    ]),
+)
+
+cxx_library(
+    name='entropy',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    exported_headers=subdir_glob([
+        ('common', 'fse.h'),
+        ('common', 'huf.h'),
+    ]),
+    srcs=[
+        'common/entropy_common.c',
+        'common/fse_decompress.c',
+        'compress/fse_compress.c',
+        'compress/huf_compress.c',
+        'decompress/huf_decompress.c',
+    ],
+    deps=[
+        ':debug',
+        ':bitstream',
+        ':compiler',
+        ':errors',
+        ':mem',
+    ],
+)
+
+cxx_library(
+    name='errors',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    exported_headers=[
+        'zstd_errors.h',
+        'common/error_private.h',
+    ]
+    srcs=['common/error_private.c'],
+)
+
+cxx_library(
+    name='mem',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    exported_headers=subdir_glob([
+        ('common', 'mem.h'),
+    ]),
+)
+
+cxx_library(
+    name='pool',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    exported_headers=subdir_glob([
+        ('common', 'pool.h'),
+    ]),
+    srcs=['common/pool.c'],
+    deps=[
+        ':threading',
+        ':zstd_common',
+    ],
+)
+
+cxx_library(
+    name='threading',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    exported_headers=subdir_glob([
+        ('common', 'threading.h'),
+    ]),
+    srcs=['common/threading.c'],
+    exported_preprocessor_flags=[
+        '-DZSTD_MULTITHREAD',
+    ],
+    exported_linker_flags=[
+        '-pthread',
+    ],
+)
+
+cxx_library(
+    name='xxhash',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    exported_headers=subdir_glob([
+        ('common', 'xxhash.h'),
+    ]),
+    srcs=['common/xxhash.c'],
+    exported_preprocessor_flags=[
+        '-DXXH_NAMESPACE=ZSTD_',
+    ],
+)
+
+cxx_library(
+    name='zstd_common',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    exported_headers=subdir_glob([
+        ('', 'zstd.h'),
+        ('common', 'zstd_internal.h'),
+    ]),
+    srcs=['common/zstd_common.c'],
+    deps=[
+        ':compiler',
+        ':errors',
+        ':mem',
+    ],
+)
+
+cxx_library(
+    name='debug',
+    header_namespace='',
+    visibility=['PUBLIC'],
+    exported_headers=subdir_glob([
+        ('common', 'debug.h'),
+    ]),
+    srcs=['common/debug.c'],
+)
+
+cxx_library(
+    name='common',
+    deps=[
+        ':debug',
+        ':bitstream',
+        ':compiler',
+        ':cpu',
+        ':entropy',
+        ':errors',
+        ':mem',
+        ':pool',
+        ':threading',
+        ':xxhash',
+        ':zstd_common',
+    ]
+)

external/zstd/LICENSE

@@ -0,0 +1,30 @@
+BSD License
+
+For Zstandard software
+
+Copyright (c) Meta Platforms, Inc. and affiliates. All rights reserved.
+
+Redistribution and use in source and binary forms, with or without modification,
+are permitted provided that the following conditions are met:
+
+ * Redistributions of source code must retain the above copyright notice, this
+   list of conditions and the following disclaimer.
+
+ * Redistributions in binary form must reproduce the above copyright notice,
+   this list of conditions and the following disclaimer in the documentation
+   and/or other materials provided with the distribution.
+
+ * Neither the name Facebook, nor Meta, nor the names of its contributors may
+   be used to endorse or promote products derived from this software without
+   specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR
+ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
+ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

external/zstd/Makefile

@@ -0,0 +1,357 @@
+# ################################################################
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under both the BSD-style license (found in the
+# LICENSE file in the root directory of this source tree) and the GPLv2 (found
+# in the COPYING file in the root directory of this source tree).
+# You may select, at your option, one of the above-listed licenses.
+# ################################################################
+
+# Modules
+ZSTD_LIB_COMPRESSION ?= 1
+ZSTD_LIB_DECOMPRESSION ?= 1
+ZSTD_LIB_DICTBUILDER ?= 1
+ZSTD_LIB_DEPRECATED ?= 0
+
+# Input variables for libzstd.mk
+ifeq ($(ZSTD_LIB_COMPRESSION), 0)
+  ZSTD_LIB_DICTBUILDER = 0
+  ZSTD_LIB_DEPRECATED = 0
+endif
+
+ifeq ($(ZSTD_LIB_DECOMPRESSION), 0)
+  ZSTD_LEGACY_SUPPORT = 0
+  ZSTD_LIB_DEPRECATED = 0
+endif
+
+include libzstd.mk
+
+ZSTD_FILES := $(ZSTD_COMMON_FILES) $(ZSTD_LEGACY_FILES)
+
+ifneq ($(ZSTD_LIB_COMPRESSION), 0)
+  ZSTD_FILES += $(ZSTD_COMPRESS_FILES)
+endif
+
+ifneq ($(ZSTD_LIB_DECOMPRESSION), 0)
+  ZSTD_FILES += $(ZSTD_DECOMPRESS_FILES)
+endif
+
+ifneq ($(ZSTD_LIB_DEPRECATED), 0)
+  ZSTD_FILES += $(ZSTD_DEPRECATED_FILES)
+endif
+
+ifneq ($(ZSTD_LIB_DICTBUILDER), 0)
+  ZSTD_FILES += $(ZSTD_DICTBUILDER_FILES)
+endif
+
+ZSTD_LOCAL_SRC := $(notdir $(ZSTD_FILES))
+ZSTD_LOCAL_OBJ0 := $(ZSTD_LOCAL_SRC:.c=.o)
+ZSTD_LOCAL_OBJ := $(ZSTD_LOCAL_OBJ0:.S=.o)
+
+VERSION := $(ZSTD_VERSION)
+
+# Note: by default, the static library is built single-threaded and dynamic library is built
+# multi-threaded. It is possible to force multi or single threaded builds by appending
+# -mt or -nomt to the build target (like lib-mt for multi-threaded, lib-nomt for single-threaded).
+.PHONY: default
+default: lib-release
+
+CPPFLAGS_DYNLIB  += -DZSTD_MULTITHREAD # dynamic library build defaults to multi-threaded
+LDFLAGS_DYNLIB   += -pthread
+CPPFLAGS_STATLIB +=                    # static library build defaults to single-threaded
+
+
+ifeq ($(findstring GCC,$(CCVER)),GCC)
+decompress/zstd_decompress_block.o : CFLAGS+=-fno-tree-vectorize
+endif
+
+
+# macOS linker doesn't support -soname, and use different extension
+# see : https://developer.apple.com/library/mac/documentation/DeveloperTools/Conceptual/DynamicLibraries/100-Articles/DynamicLibraryDesignGuidelines.html
+ifeq ($(UNAME), Darwin)
+  SHARED_EXT = dylib
+  SHARED_EXT_MAJOR = $(LIBVER_MAJOR).$(SHARED_EXT)
+  SHARED_EXT_VER = $(LIBVER).$(SHARED_EXT)
+  SONAME_FLAGS = -install_name $(LIBDIR)/libzstd.$(SHARED_EXT_MAJOR) -compatibility_version $(LIBVER_MAJOR) -current_version $(LIBVER)
+else
+  ifeq ($(UNAME), AIX)
+    SONAME_FLAGS =
+  else
+    SONAME_FLAGS = -Wl,-soname=libzstd.$(SHARED_EXT).$(LIBVER_MAJOR)
+  endif
+  SHARED_EXT = so
+  SHARED_EXT_MAJOR = $(SHARED_EXT).$(LIBVER_MAJOR)
+  SHARED_EXT_VER = $(SHARED_EXT).$(LIBVER)
+endif
+
+
+.PHONY: all
+all: lib
+
+
+.PHONY: libzstd.a  # must be run every time
+libzstd.a: CPPFLAGS += $(CPPFLAGS_STATLIB)
+
+SET_CACHE_DIRECTORY = \
+   +$(MAKE) --no-print-directory $@ \
+    BUILD_DIR=obj/$(HASH_DIR) \
+    CPPFLAGS="$(CPPFLAGS)" \
+    CFLAGS="$(CFLAGS)" \
+    LDFLAGS="$(LDFLAGS)"
+
+ifndef BUILD_DIR
+# determine BUILD_DIR from compilation flags
+
+libzstd.a:
+	$(SET_CACHE_DIRECTORY)
+
+else
+# BUILD_DIR is defined
+
+ZSTD_STATLIB_DIR := $(BUILD_DIR)/static
+ZSTD_STATLIB := $(ZSTD_STATLIB_DIR)/libzstd.a
+ZSTD_STATLIB_OBJ := $(addprefix $(ZSTD_STATLIB_DIR)/,$(ZSTD_LOCAL_OBJ))
+$(ZSTD_STATLIB): ARFLAGS = rcs
+$(ZSTD_STATLIB): | $(ZSTD_STATLIB_DIR)
+$(ZSTD_STATLIB): $(ZSTD_STATLIB_OBJ)
+  # Check for multithread flag at target execution time
+	$(if $(filter -DZSTD_MULTITHREAD,$(CPPFLAGS)),\
+    @echo compiling multi-threaded static library $(LIBVER),\
+    @echo compiling single-threaded static library $(LIBVER))
+	$(AR) $(ARFLAGS) $@ $^
+
+libzstd.a: $(ZSTD_STATLIB)
+	cp -f $< $@
+
+endif
+
+ifneq (,$(filter Windows%,$(TARGET_SYSTEM)))
+
+LIBZSTD = dll/libzstd.dll
+$(LIBZSTD): $(ZSTD_FILES)
+	@echo compiling dynamic library $(LIBVER)
+	$(CC) $(FLAGS) -DZSTD_DLL_EXPORT=1 -Wl,--out-implib,dll/libzstd.dll.a -shared $^ -o $@
+
+else  # not Windows
+
+LIBZSTD = libzstd.$(SHARED_EXT_VER)
+.PHONY: $(LIBZSTD)  # must be run every time
+$(LIBZSTD): CPPFLAGS += $(CPPFLAGS_DYNLIB)
+$(LIBZSTD): CFLAGS   += -fPIC -fvisibility=hidden
+$(LIBZSTD): LDFLAGS  += -shared $(LDFLAGS_DYNLIB)
+
+ifndef BUILD_DIR
+# determine BUILD_DIR from compilation flags
+
+$(LIBZSTD):
+	$(SET_CACHE_DIRECTORY)
+
+else
+# BUILD_DIR is defined
+
+ZSTD_DYNLIB_DIR := $(BUILD_DIR)/dynamic
+ZSTD_DYNLIB := $(ZSTD_DYNLIB_DIR)/$(LIBZSTD)
+ZSTD_DYNLIB_OBJ := $(addprefix $(ZSTD_DYNLIB_DIR)/,$(ZSTD_LOCAL_OBJ))
+
+$(ZSTD_DYNLIB): | $(ZSTD_DYNLIB_DIR)
+$(ZSTD_DYNLIB): $(ZSTD_DYNLIB_OBJ)
+# Check for multithread flag at target execution time
+	$(if $(filter -DZSTD_MULTITHREAD,$(CPPFLAGS)),\
+    @echo compiling multi-threaded dynamic library $(LIBVER),\
+    @echo compiling single-threaded dynamic library $(LIBVER))
+	$(CC) $(FLAGS) $^ $(LDFLAGS) $(SONAME_FLAGS) -o $@
+	@echo creating versioned links
+	ln -sf $@ libzstd.$(SHARED_EXT_MAJOR)
+	ln -sf $@ libzstd.$(SHARED_EXT)
+
+$(LIBZSTD): $(ZSTD_DYNLIB)
+	cp -f $< $@
+
+endif  # ifndef BUILD_DIR
+endif  # if windows
+
+.PHONY: libzstd
+libzstd : $(LIBZSTD)
+
+.PHONY: lib
+lib : libzstd.a libzstd
+
+
+# note : do not define lib-mt or lib-release as .PHONY
+# make does not consider implicit pattern rule for .PHONY target
+
+%-mt : CPPFLAGS_DYNLIB  := -DZSTD_MULTITHREAD
+%-mt : CPPFLAGS_STATLIB := -DZSTD_MULTITHREAD
+%-mt : LDFLAGS_DYNLIB   := -pthread
+%-mt : %
+	@echo multi-threaded build completed
+
+%-nomt : CPPFLAGS_DYNLIB  :=
+%-nomt : LDFLAGS_DYNLIB   :=
+%-nomt : CPPFLAGS_STATLIB :=
+%-nomt : %
+	@echo single-threaded build completed
+
+%-release : DEBUGFLAGS :=
+%-release : %
+	@echo release build completed
+
+
+# Generate .h dependencies automatically
+
+DEPFLAGS = -MT $@ -MMD -MP -MF
+
+$(ZSTD_DYNLIB_DIR)/%.o : %.c $(ZSTD_DYNLIB_DIR)/%.d | $(ZSTD_DYNLIB_DIR)
+	@echo CC $@
+	$(COMPILE.c) $(DEPFLAGS) $(ZSTD_DYNLIB_DIR)/$*.d $(OUTPUT_OPTION) $<
+
+$(ZSTD_STATLIB_DIR)/%.o : %.c $(ZSTD_STATLIB_DIR)/%.d | $(ZSTD_STATLIB_DIR)
+	@echo CC $@
+	$(COMPILE.c) $(DEPFLAGS) $(ZSTD_STATLIB_DIR)/$*.d $(OUTPUT_OPTION) $<
+
+$(ZSTD_DYNLIB_DIR)/%.o : %.S | $(ZSTD_DYNLIB_DIR)
+	@echo AS $@
+	$(COMPILE.S) $(OUTPUT_OPTION) $<
+
+$(ZSTD_STATLIB_DIR)/%.o : %.S | $(ZSTD_STATLIB_DIR)
+	@echo AS $@
+	$(COMPILE.S) $(OUTPUT_OPTION) $<
+
+MKDIR ?= mkdir
+$(BUILD_DIR) $(ZSTD_DYNLIB_DIR) $(ZSTD_STATLIB_DIR):
+	$(MKDIR) -p $@
+
+DEPFILES := $(ZSTD_DYNLIB_OBJ:.o=.d) $(ZSTD_STATLIB_OBJ:.o=.d)
+$(DEPFILES):
+
+include $(wildcard $(DEPFILES))
+
+
+# Special case : building library in single-thread mode _and_ without zstdmt_compress.c
+ZSTDMT_FILES = compress/zstdmt_compress.c
+ZSTD_NOMT_FILES = $(filter-out $(ZSTDMT_FILES),$(ZSTD_FILES))
+libzstd-nomt: CFLAGS += -fPIC -fvisibility=hidden
+libzstd-nomt: LDFLAGS += -shared
+libzstd-nomt: $(ZSTD_NOMT_FILES)
+	@echo compiling single-thread dynamic library $(LIBVER)
+	@echo files : $(ZSTD_NOMT_FILES)
+	$(CC) $(FLAGS) $^ $(LDFLAGS) $(SONAME_FLAGS) -o $@
+
+.PHONY: clean
+clean:
+	$(RM) -r *.dSYM   # macOS-specific
+	$(RM) core *.o *.a *.gcda *.$(SHARED_EXT) *.$(SHARED_EXT).* libzstd.pc
+	$(RM) dll/libzstd.dll dll/libzstd.lib libzstd-nomt*
+	$(RM) -r obj/*
+	@echo Cleaning library completed
+
+#-----------------------------------------------------------------------------
+# make install is validated only for below listed environments
+#-----------------------------------------------------------------------------
+ifneq (,$(filter $(UNAME),Linux Darwin GNU/kFreeBSD GNU OpenBSD FreeBSD NetBSD DragonFly SunOS Haiku AIX))
+
+lib: libzstd.pc
+
+HAS_EXPLICIT_EXEC_PREFIX := $(if $(or $(EXEC_PREFIX),$(exec_prefix)),1,)
+
+DESTDIR     ?=
+# directory variables : GNU conventions prefer lowercase
+# see https://www.gnu.org/prep/standards/html_node/Makefile-Conventions.html
+# support both lower and uppercase (BSD), use uppercase in script
+prefix      ?= /usr/local
+PREFIX      ?= $(prefix)
+exec_prefix ?= $(PREFIX)
+EXEC_PREFIX ?= $(exec_prefix)
+libdir      ?= $(EXEC_PREFIX)/lib
+LIBDIR      ?= $(libdir)
+includedir  ?= $(PREFIX)/include
+INCLUDEDIR  ?= $(includedir)
+
+PCINCDIR := $(patsubst $(PREFIX)%,%,$(INCLUDEDIR))
+PCLIBDIR := $(patsubst $(EXEC_PREFIX)%,%,$(LIBDIR))
+
+# If we successfully stripped off a prefix, we'll add a reference to the
+# relevant pc variable.
+PCINCPREFIX := $(if $(findstring $(INCLUDEDIR),$(PCINCDIR)),,$${prefix})
+PCLIBPREFIX := $(if $(findstring $(LIBDIR),$(PCLIBDIR)),,$${exec_prefix})
+
+# If no explicit EXEC_PREFIX was set by the caller, write it out as a reference
+# to PREFIX, rather than as a resolved value.
+PCEXEC_PREFIX := $(if $(HAS_EXPLICIT_EXEC_PREFIX),$(EXEC_PREFIX),$${prefix})
+
+ifneq (,$(filter $(UNAME),FreeBSD NetBSD DragonFly))
+  PKGCONFIGDIR ?= $(PREFIX)/libdata/pkgconfig
+else
+  PKGCONFIGDIR ?= $(LIBDIR)/pkgconfig
+endif
+
+ifneq (,$(filter $(UNAME),SunOS))
+  INSTALL ?= ginstall
+else
+  INSTALL ?= install
+endif
+
+INSTALL_PROGRAM ?= $(INSTALL)
+INSTALL_DATA    ?= $(INSTALL) -m 644
+
+
+libzstd.pc: libzstd.pc.in
+	@echo creating pkgconfig
+	@sed \
+	        -e 's|@PREFIX@|$(PREFIX)|' \
+	        -e 's|@EXEC_PREFIX@|$(PCEXEC_PREFIX)|' \
+	        -e 's|@INCLUDEDIR@|$(PCINCPREFIX)$(PCINCDIR)|' \
+	        -e 's|@LIBDIR@|$(PCLIBPREFIX)$(PCLIBDIR)|' \
+	        -e 's|@VERSION@|$(VERSION)|' \
+	        -e 's|@LIBS_PRIVATE@|$(LDFLAGS_DYNLIB)|' \
+	        $< >$@
+
+.PHONY: install
+install: install-pc install-static install-shared install-includes
+	@echo zstd static and shared library installed
+
+.PHONY: install-pc
+install-pc: libzstd.pc
+	[ -e $(DESTDIR)$(PKGCONFIGDIR) ] || $(INSTALL) -d -m 755 $(DESTDIR)$(PKGCONFIGDIR)/
+	$(INSTALL_DATA) libzstd.pc $(DESTDIR)$(PKGCONFIGDIR)/
+
+.PHONY: install-static
+install-static:
+	# only generate libzstd.a if it's not already present
+	[ -e libzstd.a ] || $(MAKE) libzstd.a-release
+	[ -e $(DESTDIR)$(LIBDIR) ] || $(INSTALL) -d -m 755 $(DESTDIR)$(LIBDIR)/
+	@echo Installing static library
+	$(INSTALL_DATA) libzstd.a $(DESTDIR)$(LIBDIR)
+
+.PHONY: install-shared
+install-shared:
+	# only generate libzstd.so if it's not already present
+	[ -e $(LIBZSTD) ] || $(MAKE) libzstd-release
+	[ -e $(DESTDIR)$(LIBDIR) ] || $(INSTALL) -d -m 755 $(DESTDIR)$(LIBDIR)/
+	@echo Installing shared library
+	$(INSTALL_PROGRAM) $(LIBZSTD) $(DESTDIR)$(LIBDIR)
+	ln -sf $(LIBZSTD) $(DESTDIR)$(LIBDIR)/libzstd.$(SHARED_EXT_MAJOR)
+	ln -sf $(LIBZSTD) $(DESTDIR)$(LIBDIR)/libzstd.$(SHARED_EXT)
+
+.PHONY: install-includes
+install-includes:
+	[ -e $(DESTDIR)$(INCLUDEDIR) ] || $(INSTALL) -d -m 755 $(DESTDIR)$(INCLUDEDIR)/
+	@echo Installing includes
+	$(INSTALL_DATA) zstd.h $(DESTDIR)$(INCLUDEDIR)
+	$(INSTALL_DATA) zstd_errors.h $(DESTDIR)$(INCLUDEDIR)
+	$(INSTALL_DATA) zdict.h $(DESTDIR)$(INCLUDEDIR)
+
+.PHONY: uninstall
+uninstall:
+	$(RM) $(DESTDIR)$(LIBDIR)/libzstd.a
+	$(RM) $(DESTDIR)$(LIBDIR)/libzstd.$(SHARED_EXT)
+	$(RM) $(DESTDIR)$(LIBDIR)/libzstd.$(SHARED_EXT_MAJOR)
+	$(RM) $(DESTDIR)$(LIBDIR)/$(LIBZSTD)
+	$(RM) $(DESTDIR)$(PKGCONFIGDIR)/libzstd.pc
+	$(RM) $(DESTDIR)$(INCLUDEDIR)/zstd.h
+	$(RM) $(DESTDIR)$(INCLUDEDIR)/zstd_errors.h
+	$(RM) $(DESTDIR)$(INCLUDEDIR)/zdict.h
+	@echo zstd libraries successfully uninstalled
+
+endif

external/zstd/README.md

@@ -0,0 +1,224 @@
+Zstandard library files
+================================
+
+The __lib__ directory is split into several sub-directories,
+in order to make it easier to select or exclude features.
+
+
+#### Building
+
+`Makefile` script is provided, supporting [Makefile conventions](https://www.gnu.org/prep/standards/html_node/Makefile-Conventions.html#Makefile-Conventions),
+including commands variables, staged install, directory variables and standard targets.
+- `make` : generates both static and dynamic libraries
+- `make install` : install libraries and headers in target system directories
+
+`libzstd` default scope is pretty large, including compression, decompression, dictionary builder,
+and support for decoding legacy formats >= v0.5.0.
+The scope can be reduced on demand (see paragraph _modular build_).
+
+
+#### Multithreading support
+
+When building with `make`, by default the dynamic library is multithreaded and static library is single-threaded (for compatibility reasons).
+
+Enabling multithreading requires 2 conditions :
+- set build macro `ZSTD_MULTITHREAD` (`-DZSTD_MULTITHREAD` for `gcc`)
+- for POSIX systems : compile with pthread (`-pthread` compilation flag for `gcc`)
+
+For convenience, we provide a build target to generate multi and single threaded libraries:
+- Force enable multithreading on both dynamic and static libraries by appending `-mt` to the target, e.g. `make lib-mt`.
+- Force disable multithreading on both dynamic and static libraries by appending `-nomt` to the target, e.g. `make lib-nomt`.
+- By default, as mentioned before, dynamic library is multithreaded, and static library is single-threaded, e.g. `make lib`.
+
+When linking a POSIX program with a multithreaded version of `libzstd`,
+note that it's necessary to invoke the `-pthread` flag during link stage.
+
+Multithreading capabilities are exposed
+via the [advanced API defined in `lib/zstd.h`](https://github.com/facebook/zstd/blob/v1.4.3/lib/zstd.h#L351).
+
+
+#### API
+
+Zstandard's stable API is exposed within [lib/zstd.h](zstd.h).
+
+
+#### Advanced API
+
+Optional advanced features are exposed via :
+
+- `lib/zstd_errors.h` : translates `size_t` function results
+                        into a `ZSTD_ErrorCode`, for accurate error handling.
+
+- `ZSTD_STATIC_LINKING_ONLY` : if this macro is defined _before_ including `zstd.h`,
+                          it unlocks access to the experimental API,
+                          exposed in the second part of `zstd.h`.
+                          All definitions in the experimental APIs are unstable,
+                          they may still change in the future, or even be removed.
+                          As a consequence, experimental definitions shall ___never be used with dynamic library___ !
+                          Only static linking is allowed.
+
+
+#### Modular build
+
+It's possible to compile only a limited set of features within `libzstd`.
+The file structure is designed to make this selection manually achievable for any build system :
+
+- Directory `lib/common` is always required, for all variants.
+
+- Compression source code lies in `lib/compress`
+
+- Decompression source code lies in `lib/decompress`
+
+- It's possible to include only `compress` or only `decompress`, they don't depend on each other.
+
+- `lib/dictBuilder` : makes it possible to generate dictionaries from a set of samples.
+        The API is exposed in `lib/dictBuilder/zdict.h`.
+        This module depends on both `lib/common` and `lib/compress` .
+
+- `lib/legacy` : makes it possible to decompress legacy zstd formats, starting from `v0.1.0`.
+        This module depends on `lib/common` and `lib/decompress`.
+        To enable this feature, define `ZSTD_LEGACY_SUPPORT` during compilation.
+        Specifying a number limits versions supported to that version onward.
+        For example, `ZSTD_LEGACY_SUPPORT=2` means : "support legacy formats >= v0.2.0".
+        Conversely, `ZSTD_LEGACY_SUPPORT=0` means "do __not__ support legacy formats".
+        By default, this build macro is set as `ZSTD_LEGACY_SUPPORT=5`.
+        Decoding supported legacy format is a transparent capability triggered within decompression functions.
+        It's also allowed to invoke legacy API directly, exposed in `lib/legacy/zstd_legacy.h`.
+        Each version does also provide its own set of advanced API.
+        For example, advanced API for version `v0.4` is exposed in `lib/legacy/zstd_v04.h` .
+
+- While invoking `make libzstd`, it's possible to define build macros
+        `ZSTD_LIB_COMPRESSION, ZSTD_LIB_DECOMPRESSION`, `ZSTD_LIB_DICTBUILDER`,
+        and `ZSTD_LIB_DEPRECATED` as `0` to forgo compilation of the
+        corresponding features. This will also disable compilation of all
+        dependencies (e.g. `ZSTD_LIB_COMPRESSION=0` will also disable
+        dictBuilder).
+
+- There are a number of options that can help minimize the binary size of
+  `libzstd`.
+
+  The first step is to select the components needed (using the above-described
+  `ZSTD_LIB_COMPRESSION` etc.).
+
+  The next step is to set `ZSTD_LIB_MINIFY` to `1` when invoking `make`. This
+  disables various optional components and changes the compilation flags to
+  prioritize space-saving.
+
+  Detailed options: Zstandard's code and build environment is set up by default
+  to optimize above all else for performance. In pursuit of this goal, Zstandard
+  makes significant trade-offs in code size. For example, Zstandard often has
+  more than one implementation of a particular component, with each
+  implementation optimized for different scenarios. For example, the Huffman
+  decoder has complementary implementations that decode the stream one symbol at
+  a time or two symbols at a time. Zstd normally includes both (and dispatches
+  between them at runtime), but by defining `HUF_FORCE_DECOMPRESS_X1` or
+  `HUF_FORCE_DECOMPRESS_X2`, you can force the use of one or the other, avoiding
+  compilation of the other. Similarly, `ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT`
+  and `ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG` force the compilation and use of
+  only one or the other of two decompression implementations. The smallest
+  binary is achieved by using `HUF_FORCE_DECOMPRESS_X1` and
+  `ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT` (implied by `ZSTD_LIB_MINIFY`).
+
+  For squeezing the last ounce of size out, you can also define
+  `ZSTD_NO_INLINE`, which disables inlining, and `ZSTD_STRIP_ERROR_STRINGS`,
+  which removes the error messages that are otherwise returned by
+  `ZSTD_getErrorName` (implied by `ZSTD_LIB_MINIFY`).
+
+  Finally, when integrating into your application, make sure you're doing link-
+  time optimization and unused symbol garbage collection (via some combination of,
+  e.g., `-flto`, `-ffat-lto-objects`, `-fuse-linker-plugin`,
+  `-ffunction-sections`, `-fdata-sections`, `-fmerge-all-constants`,
+  `-Wl,--gc-sections`, `-Wl,-z,norelro`, and an archiver that understands
+  the compiler's intermediate representation, e.g., `AR=gcc-ar`). Consult your
+  compiler's documentation.
+
+- While invoking `make libzstd`, the build macro `ZSTD_LEGACY_MULTITHREADED_API=1`
+  will expose the deprecated `ZSTDMT` API exposed by `zstdmt_compress.h` in
+  the shared library, which is now hidden by default.
+
+- The build macro `DYNAMIC_BMI2` can be set to 1 or 0 in order to generate binaries
+  which can detect at runtime the presence of BMI2 instructions, and use them only if present.
+  These instructions contribute to better performance, notably on the decoder side.
+  By default, this feature is automatically enabled on detecting
+  the right instruction set (x64) and compiler (clang or gcc >= 5).
+  It's obviously disabled for different cpus,
+  or when BMI2 instruction set is _required_ by the compiler command line
+  (in this case, only the BMI2 code path is generated).
+  Setting this macro will either force to generate the BMI2 dispatcher (1)
+  or prevent it (0). It overrides automatic detection.
+
+- The build macro `ZSTD_NO_UNUSED_FUNCTIONS` can be defined to hide the definitions of functions
+  that zstd does not use. Not all unused functions are hidden, but they can be if needed.
+  Currently, this macro will hide function definitions in FSE and HUF that use an excessive
+  amount of stack space.
+
+- The build macro `ZSTD_NO_INTRINSICS` can be defined to disable all explicit intrinsics.
+  Compiler builtins are still used.
+
+- The build macro `ZSTD_DECODER_INTERNAL_BUFFER` can be set to control
+  the amount of extra memory used during decompression to store literals.
+  This defaults to 64kB.  Reducing this value reduces the memory footprint of
+  `ZSTD_DCtx` decompression contexts,
+  but might also result in a small decompression speed cost.
+
+- The C compiler macros `ZSTDLIB_VISIBLE`, `ZSTDERRORLIB_VISIBLE` and `ZDICTLIB_VISIBLE`
+  can be overridden to control the visibility of zstd's API. Additionally,
+  `ZSTDLIB_STATIC_API` and `ZDICTLIB_STATIC_API` can be overridden to control the visibility
+  of zstd's static API. Specifically, it can be set to `ZSTDLIB_HIDDEN` to hide the symbols
+  from the shared library. These macros default to `ZSTDLIB_VISIBILITY`,
+  `ZSTDERRORLIB_VSIBILITY`, and `ZDICTLIB_VISIBILITY` if unset, for backwards compatibility
+  with the old macro names.
+
+#### Windows : using MinGW+MSYS to create DLL
+
+DLL can be created using MinGW+MSYS with the `make libzstd` command.
+This command creates `dll\libzstd.dll` and the import library `dll\libzstd.lib`.
+The import library is only required with Visual C++.
+The header file `zstd.h` and the dynamic library `dll\libzstd.dll` are required to
+compile a project using gcc/MinGW.
+The dynamic library has to be added to linking options.
+It means that if a project that uses ZSTD consists of a single `test-dll.c`
+file it should be linked with `dll\libzstd.dll`. For example:
+```
+    gcc $(CFLAGS) -Iinclude/ test-dll.c -o test-dll dll\libzstd.dll
+```
+The compiled executable will require ZSTD DLL which is available at `dll\libzstd.dll`.
+
+
+#### Advanced Build options
+
+The build system requires a hash function in order to
+separate object files created with different compilation flags.
+By default, it tries to use `md5sum` or equivalent.
+The hash function can be manually switched by setting the `HASH` variable.
+For example : `make HASH=xxhsum`
+The hash function needs to generate at least 64-bit using hexadecimal format.
+When no hash function is found,
+the Makefile just generates all object files into the same default directory,
+irrespective of compilation flags.
+This functionality only matters if `libzstd` is compiled multiple times
+with different build flags.
+
+The build directory, where object files are stored
+can also be manually controlled using variable `BUILD_DIR`,
+for example `make BUILD_DIR=objectDir/v1`.
+In which case, the hash function doesn't matter.
+
+
+#### Deprecated API
+
+Obsolete API on their way out are stored in directory `lib/deprecated`.
+At this stage, it contains older streaming prototypes, in `lib/deprecated/zbuff.h`.
+These prototypes will be removed in some future version.
+Consider migrating code towards supported streaming API exposed in `zstd.h`.
+
+
+#### Miscellaneous
+
+The other files are not source code. There are :
+
+ - `BUCK` : support for `buck` build system (https://buckbuild.com/)
+ - `Makefile` : `make` script to build and install zstd library (static and dynamic)
+ - `README.md` : this file
+ - `dll/` : resources directory for Windows compilation
+ - `libzstd.pc.in` : script for `pkg-config` (used in `make install`)

external/zstd/common/bits.h

@@ -0,0 +1,175 @@
+/*
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * All rights reserved.
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+ */
+
+#ifndef ZSTD_BITS_H
+#define ZSTD_BITS_H
+
+#include "mem.h"
+
+MEM_STATIC unsigned ZSTD_countTrailingZeros32_fallback(U32 val)
+{
+    assert(val != 0);
+    {
+        static const int DeBruijnBytePos[32] = {0, 1, 28, 2, 29, 14, 24, 3,
+                                                30, 22, 20, 15, 25, 17, 4, 8,
+                                                31, 27, 13, 23, 21, 19, 16, 7,
+                                                26, 12, 18, 6, 11, 5, 10, 9};
+        return DeBruijnBytePos[((U32) ((val & -(S32) val) * 0x077CB531U)) >> 27];
+    }
+}
+
+MEM_STATIC unsigned ZSTD_countTrailingZeros32(U32 val)
+{
+    assert(val != 0);
+#   if defined(_MSC_VER)
+#       if STATIC_BMI2 == 1
+            return _tzcnt_u32(val);
+#       else
+            if (val != 0) {
+                unsigned long r;
+                _BitScanForward(&r, val);
+                return (unsigned)r;
+            } else {
+                /* Should not reach this code path */
+                __assume(0);
+            }
+#       endif
+#   elif defined(__GNUC__) && (__GNUC__ >= 4)
+        return (unsigned)__builtin_ctz(val);
+#   else
+        return ZSTD_countTrailingZeros32_fallback(val);
+#   endif
+}
+
+MEM_STATIC unsigned ZSTD_countLeadingZeros32_fallback(U32 val) {
+    assert(val != 0);
+    {
+        static const U32 DeBruijnClz[32] = {0, 9, 1, 10, 13, 21, 2, 29,
+                                            11, 14, 16, 18, 22, 25, 3, 30,
+                                            8, 12, 20, 28, 15, 17, 24, 7,
+                                            19, 27, 23, 6, 26, 5, 4, 31};
+        val |= val >> 1;
+        val |= val >> 2;
+        val |= val >> 4;
+        val |= val >> 8;
+        val |= val >> 16;
+        return 31 - DeBruijnClz[(val * 0x07C4ACDDU) >> 27];
+    }
+}
+
+MEM_STATIC unsigned ZSTD_countLeadingZeros32(U32 val)
+{
+    assert(val != 0);
+#   if defined(_MSC_VER)
+#       if STATIC_BMI2 == 1
+            return _lzcnt_u32(val);
+#       else
+            if (val != 0) {
+                unsigned long r;
+                _BitScanReverse(&r, val);
+                return (unsigned)(31 - r);
+            } else {
+                /* Should not reach this code path */
+                __assume(0);
+            }
+#       endif
+#   elif defined(__GNUC__) && (__GNUC__ >= 4)
+        return (unsigned)__builtin_clz(val);
+#   else
+        return ZSTD_countLeadingZeros32_fallback(val);
+#   endif
+}
+
+MEM_STATIC unsigned ZSTD_countTrailingZeros64(U64 val)
+{
+    assert(val != 0);
+#   if defined(_MSC_VER) && defined(_WIN64)
+#       if STATIC_BMI2 == 1
+            return _tzcnt_u64(val);
+#       else
+            if (val != 0) {
+                unsigned long r;
+                _BitScanForward64(&r, val);
+                return (unsigned)r;
+            } else {
+                /* Should not reach this code path */
+                __assume(0);
+            }
+#       endif
+#   elif defined(__GNUC__) && (__GNUC__ >= 4) && defined(__LP64__)
+        return (unsigned)__builtin_ctzll(val);
+#   else
+        {
+            U32 mostSignificantWord = (U32)(val >> 32);
+            U32 leastSignificantWord = (U32)val;
+            if (leastSignificantWord == 0) {
+                return 32 + ZSTD_countTrailingZeros32(mostSignificantWord);
+            } else {
+                return ZSTD_countTrailingZeros32(leastSignificantWord);
+            }
+        }
+#   endif
+}
+
+MEM_STATIC unsigned ZSTD_countLeadingZeros64(U64 val)
+{
+    assert(val != 0);
+#   if defined(_MSC_VER) && defined(_WIN64)
+#       if STATIC_BMI2 == 1
+            return _lzcnt_u64(val);
+#       else
+            if (val != 0) {
+                unsigned long r;
+                _BitScanReverse64(&r, val);
+                return (unsigned)(63 - r);
+            } else {
+                /* Should not reach this code path */
+                __assume(0);
+            }
+#       endif
+#   elif defined(__GNUC__) && (__GNUC__ >= 4)
+        return (unsigned)(__builtin_clzll(val));
+#   else
+        {
+            U32 mostSignificantWord = (U32)(val >> 32);
+            U32 leastSignificantWord = (U32)val;
+            if (mostSignificantWord == 0) {
+                return 32 + ZSTD_countLeadingZeros32(leastSignificantWord);
+            } else {
+                return ZSTD_countLeadingZeros32(mostSignificantWord);
+            }
+        }
+#   endif
+}
+
+MEM_STATIC unsigned ZSTD_NbCommonBytes(size_t val)
+{
+    if (MEM_isLittleEndian()) {
+        if (MEM_64bits()) {
+            return ZSTD_countTrailingZeros64((U64)val) >> 3;
+        } else {
+            return ZSTD_countTrailingZeros32((U32)val) >> 3;
+        }
+    } else {  /* Big Endian CPU */
+        if (MEM_64bits()) {
+            return ZSTD_countLeadingZeros64((U64)val) >> 3;
+        } else {
+            return ZSTD_countLeadingZeros32((U32)val) >> 3;
+        }
+    }
+}
+
+MEM_STATIC unsigned ZSTD_highbit32(U32 val)   /* compress, dictBuilder, decodeCorpus */
+{
+    assert(val != 0);
+    return 31 - ZSTD_countLeadingZeros32(val);
+}
+
+#endif /* ZSTD_BITS_H */

external/zstd/common/bitstream.h

@@ -0,0 +1,437 @@
+/* ******************************************************************
+ * bitstream
+ * Part of FSE library
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ *
+ * You can contact the author at :
+ * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+****************************************************************** */
+#ifndef BITSTREAM_H_MODULE
+#define BITSTREAM_H_MODULE
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+/*
+*  This API consists of small unitary functions, which must be inlined for best performance.
+*  Since link-time-optimization is not available for all compilers,
+*  these functions are defined into a .h to be included.
+*/
+
+/*-****************************************
+*  Dependencies
+******************************************/
+#include "mem.h"            /* unaligned access routines */
+#include "compiler.h"       /* UNLIKELY() */
+#include "debug.h"          /* assert(), DEBUGLOG(), RAWLOG() */
+#include "error_private.h"  /* error codes and messages */
+#include "bits.h"           /* ZSTD_highbit32 */
+
+
+/*=========================================
+*  Target specific
+=========================================*/
+#ifndef ZSTD_NO_INTRINSICS
+#  if (defined(__BMI__) || defined(__BMI2__)) && defined(__GNUC__)
+#    include <immintrin.h>   /* support for bextr (experimental)/bzhi */
+#  elif defined(__ICCARM__)
+#    include <intrinsics.h>
+#  endif
+#endif
+
+#define STREAM_ACCUMULATOR_MIN_32  25
+#define STREAM_ACCUMULATOR_MIN_64  57
+#define STREAM_ACCUMULATOR_MIN    ((U32)(MEM_32bits() ? STREAM_ACCUMULATOR_MIN_32 : STREAM_ACCUMULATOR_MIN_64))
+
+
+/*-******************************************
+*  bitStream encoding API (write forward)
+********************************************/
+/* bitStream can mix input from multiple sources.
+ * A critical property of these streams is that they encode and decode in **reverse** direction.
+ * So the first bit sequence you add will be the last to be read, like a LIFO stack.
+ */
+typedef struct {
+    size_t bitContainer;
+    unsigned bitPos;
+    char*  startPtr;
+    char*  ptr;
+    char*  endPtr;
+} BIT_CStream_t;
+
+MEM_STATIC size_t BIT_initCStream(BIT_CStream_t* bitC, void* dstBuffer, size_t dstCapacity);
+MEM_STATIC void   BIT_addBits(BIT_CStream_t* bitC, size_t value, unsigned nbBits);
+MEM_STATIC void   BIT_flushBits(BIT_CStream_t* bitC);
+MEM_STATIC size_t BIT_closeCStream(BIT_CStream_t* bitC);
+
+/* Start with initCStream, providing the size of buffer to write into.
+*  bitStream will never write outside of this buffer.
+*  `dstCapacity` must be >= sizeof(bitD->bitContainer), otherwise @return will be an error code.
+*
+*  bits are first added to a local register.
+*  Local register is size_t, hence 64-bits on 64-bits systems, or 32-bits on 32-bits systems.
+*  Writing data into memory is an explicit operation, performed by the flushBits function.
+*  Hence keep track how many bits are potentially stored into local register to avoid register overflow.
+*  After a flushBits, a maximum of 7 bits might still be stored into local register.
+*
+*  Avoid storing elements of more than 24 bits if you want compatibility with 32-bits bitstream readers.
+*
+*  Last operation is to close the bitStream.
+*  The function returns the final size of CStream in bytes.
+*  If data couldn't fit into `dstBuffer`, it will return a 0 ( == not storable)
+*/
+
+
+/*-********************************************
+*  bitStream decoding API (read backward)
+**********************************************/
+typedef struct {
+    size_t   bitContainer;
+    unsigned bitsConsumed;
+    const char* ptr;
+    const char* start;
+    const char* limitPtr;
+} BIT_DStream_t;
+
+typedef enum { BIT_DStream_unfinished = 0,
+               BIT_DStream_endOfBuffer = 1,
+               BIT_DStream_completed = 2,
+               BIT_DStream_overflow = 3 } BIT_DStream_status;  /* result of BIT_reloadDStream() */
+               /* 1,2,4,8 would be better for bitmap combinations, but slows down performance a bit ... :( */
+
+MEM_STATIC size_t   BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, size_t srcSize);
+MEM_STATIC size_t   BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits);
+MEM_STATIC BIT_DStream_status BIT_reloadDStream(BIT_DStream_t* bitD);
+MEM_STATIC unsigned BIT_endOfDStream(const BIT_DStream_t* bitD);
+
+
+/* Start by invoking BIT_initDStream().
+*  A chunk of the bitStream is then stored into a local register.
+*  Local register size is 64-bits on 64-bits systems, 32-bits on 32-bits systems (size_t).
+*  You can then retrieve bitFields stored into the local register, **in reverse order**.
+*  Local register is explicitly reloaded from memory by the BIT_reloadDStream() method.
+*  A reload guarantee a minimum of ((8*sizeof(bitD->bitContainer))-7) bits when its result is BIT_DStream_unfinished.
+*  Otherwise, it can be less than that, so proceed accordingly.
+*  Checking if DStream has reached its end can be performed with BIT_endOfDStream().
+*/
+
+
+/*-****************************************
+*  unsafe API
+******************************************/
+MEM_STATIC void BIT_addBitsFast(BIT_CStream_t* bitC, size_t value, unsigned nbBits);
+/* faster, but works only if value is "clean", meaning all high bits above nbBits are 0 */
+
+MEM_STATIC void BIT_flushBitsFast(BIT_CStream_t* bitC);
+/* unsafe version; does not check buffer overflow */
+
+MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits);
+/* faster, but works only if nbBits >= 1 */
+
+/*=====    Local Constants   =====*/
+static const unsigned BIT_mask[] = {
+    0,          1,         3,         7,         0xF,       0x1F,
+    0x3F,       0x7F,      0xFF,      0x1FF,     0x3FF,     0x7FF,
+    0xFFF,      0x1FFF,    0x3FFF,    0x7FFF,    0xFFFF,    0x1FFFF,
+    0x3FFFF,    0x7FFFF,   0xFFFFF,   0x1FFFFF,  0x3FFFFF,  0x7FFFFF,
+    0xFFFFFF,   0x1FFFFFF, 0x3FFFFFF, 0x7FFFFFF, 0xFFFFFFF, 0x1FFFFFFF,
+    0x3FFFFFFF, 0x7FFFFFFF}; /* up to 31 bits */
+#define BIT_MASK_SIZE (sizeof(BIT_mask) / sizeof(BIT_mask[0]))
+
+/*-**************************************************************
+*  bitStream encoding
+****************************************************************/
+/*! BIT_initCStream() :
+ *  `dstCapacity` must be > sizeof(size_t)
+ *  @return : 0 if success,
+ *            otherwise an error code (can be tested using ERR_isError()) */
+MEM_STATIC size_t BIT_initCStream(BIT_CStream_t* bitC,
+                                  void* startPtr, size_t dstCapacity)
+{
+    bitC->bitContainer = 0;
+    bitC->bitPos = 0;
+    bitC->startPtr = (char*)startPtr;
+    bitC->ptr = bitC->startPtr;
+    bitC->endPtr = bitC->startPtr + dstCapacity - sizeof(bitC->bitContainer);
+    if (dstCapacity <= sizeof(bitC->bitContainer)) return ERROR(dstSize_tooSmall);
+    return 0;
+}
+
+MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits)
+{
+#if defined(STATIC_BMI2) && STATIC_BMI2 == 1 && !defined(ZSTD_NO_INTRINSICS)
+    return  _bzhi_u64(bitContainer, nbBits);
+#else
+    assert(nbBits < BIT_MASK_SIZE);
+    return bitContainer & BIT_mask[nbBits];
+#endif
+}
+
+/*! BIT_addBits() :
+ *  can add up to 31 bits into `bitC`.
+ *  Note : does not check for register overflow ! */
+MEM_STATIC void BIT_addBits(BIT_CStream_t* bitC,
+                            size_t value, unsigned nbBits)
+{
+    DEBUG_STATIC_ASSERT(BIT_MASK_SIZE == 32);
+    assert(nbBits < BIT_MASK_SIZE);
+    assert(nbBits + bitC->bitPos < sizeof(bitC->bitContainer) * 8);
+    bitC->bitContainer |= BIT_getLowerBits(value, nbBits) << bitC->bitPos;
+    bitC->bitPos += nbBits;
+}
+
+/*! BIT_addBitsFast() :
+ *  works only if `value` is _clean_,
+ *  meaning all high bits above nbBits are 0 */
+MEM_STATIC void BIT_addBitsFast(BIT_CStream_t* bitC,
+                                size_t value, unsigned nbBits)
+{
+    assert((value>>nbBits) == 0);
+    assert(nbBits + bitC->bitPos < sizeof(bitC->bitContainer) * 8);
+    bitC->bitContainer |= value << bitC->bitPos;
+    bitC->bitPos += nbBits;
+}
+
+/*! BIT_flushBitsFast() :
+ *  assumption : bitContainer has not overflowed
+ *  unsafe version; does not check buffer overflow */
+MEM_STATIC void BIT_flushBitsFast(BIT_CStream_t* bitC)
+{
+    size_t const nbBytes = bitC->bitPos >> 3;
+    assert(bitC->bitPos < sizeof(bitC->bitContainer) * 8);
+    assert(bitC->ptr <= bitC->endPtr);
+    MEM_writeLEST(bitC->ptr, bitC->bitContainer);
+    bitC->ptr += nbBytes;
+    bitC->bitPos &= 7;
+    bitC->bitContainer >>= nbBytes*8;
+}
+
+/*! BIT_flushBits() :
+ *  assumption : bitContainer has not overflowed
+ *  safe version; check for buffer overflow, and prevents it.
+ *  note : does not signal buffer overflow.
+ *  overflow will be revealed later on using BIT_closeCStream() */
+MEM_STATIC void BIT_flushBits(BIT_CStream_t* bitC)
+{
+    size_t const nbBytes = bitC->bitPos >> 3;
+    assert(bitC->bitPos < sizeof(bitC->bitContainer) * 8);
+    assert(bitC->ptr <= bitC->endPtr);
+    MEM_writeLEST(bitC->ptr, bitC->bitContainer);
+    bitC->ptr += nbBytes;
+    if (bitC->ptr > bitC->endPtr) bitC->ptr = bitC->endPtr;
+    bitC->bitPos &= 7;
+    bitC->bitContainer >>= nbBytes*8;
+}
+
+/*! BIT_closeCStream() :
+ *  @return : size of CStream, in bytes,
+ *            or 0 if it could not fit into dstBuffer */
+MEM_STATIC size_t BIT_closeCStream(BIT_CStream_t* bitC)
+{
+    BIT_addBitsFast(bitC, 1, 1);   /* endMark */
+    BIT_flushBits(bitC);
+    if (bitC->ptr >= bitC->endPtr) return 0; /* overflow detected */
+    return (bitC->ptr - bitC->startPtr) + (bitC->bitPos > 0);
+}
+
+
+/*-********************************************************
+*  bitStream decoding
+**********************************************************/
+/*! BIT_initDStream() :
+ *  Initialize a BIT_DStream_t.
+ * `bitD` : a pointer to an already allocated BIT_DStream_t structure.
+ * `srcSize` must be the *exact* size of the bitStream, in bytes.
+ * @return : size of stream (== srcSize), or an errorCode if a problem is detected
+ */
+MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, size_t srcSize)
+{
+    if (srcSize < 1) { ZSTD_memset(bitD, 0, sizeof(*bitD)); return ERROR(srcSize_wrong); }
+
+    bitD->start = (const char*)srcBuffer;
+    bitD->limitPtr = bitD->start + sizeof(bitD->bitContainer);
+
+    if (srcSize >=  sizeof(bitD->bitContainer)) {  /* normal case */
+        bitD->ptr   = (const char*)srcBuffer + srcSize - sizeof(bitD->bitContainer);
+        bitD->bitContainer = MEM_readLEST(bitD->ptr);
+        { BYTE const lastByte = ((const BYTE*)srcBuffer)[srcSize-1];
+          bitD->bitsConsumed = lastByte ? 8 - ZSTD_highbit32(lastByte) : 0;  /* ensures bitsConsumed is always set */
+          if (lastByte == 0) return ERROR(GENERIC); /* endMark not present */ }
+    } else {
+        bitD->ptr   = bitD->start;
+        bitD->bitContainer = *(const BYTE*)(bitD->start);
+        switch(srcSize)
+        {
+        case 7: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[6]) << (sizeof(bitD->bitContainer)*8 - 16);
+                ZSTD_FALLTHROUGH;
+
+        case 6: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[5]) << (sizeof(bitD->bitContainer)*8 - 24);
+                ZSTD_FALLTHROUGH;
+
+        case 5: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[4]) << (sizeof(bitD->bitContainer)*8 - 32);
+                ZSTD_FALLTHROUGH;
+
+        case 4: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[3]) << 24;
+                ZSTD_FALLTHROUGH;
+
+        case 3: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[2]) << 16;
+                ZSTD_FALLTHROUGH;
+
+        case 2: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[1]) <<  8;
+                ZSTD_FALLTHROUGH;
+
+        default: break;
+        }
+        {   BYTE const lastByte = ((const BYTE*)srcBuffer)[srcSize-1];
+            bitD->bitsConsumed = lastByte ? 8 - ZSTD_highbit32(lastByte) : 0;
+            if (lastByte == 0) return ERROR(corruption_detected);  /* endMark not present */
+        }
+        bitD->bitsConsumed += (U32)(sizeof(bitD->bitContainer) - srcSize)*8;
+    }
+
+    return srcSize;
+}
+
+MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getUpperBits(size_t bitContainer, U32 const start)
+{
+    return bitContainer >> start;
+}
+
+MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getMiddleBits(size_t bitContainer, U32 const start, U32 const nbBits)
+{
+    U32 const regMask = sizeof(bitContainer)*8 - 1;
+    /* if start > regMask, bitstream is corrupted, and result is undefined */
+    assert(nbBits < BIT_MASK_SIZE);
+    /* x86 transform & ((1 << nbBits) - 1) to bzhi instruction, it is better
+     * than accessing memory. When bmi2 instruction is not present, we consider
+     * such cpus old (pre-Haswell, 2013) and their performance is not of that
+     * importance.
+     */
+#if defined(__x86_64__) || defined(_M_X86)
+    return (bitContainer >> (start & regMask)) & ((((U64)1) << nbBits) - 1);
+#else
+    return (bitContainer >> (start & regMask)) & BIT_mask[nbBits];
+#endif
+}
+
+/*! BIT_lookBits() :
+ *  Provides next n bits from local register.
+ *  local register is not modified.
+ *  On 32-bits, maxNbBits==24.
+ *  On 64-bits, maxNbBits==56.
+ * @return : value extracted */
+MEM_STATIC  FORCE_INLINE_ATTR size_t BIT_lookBits(const BIT_DStream_t*  bitD, U32 nbBits)
+{
+    /* arbitrate between double-shift and shift+mask */
+#if 1
+    /* if bitD->bitsConsumed + nbBits > sizeof(bitD->bitContainer)*8,
+     * bitstream is likely corrupted, and result is undefined */
+    return BIT_getMiddleBits(bitD->bitContainer, (sizeof(bitD->bitContainer)*8) - bitD->bitsConsumed - nbBits, nbBits);
+#else
+    /* this code path is slower on my os-x laptop */
+    U32 const regMask = sizeof(bitD->bitContainer)*8 - 1;
+    return ((bitD->bitContainer << (bitD->bitsConsumed & regMask)) >> 1) >> ((regMask-nbBits) & regMask);
+#endif
+}
+
+/*! BIT_lookBitsFast() :
+ *  unsafe version; only works if nbBits >= 1 */
+MEM_STATIC size_t BIT_lookBitsFast(const BIT_DStream_t* bitD, U32 nbBits)
+{
+    U32 const regMask = sizeof(bitD->bitContainer)*8 - 1;
+    assert(nbBits >= 1);
+    return (bitD->bitContainer << (bitD->bitsConsumed & regMask)) >> (((regMask+1)-nbBits) & regMask);
+}
+
+MEM_STATIC FORCE_INLINE_ATTR void BIT_skipBits(BIT_DStream_t* bitD, U32 nbBits)
+{
+    bitD->bitsConsumed += nbBits;
+}
+
+/*! BIT_readBits() :
+ *  Read (consume) next n bits from local register and update.
+ *  Pay attention to not read more than nbBits contained into local register.
+ * @return : extracted value. */
+MEM_STATIC FORCE_INLINE_ATTR size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits)
+{
+    size_t const value = BIT_lookBits(bitD, nbBits);
+    BIT_skipBits(bitD, nbBits);
+    return value;
+}
+
+/*! BIT_readBitsFast() :
+ *  unsafe version; only works if nbBits >= 1 */
+MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits)
+{
+    size_t const value = BIT_lookBitsFast(bitD, nbBits);
+    assert(nbBits >= 1);
+    BIT_skipBits(bitD, nbBits);
+    return value;
+}
+
+/*! BIT_reloadDStreamFast() :
+ *  Similar to BIT_reloadDStream(), but with two differences:
+ *  1. bitsConsumed <= sizeof(bitD->bitContainer)*8 must hold!
+ *  2. Returns BIT_DStream_overflow when bitD->ptr < bitD->limitPtr, at this
+ *     point you must use BIT_reloadDStream() to reload.
+ */
+MEM_STATIC BIT_DStream_status BIT_reloadDStreamFast(BIT_DStream_t* bitD)
+{
+    if (UNLIKELY(bitD->ptr < bitD->limitPtr))
+        return BIT_DStream_overflow;
+    assert(bitD->bitsConsumed <= sizeof(bitD->bitContainer)*8);
+    bitD->ptr -= bitD->bitsConsumed >> 3;
+    bitD->bitsConsumed &= 7;
+    bitD->bitContainer = MEM_readLEST(bitD->ptr);
+    return BIT_DStream_unfinished;
+}
+
+/*! BIT_reloadDStream() :
+ *  Refill `bitD` from buffer previously set in BIT_initDStream() .
+ *  This function is safe, it guarantees it will not read beyond src buffer.
+ * @return : status of `BIT_DStream_t` internal register.
+ *           when status == BIT_DStream_unfinished, internal register is filled with at least 25 or 57 bits */
+MEM_STATIC BIT_DStream_status BIT_reloadDStream(BIT_DStream_t* bitD)
+{
+    if (bitD->bitsConsumed > (sizeof(bitD->bitContainer)*8))  /* overflow detected, like end of stream */
+        return BIT_DStream_overflow;
+
+    if (bitD->ptr >= bitD->limitPtr) {
+        return BIT_reloadDStreamFast(bitD);
+    }
+    if (bitD->ptr == bitD->start) {
+        if (bitD->bitsConsumed < sizeof(bitD->bitContainer)*8) return BIT_DStream_endOfBuffer;
+        return BIT_DStream_completed;
+    }
+    /* start < ptr < limitPtr */
+    {   U32 nbBytes = bitD->bitsConsumed >> 3;
+        BIT_DStream_status result = BIT_DStream_unfinished;
+        if (bitD->ptr - nbBytes < bitD->start) {
+            nbBytes = (U32)(bitD->ptr - bitD->start);  /* ptr > start */
+            result = BIT_DStream_endOfBuffer;
+        }
+        bitD->ptr -= nbBytes;
+        bitD->bitsConsumed -= nbBytes*8;
+        bitD->bitContainer = MEM_readLEST(bitD->ptr);   /* reminder : srcSize > sizeof(bitD->bitContainer), otherwise bitD->ptr == bitD->start */
+        return result;
+    }
+}
+
+/*! BIT_endOfDStream() :
+ * @return : 1 if DStream has _exactly_ reached its end (all bits consumed).
+ */
+MEM_STATIC unsigned BIT_endOfDStream(const BIT_DStream_t* DStream)
+{
+    return ((DStream->ptr == DStream->start) && (DStream->bitsConsumed == sizeof(DStream->bitContainer)*8));
+}
+
+#if defined (__cplusplus)
+}
+#endif
+
+#endif /* BITSTREAM_H_MODULE */

external/zstd/common/compiler.h

@@ -0,0 +1,343 @@
+/*
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * All rights reserved.
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+ */
+
+#ifndef ZSTD_COMPILER_H
+#define ZSTD_COMPILER_H
+
+#include "portability_macros.h"
+
+/*-*******************************************************
+*  Compiler specifics
+*********************************************************/
+/* force inlining */
+
+#if !defined(ZSTD_NO_INLINE)
+#if (defined(__GNUC__) && !defined(__STRICT_ANSI__)) || defined(__cplusplus) || defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L   /* C99 */
+#  define INLINE_KEYWORD inline
+#else
+#  define INLINE_KEYWORD
+#endif
+
+#if defined(__GNUC__) || defined(__ICCARM__)
+#  define FORCE_INLINE_ATTR __attribute__((always_inline))
+#elif defined(_MSC_VER)
+#  define FORCE_INLINE_ATTR __forceinline
+#else
+#  define FORCE_INLINE_ATTR
+#endif
+
+#else
+
+#define INLINE_KEYWORD
+#define FORCE_INLINE_ATTR
+
+#endif
+
+/**
+  On MSVC qsort requires that functions passed into it use the __cdecl calling conversion(CC).
+  This explicitly marks such functions as __cdecl so that the code will still compile
+  if a CC other than __cdecl has been made the default.
+*/
+#if  defined(_MSC_VER)
+#  define WIN_CDECL __cdecl
+#else
+#  define WIN_CDECL
+#endif
+
+/**
+ * FORCE_INLINE_TEMPLATE is used to define C "templates", which take constant
+ * parameters. They must be inlined for the compiler to eliminate the constant
+ * branches.
+ */
+#define FORCE_INLINE_TEMPLATE static INLINE_KEYWORD FORCE_INLINE_ATTR
+/**
+ * HINT_INLINE is used to help the compiler generate better code. It is *not*
+ * used for "templates", so it can be tweaked based on the compilers
+ * performance.
+ *
+ * gcc-4.8 and gcc-4.9 have been shown to benefit from leaving off the
+ * always_inline attribute.
+ *
+ * clang up to 5.0.0 (trunk) benefit tremendously from the always_inline
+ * attribute.
+ */
+#if !defined(__clang__) && defined(__GNUC__) && __GNUC__ >= 4 && __GNUC_MINOR__ >= 8 && __GNUC__ < 5
+#  define HINT_INLINE static INLINE_KEYWORD
+#else
+#  define HINT_INLINE static INLINE_KEYWORD FORCE_INLINE_ATTR
+#endif
+
+/* UNUSED_ATTR tells the compiler it is okay if the function is unused. */
+#if defined(__GNUC__)
+#  define UNUSED_ATTR __attribute__((unused))
+#else
+#  define UNUSED_ATTR
+#endif
+
+/* force no inlining */
+#ifdef _MSC_VER
+#  define FORCE_NOINLINE static __declspec(noinline)
+#else
+#  if defined(__GNUC__) || defined(__ICCARM__)
+#    define FORCE_NOINLINE static __attribute__((__noinline__))
+#  else
+#    define FORCE_NOINLINE static
+#  endif
+#endif
+
+
+/* target attribute */
+#if defined(__GNUC__) || defined(__ICCARM__)
+#  define TARGET_ATTRIBUTE(target) __attribute__((__target__(target)))
+#else
+#  define TARGET_ATTRIBUTE(target)
+#endif
+
+/* Target attribute for BMI2 dynamic dispatch.
+ * Enable lzcnt, bmi, and bmi2.
+ * We test for bmi1 & bmi2. lzcnt is included in bmi1.
+ */
+#define BMI2_TARGET_ATTRIBUTE TARGET_ATTRIBUTE("lzcnt,bmi,bmi2")
+
+/* prefetch
+ * can be disabled, by declaring NO_PREFETCH build macro */
+#if defined(NO_PREFETCH)
+#  define PREFETCH_L1(ptr)  (void)(ptr)  /* disabled */
+#  define PREFETCH_L2(ptr)  (void)(ptr)  /* disabled */
+#else
+#  if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86))  /* _mm_prefetch() is not defined outside of x86/x64 */
+#    include <mmintrin.h>   /* https://msdn.microsoft.com/fr-fr/library/84szxsww(v=vs.90).aspx */
+#    define PREFETCH_L1(ptr)  _mm_prefetch((const char*)(ptr), _MM_HINT_T0)
+#    define PREFETCH_L2(ptr)  _mm_prefetch((const char*)(ptr), _MM_HINT_T1)
+#  elif defined(__GNUC__) && ( (__GNUC__ >= 4) || ( (__GNUC__ == 3) && (__GNUC_MINOR__ >= 1) ) )
+#    define PREFETCH_L1(ptr)  __builtin_prefetch((ptr), 0 /* rw==read */, 3 /* locality */)
+#    define PREFETCH_L2(ptr)  __builtin_prefetch((ptr), 0 /* rw==read */, 2 /* locality */)
+#  elif defined(__aarch64__)
+#    define PREFETCH_L1(ptr)  __asm__ __volatile__("prfm pldl1keep, %0" ::"Q"(*(ptr)))
+#    define PREFETCH_L2(ptr)  __asm__ __volatile__("prfm pldl2keep, %0" ::"Q"(*(ptr)))
+#  else
+#    define PREFETCH_L1(ptr) (void)(ptr)  /* disabled */
+#    define PREFETCH_L2(ptr) (void)(ptr)  /* disabled */
+#  endif
+#endif  /* NO_PREFETCH */
+
+#define CACHELINE_SIZE 64
+
+#define PREFETCH_AREA(p, s)  {            \
+    const char* const _ptr = (const char*)(p);  \
+    size_t const _size = (size_t)(s);     \
+    size_t _pos;                          \
+    for (_pos=0; _pos<_size; _pos+=CACHELINE_SIZE) {  \
+        PREFETCH_L2(_ptr + _pos);         \
+    }                                     \
+}
+
+/* vectorization
+ * older GCC (pre gcc-4.3 picked as the cutoff) uses a different syntax,
+ * and some compilers, like Intel ICC and MCST LCC, do not support it at all. */
+#if !defined(__INTEL_COMPILER) && !defined(__clang__) && defined(__GNUC__) && !defined(__LCC__)
+#  if (__GNUC__ == 4 && __GNUC_MINOR__ > 3) || (__GNUC__ >= 5)
+#    define DONT_VECTORIZE __attribute__((optimize("no-tree-vectorize")))
+#  else
+#    define DONT_VECTORIZE _Pragma("GCC optimize(\"no-tree-vectorize\")")
+#  endif
+#else
+#  define DONT_VECTORIZE
+#endif
+
+/* Tell the compiler that a branch is likely or unlikely.
+ * Only use these macros if it causes the compiler to generate better code.
+ * If you can remove a LIKELY/UNLIKELY annotation without speed changes in gcc
+ * and clang, please do.
+ */
+#if defined(__GNUC__)
+#define LIKELY(x) (__builtin_expect((x), 1))
+#define UNLIKELY(x) (__builtin_expect((x), 0))
+#else
+#define LIKELY(x) (x)
+#define UNLIKELY(x) (x)
+#endif
+
+#if __has_builtin(__builtin_unreachable) || (defined(__GNUC__) && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 5)))
+#  define ZSTD_UNREACHABLE { assert(0), __builtin_unreachable(); }
+#else
+#  define ZSTD_UNREACHABLE { assert(0); }
+#endif
+
+/* disable warnings */
+#ifdef _MSC_VER    /* Visual Studio */
+#  include <intrin.h>                    /* For Visual 2005 */
+#  pragma warning(disable : 4100)        /* disable: C4100: unreferenced formal parameter */
+#  pragma warning(disable : 4127)        /* disable: C4127: conditional expression is constant */
+#  pragma warning(disable : 4204)        /* disable: C4204: non-constant aggregate initializer */
+#  pragma warning(disable : 4214)        /* disable: C4214: non-int bitfields */
+#  pragma warning(disable : 4324)        /* disable: C4324: padded structure */
+#endif
+
+/*Like DYNAMIC_BMI2 but for compile time determination of BMI2 support*/
+#ifndef STATIC_BMI2
+#  if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86))
+#    ifdef __AVX2__  //MSVC does not have a BMI2 specific flag, but every CPU that supports AVX2 also supports BMI2
+#       define STATIC_BMI2 1
+#    endif
+#  elif defined(__BMI2__) && defined(__x86_64__) && defined(__GNUC__)
+#    define STATIC_BMI2 1
+#  endif
+#endif
+
+#ifndef STATIC_BMI2
+    #define STATIC_BMI2 0
+#endif
+
+/* compile time determination of SIMD support */
+#if !defined(ZSTD_NO_INTRINSICS)
+#  if defined(__SSE2__) || defined(_M_AMD64) || (defined (_M_IX86) && defined(_M_IX86_FP) && (_M_IX86_FP >= 2))
+#    define ZSTD_ARCH_X86_SSE2
+#  endif
+#  if defined(__ARM_NEON) || defined(_M_ARM64)
+#    define ZSTD_ARCH_ARM_NEON
+#  endif
+#
+#  if defined(ZSTD_ARCH_X86_SSE2)
+#    include <emmintrin.h>
+#  elif defined(ZSTD_ARCH_ARM_NEON)
+#    include <arm_neon.h>
+#  endif
+#endif
+
+/* C-language Attributes are added in C23. */
+#if defined(__STDC_VERSION__) && (__STDC_VERSION__ > 201710L) && defined(__has_c_attribute)
+# define ZSTD_HAS_C_ATTRIBUTE(x) __has_c_attribute(x)
+#else
+# define ZSTD_HAS_C_ATTRIBUTE(x) 0
+#endif
+
+/* Only use C++ attributes in C++. Some compilers report support for C++
+ * attributes when compiling with C.
+ */
+#if defined(__cplusplus) && defined(__has_cpp_attribute)
+# define ZSTD_HAS_CPP_ATTRIBUTE(x) __has_cpp_attribute(x)
+#else
+# define ZSTD_HAS_CPP_ATTRIBUTE(x) 0
+#endif
+
+/* Define ZSTD_FALLTHROUGH macro for annotating switch case with the 'fallthrough' attribute.
+ * - C23: https://en.cppreference.com/w/c/language/attributes/fallthrough
+ * - CPP17: https://en.cppreference.com/w/cpp/language/attributes/fallthrough
+ * - Else: __attribute__((__fallthrough__))
+ */
+#ifndef ZSTD_FALLTHROUGH
+# if ZSTD_HAS_C_ATTRIBUTE(fallthrough)
+#  define ZSTD_FALLTHROUGH [[fallthrough]]
+# elif ZSTD_HAS_CPP_ATTRIBUTE(fallthrough)
+#  define ZSTD_FALLTHROUGH [[fallthrough]]
+# elif __has_attribute(__fallthrough__)
+/* Leading semicolon is to satisfy gcc-11 with -pedantic. Without the semicolon
+ * gcc complains about: a label can only be part of a statement and a declaration is not a statement.
+ */
+#  define ZSTD_FALLTHROUGH ; __attribute__((__fallthrough__))
+# else
+#  define ZSTD_FALLTHROUGH
+# endif
+#endif
+
+/*-**************************************************************
+*  Alignment check
+*****************************************************************/
+
+/* this test was initially positioned in mem.h,
+ * but this file is removed (or replaced) for linux kernel
+ * so it's now hosted in compiler.h,
+ * which remains valid for both user & kernel spaces.
+ */
+
+#ifndef ZSTD_ALIGNOF
+# if defined(__GNUC__) || defined(_MSC_VER)
+/* covers gcc, clang & MSVC */
+/* note : this section must come first, before C11,
+ * due to a limitation in the kernel source generator */
+#  define ZSTD_ALIGNOF(T) __alignof(T)
+
+# elif defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 201112L)
+/* C11 support */
+#  include <stdalign.h>
+#  define ZSTD_ALIGNOF(T) alignof(T)
+
+# else
+/* No known support for alignof() - imperfect backup */
+#  define ZSTD_ALIGNOF(T) (sizeof(void*) < sizeof(T) ? sizeof(void*) : sizeof(T))
+
+# endif
+#endif /* ZSTD_ALIGNOF */
+
+/*-**************************************************************
+*  Sanitizer
+*****************************************************************/
+
+#if ZSTD_MEMORY_SANITIZER
+/* Not all platforms that support msan provide sanitizers/msan_interface.h.
+ * We therefore declare the functions we need ourselves, rather than trying to
+ * include the header file... */
+#include <stddef.h>  /* size_t */
+#define ZSTD_DEPS_NEED_STDINT
+#include "zstd_deps.h"  /* intptr_t */
+
+/* Make memory region fully initialized (without changing its contents). */
+void __msan_unpoison(const volatile void *a, size_t size);
+
+/* Make memory region fully uninitialized (without changing its contents).
+   This is a legacy interface that does not update origin information. Use
+   __msan_allocated_memory() instead. */
+void __msan_poison(const volatile void *a, size_t size);
+
+/* Returns the offset of the first (at least partially) poisoned byte in the
+   memory range, or -1 if the whole range is good. */
+intptr_t __msan_test_shadow(const volatile void *x, size_t size);
+#endif
+
+#if ZSTD_ADDRESS_SANITIZER
+/* Not all platforms that support asan provide sanitizers/asan_interface.h.
+ * We therefore declare the functions we need ourselves, rather than trying to
+ * include the header file... */
+#include <stddef.h>  /* size_t */
+
+/**
+ * Marks a memory region (<c>[addr, addr+size)</c>) as unaddressable.
+ *
+ * This memory must be previously allocated by your program. Instrumented
+ * code is forbidden from accessing addresses in this region until it is
+ * unpoisoned. This function is not guaranteed to poison the entire region -
+ * it could poison only a subregion of <c>[addr, addr+size)</c> due to ASan
+ * alignment restrictions.
+ *
+ * \note This function is not thread-safe because no two threads can poison or
+ * unpoison memory in the same memory region simultaneously.
+ *
+ * \param addr Start of memory region.
+ * \param size Size of memory region. */
+void __asan_poison_memory_region(void const volatile *addr, size_t size);
+
+/**
+ * Marks a memory region (<c>[addr, addr+size)</c>) as addressable.
+ *
+ * This memory must be previously allocated by your program. Accessing
+ * addresses in this region is allowed until this region is poisoned again.
+ * This function could unpoison a super-region of <c>[addr, addr+size)</c> due
+ * to ASan alignment restrictions.
+ *
+ * \note This function is not thread-safe because no two threads can
+ * poison or unpoison memory in the same memory region simultaneously.
+ *
+ * \param addr Start of memory region.
+ * \param size Size of memory region. */
+void __asan_unpoison_memory_region(void const volatile *addr, size_t size);
+#endif
+
+#endif /* ZSTD_COMPILER_H */

external/zstd/common/cpu.h

@@ -0,0 +1,213 @@
+/*
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * All rights reserved.
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+ */
+
+#ifndef ZSTD_COMMON_CPU_H
+#define ZSTD_COMMON_CPU_H
+
+/**
+ * Implementation taken from folly/CpuId.h
+ * https://github.com/facebook/folly/blob/master/folly/CpuId.h
+ */
+
+#include "mem.h"
+
+#ifdef _MSC_VER
+#include <intrin.h>
+#endif
+
+typedef struct {
+    U32 f1c;
+    U32 f1d;
+    U32 f7b;
+    U32 f7c;
+} ZSTD_cpuid_t;
+
+MEM_STATIC ZSTD_cpuid_t ZSTD_cpuid(void) {
+    U32 f1c = 0;
+    U32 f1d = 0;
+    U32 f7b = 0;
+    U32 f7c = 0;
+#if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_IX86))
+    int reg[4];
+    __cpuid((int*)reg, 0);
+    {
+        int const n = reg[0];
+        if (n >= 1) {
+            __cpuid((int*)reg, 1);
+            f1c = (U32)reg[2];
+            f1d = (U32)reg[3];
+        }
+        if (n >= 7) {
+            __cpuidex((int*)reg, 7, 0);
+            f7b = (U32)reg[1];
+            f7c = (U32)reg[2];
+        }
+    }
+#elif defined(__i386__) && defined(__PIC__) && !defined(__clang__) && defined(__GNUC__)
+    /* The following block like the normal cpuid branch below, but gcc
+     * reserves ebx for use of its pic register so we must specially
+     * handle the save and restore to avoid clobbering the register
+     */
+    U32 n;
+    __asm__(
+        "pushl %%ebx\n\t"
+        "cpuid\n\t"
+        "popl %%ebx\n\t"
+        : "=a"(n)
+        : "a"(0)
+        : "ecx", "edx");
+    if (n >= 1) {
+      U32 f1a;
+      __asm__(
+          "pushl %%ebx\n\t"
+          "cpuid\n\t"
+          "popl %%ebx\n\t"
+          : "=a"(f1a), "=c"(f1c), "=d"(f1d)
+          : "a"(1));
+    }
+    if (n >= 7) {
+      __asm__(
+          "pushl %%ebx\n\t"
+          "cpuid\n\t"
+          "movl %%ebx, %%eax\n\t"
+          "popl %%ebx"
+          : "=a"(f7b), "=c"(f7c)
+          : "a"(7), "c"(0)
+          : "edx");
+    }
+#elif defined(__x86_64__) || defined(_M_X64) || defined(__i386__)
+    U32 n;
+    __asm__("cpuid" : "=a"(n) : "a"(0) : "ebx", "ecx", "edx");
+    if (n >= 1) {
+      U32 f1a;
+      __asm__("cpuid" : "=a"(f1a), "=c"(f1c), "=d"(f1d) : "a"(1) : "ebx");
+    }
+    if (n >= 7) {
+      U32 f7a;
+      __asm__("cpuid"
+              : "=a"(f7a), "=b"(f7b), "=c"(f7c)
+              : "a"(7), "c"(0)
+              : "edx");
+    }
+#endif
+    {
+        ZSTD_cpuid_t cpuid;
+        cpuid.f1c = f1c;
+        cpuid.f1d = f1d;
+        cpuid.f7b = f7b;
+        cpuid.f7c = f7c;
+        return cpuid;
+    }
+}
+
+#define X(name, r, bit)                                                        \
+  MEM_STATIC int ZSTD_cpuid_##name(ZSTD_cpuid_t const cpuid) {                 \
+    return ((cpuid.r) & (1U << bit)) != 0;                                     \
+  }
+
+/* cpuid(1): Processor Info and Feature Bits. */
+#define C(name, bit) X(name, f1c, bit)
+  C(sse3, 0)
+  C(pclmuldq, 1)
+  C(dtes64, 2)
+  C(monitor, 3)
+  C(dscpl, 4)
+  C(vmx, 5)
+  C(smx, 6)
+  C(eist, 7)
+  C(tm2, 8)
+  C(ssse3, 9)
+  C(cnxtid, 10)
+  C(fma, 12)
+  C(cx16, 13)
+  C(xtpr, 14)
+  C(pdcm, 15)
+  C(pcid, 17)
+  C(dca, 18)
+  C(sse41, 19)
+  C(sse42, 20)
+  C(x2apic, 21)
+  C(movbe, 22)
+  C(popcnt, 23)
+  C(tscdeadline, 24)
+  C(aes, 25)
+  C(xsave, 26)
+  C(osxsave, 27)
+  C(avx, 28)
+  C(f16c, 29)
+  C(rdrand, 30)
+#undef C
+#define D(name, bit) X(name, f1d, bit)
+  D(fpu, 0)
+  D(vme, 1)
+  D(de, 2)
+  D(pse, 3)
+  D(tsc, 4)
+  D(msr, 5)
+  D(pae, 6)
+  D(mce, 7)
+  D(cx8, 8)
+  D(apic, 9)
+  D(sep, 11)
+  D(mtrr, 12)
+  D(pge, 13)
+  D(mca, 14)
+  D(cmov, 15)
+  D(pat, 16)
+  D(pse36, 17)
+  D(psn, 18)
+  D(clfsh, 19)
+  D(ds, 21)
+  D(acpi, 22)
+  D(mmx, 23)
+  D(fxsr, 24)
+  D(sse, 25)
+  D(sse2, 26)
+  D(ss, 27)
+  D(htt, 28)
+  D(tm, 29)
+  D(pbe, 31)
+#undef D
+
+/* cpuid(7): Extended Features. */
+#define B(name, bit) X(name, f7b, bit)
+  B(bmi1, 3)
+  B(hle, 4)
+  B(avx2, 5)
+  B(smep, 7)
+  B(bmi2, 8)
+  B(erms, 9)
+  B(invpcid, 10)
+  B(rtm, 11)
+  B(mpx, 14)
+  B(avx512f, 16)
+  B(avx512dq, 17)
+  B(rdseed, 18)
+  B(adx, 19)
+  B(smap, 20)
+  B(avx512ifma, 21)
+  B(pcommit, 22)
+  B(clflushopt, 23)
+  B(clwb, 24)
+  B(avx512pf, 26)
+  B(avx512er, 27)
+  B(avx512cd, 28)
+  B(sha, 29)
+  B(avx512bw, 30)
+  B(avx512vl, 31)
+#undef B
+#define C(name, bit) X(name, f7c, bit)
+  C(prefetchwt1, 0)
+  C(avx512vbmi, 1)
+#undef C
+
+#undef X
+
+#endif /* ZSTD_COMMON_CPU_H */

external/zstd/common/debug.c

@@ -0,0 +1,24 @@
+/* ******************************************************************
+ * debug
+ * Part of FSE library
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ *
+ * You can contact the author at :
+ * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+****************************************************************** */
+
+
+/*
+ * This module only hosts one global variable
+ * which can be used to dynamically influence the verbosity of traces,
+ * such as DEBUGLOG and RAWLOG
+ */
+
+#include "debug.h"
+
+int g_debuglevel = DEBUGLEVEL;

external/zstd/common/debug.h

@@ -0,0 +1,107 @@
+/* ******************************************************************
+ * debug
+ * Part of FSE library
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ *
+ * You can contact the author at :
+ * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+****************************************************************** */
+
+
+/*
+ * The purpose of this header is to enable debug functions.
+ * They regroup assert(), DEBUGLOG() and RAWLOG() for run-time,
+ * and DEBUG_STATIC_ASSERT() for compile-time.
+ *
+ * By default, DEBUGLEVEL==0, which means run-time debug is disabled.
+ *
+ * Level 1 enables assert() only.
+ * Starting level 2, traces can be generated and pushed to stderr.
+ * The higher the level, the more verbose the traces.
+ *
+ * It's possible to dynamically adjust level using variable g_debug_level,
+ * which is only declared if DEBUGLEVEL>=2,
+ * and is a global variable, not multi-thread protected (use with care)
+ */
+
+#ifndef DEBUG_H_12987983217
+#define DEBUG_H_12987983217
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
+
+/* static assert is triggered at compile time, leaving no runtime artefact.
+ * static assert only works with compile-time constants.
+ * Also, this variant can only be used inside a function. */
+#define DEBUG_STATIC_ASSERT(c) (void)sizeof(char[(c) ? 1 : -1])
+
+
+/* DEBUGLEVEL is expected to be defined externally,
+ * typically through compiler command line.
+ * Value must be a number. */
+#ifndef DEBUGLEVEL
+#  define DEBUGLEVEL 0
+#endif
+
+
+/* recommended values for DEBUGLEVEL :
+ * 0 : release mode, no debug, all run-time checks disabled
+ * 1 : enables assert() only, no display
+ * 2 : reserved, for currently active debug path
+ * 3 : events once per object lifetime (CCtx, CDict, etc.)
+ * 4 : events once per frame
+ * 5 : events once per block
+ * 6 : events once per sequence (verbose)
+ * 7+: events at every position (*very* verbose)
+ *
+ * It's generally inconvenient to output traces > 5.
+ * In which case, it's possible to selectively trigger high verbosity levels
+ * by modifying g_debug_level.
+ */
+
+#if (DEBUGLEVEL>=1)
+#  define ZSTD_DEPS_NEED_ASSERT
+#  include "zstd_deps.h"
+#else
+#  ifndef assert   /* assert may be already defined, due to prior #include <assert.h> */
+#    define assert(condition) ((void)0)   /* disable assert (default) */
+#  endif
+#endif
+
+#if (DEBUGLEVEL>=2)
+#  define ZSTD_DEPS_NEED_IO
+#  include "zstd_deps.h"
+extern int g_debuglevel; /* the variable is only declared,
+                            it actually lives in debug.c,
+                            and is shared by the whole process.
+                            It's not thread-safe.
+                            It's useful when enabling very verbose levels
+                            on selective conditions (such as position in src) */
+
+#  define RAWLOG(l, ...) {                                       \
+                if (l<=g_debuglevel) {                           \
+                    ZSTD_DEBUG_PRINT(__VA_ARGS__);               \
+            }   }
+#  define DEBUGLOG(l, ...) {                                     \
+                if (l<=g_debuglevel) {                           \
+                    ZSTD_DEBUG_PRINT(__FILE__ ": " __VA_ARGS__); \
+                    ZSTD_DEBUG_PRINT(" \n");                     \
+            }   }
+#else
+#  define RAWLOG(l, ...)      {}    /* disabled */
+#  define DEBUGLOG(l, ...)    {}    /* disabled */
+#endif
+
+
+#if defined (__cplusplus)
+}
+#endif
+
+#endif /* DEBUG_H_12987983217 */

external/zstd/common/entropy_common.c

@@ -0,0 +1,341 @@
+/* ******************************************************************
+ * Common functions of New Generation Entropy library
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ *
+ *  You can contact the author at :
+ *  - FSE+HUF source repository : https://github.com/Cyan4973/FiniteStateEntropy
+ *  - Public forum : https://groups.google.com/forum/#!forum/lz4c
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+****************************************************************** */
+
+/* *************************************
+*  Dependencies
+***************************************/
+#include "mem.h"
+#include "error_private.h"       /* ERR_*, ERROR */
+#define FSE_STATIC_LINKING_ONLY  /* FSE_MIN_TABLELOG */
+#include "fse.h"
+#define HUF_STATIC_LINKING_ONLY  /* HUF_TABLELOG_ABSOLUTEMAX */
+#include "huf.h"
+#include "bits.h"                /* ZSDT_highbit32, ZSTD_countTrailingZeros32 */
+
+
+/*===   Version   ===*/
+unsigned FSE_versionNumber(void) { return FSE_VERSION_NUMBER; }
+
+
+/*===   Error Management   ===*/
+unsigned FSE_isError(size_t code) { return ERR_isError(code); }
+const char* FSE_getErrorName(size_t code) { return ERR_getErrorName(code); }
+
+unsigned HUF_isError(size_t code) { return ERR_isError(code); }
+const char* HUF_getErrorName(size_t code) { return ERR_getErrorName(code); }
+
+
+/*-**************************************************************
+*  FSE NCount encoding-decoding
+****************************************************************/
+FORCE_INLINE_TEMPLATE
+size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
+                           const void* headerBuffer, size_t hbSize)
+{
+    const BYTE* const istart = (const BYTE*) headerBuffer;
+    const BYTE* const iend = istart + hbSize;
+    const BYTE* ip = istart;
+    int nbBits;
+    int remaining;
+    int threshold;
+    U32 bitStream;
+    int bitCount;
+    unsigned charnum = 0;
+    unsigned const maxSV1 = *maxSVPtr + 1;
+    int previous0 = 0;
+
+    if (hbSize < 8) {
+        /* This function only works when hbSize >= 8 */
+        char buffer[8] = {0};
+        ZSTD_memcpy(buffer, headerBuffer, hbSize);
+        {   size_t const countSize = FSE_readNCount(normalizedCounter, maxSVPtr, tableLogPtr,
+                                                    buffer, sizeof(buffer));
+            if (FSE_isError(countSize)) return countSize;
+            if (countSize > hbSize) return ERROR(corruption_detected);
+            return countSize;
+    }   }
+    assert(hbSize >= 8);
+
+    /* init */
+    ZSTD_memset(normalizedCounter, 0, (*maxSVPtr+1) * sizeof(normalizedCounter[0]));   /* all symbols not present in NCount have a frequency of 0 */
+    bitStream = MEM_readLE32(ip);
+    nbBits = (bitStream & 0xF) + FSE_MIN_TABLELOG;   /* extract tableLog */
+    if (nbBits > FSE_TABLELOG_ABSOLUTE_MAX) return ERROR(tableLog_tooLarge);
+    bitStream >>= 4;
+    bitCount = 4;
+    *tableLogPtr = nbBits;
+    remaining = (1<<nbBits)+1;
+    threshold = 1<<nbBits;
+    nbBits++;
+
+    for (;;) {
+        if (previous0) {
+            /* Count the number of repeats. Each time the
+             * 2-bit repeat code is 0b11 there is another
+             * repeat.
+             * Avoid UB by setting the high bit to 1.
+             */
+            int repeats = ZSTD_countTrailingZeros32(~bitStream | 0x80000000) >> 1;
+            while (repeats >= 12) {
+                charnum += 3 * 12;
+                if (LIKELY(ip <= iend-7)) {
+                    ip += 3;
+                } else {
+                    bitCount -= (int)(8 * (iend - 7 - ip));
+                    bitCount &= 31;
+                    ip = iend - 4;
+                }
+                bitStream = MEM_readLE32(ip) >> bitCount;
+                repeats = ZSTD_countTrailingZeros32(~bitStream | 0x80000000) >> 1;
+            }
+            charnum += 3 * repeats;
+            bitStream >>= 2 * repeats;
+            bitCount += 2 * repeats;
+
+            /* Add the final repeat which isn't 0b11. */
+            assert((bitStream & 3) < 3);
+            charnum += bitStream & 3;
+            bitCount += 2;
+
+            /* This is an error, but break and return an error
+             * at the end, because returning out of a loop makes
+             * it harder for the compiler to optimize.
+             */
+            if (charnum >= maxSV1) break;
+
+            /* We don't need to set the normalized count to 0
+             * because we already memset the whole buffer to 0.
+             */
+
+            if (LIKELY(ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) {
+                assert((bitCount >> 3) <= 3); /* For first condition to work */
+                ip += bitCount>>3;
+                bitCount &= 7;
+            } else {
+                bitCount -= (int)(8 * (iend - 4 - ip));
+                bitCount &= 31;
+                ip = iend - 4;
+            }
+            bitStream = MEM_readLE32(ip) >> bitCount;
+        }
+        {
+            int const max = (2*threshold-1) - remaining;
+            int count;
+
+            if ((bitStream & (threshold-1)) < (U32)max) {
+                count = bitStream & (threshold-1);
+                bitCount += nbBits-1;
+            } else {
+                count = bitStream & (2*threshold-1);
+                if (count >= threshold) count -= max;
+                bitCount += nbBits;
+            }
+
+            count--;   /* extra accuracy */
+            /* When it matters (small blocks), this is a
+             * predictable branch, because we don't use -1.
+             */
+            if (count >= 0) {
+                remaining -= count;
+            } else {
+                assert(count == -1);
+                remaining += count;
+            }
+            normalizedCounter[charnum++] = (short)count;
+            previous0 = !count;
+
+            assert(threshold > 1);
+            if (remaining < threshold) {
+                /* This branch can be folded into the
+                 * threshold update condition because we
+                 * know that threshold > 1.
+                 */
+                if (remaining <= 1) break;
+                nbBits = ZSTD_highbit32(remaining) + 1;
+                threshold = 1 << (nbBits - 1);
+            }
+            if (charnum >= maxSV1) break;
+
+            if (LIKELY(ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) {
+                ip += bitCount>>3;
+                bitCount &= 7;
+            } else {
+                bitCount -= (int)(8 * (iend - 4 - ip));
+                bitCount &= 31;
+                ip = iend - 4;
+            }
+            bitStream = MEM_readLE32(ip) >> bitCount;
+    }   }
+    if (remaining != 1) return ERROR(corruption_detected);
+    /* Only possible when there are too many zeros. */
+    if (charnum > maxSV1) return ERROR(maxSymbolValue_tooSmall);
+    if (bitCount > 32) return ERROR(corruption_detected);
+    *maxSVPtr = charnum-1;
+
+    ip += (bitCount+7)>>3;
+    return ip-istart;
+}
+
+/* Avoids the FORCE_INLINE of the _body() function. */
+static size_t FSE_readNCount_body_default(
+        short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
+        const void* headerBuffer, size_t hbSize)
+{
+    return FSE_readNCount_body(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
+}
+
+#if DYNAMIC_BMI2
+BMI2_TARGET_ATTRIBUTE static size_t FSE_readNCount_body_bmi2(
+        short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
+        const void* headerBuffer, size_t hbSize)
+{
+    return FSE_readNCount_body(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
+}
+#endif
+
+size_t FSE_readNCount_bmi2(
+        short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
+        const void* headerBuffer, size_t hbSize, int bmi2)
+{
+#if DYNAMIC_BMI2
+    if (bmi2) {
+        return FSE_readNCount_body_bmi2(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
+    }
+#endif
+    (void)bmi2;
+    return FSE_readNCount_body_default(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
+}
+
+size_t FSE_readNCount(
+        short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
+        const void* headerBuffer, size_t hbSize)
+{
+    return FSE_readNCount_bmi2(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize, /* bmi2 */ 0);
+}
+
+
+/*! HUF_readStats() :
+    Read compact Huffman tree, saved by HUF_writeCTable().
+    `huffWeight` is destination buffer.
+    `rankStats` is assumed to be a table of at least HUF_TABLELOG_MAX U32.
+    @return : size read from `src` , or an error Code .
+    Note : Needed by HUF_readCTable() and HUF_readDTableX?() .
+*/
+size_t HUF_readStats(BYTE* huffWeight, size_t hwSize, U32* rankStats,
+                     U32* nbSymbolsPtr, U32* tableLogPtr,
+                     const void* src, size_t srcSize)
+{
+    U32 wksp[HUF_READ_STATS_WORKSPACE_SIZE_U32];
+    return HUF_readStats_wksp(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, wksp, sizeof(wksp), /* bmi2 */ 0);
+}
+
+FORCE_INLINE_TEMPLATE size_t
+HUF_readStats_body(BYTE* huffWeight, size_t hwSize, U32* rankStats,
+                   U32* nbSymbolsPtr, U32* tableLogPtr,
+                   const void* src, size_t srcSize,
+                   void* workSpace, size_t wkspSize,
+                   int bmi2)
+{
+    U32 weightTotal;
+    const BYTE* ip = (const BYTE*) src;
+    size_t iSize;
+    size_t oSize;
+
+    if (!srcSize) return ERROR(srcSize_wrong);
+    iSize = ip[0];
+    /* ZSTD_memset(huffWeight, 0, hwSize);   *//* is not necessary, even though some analyzer complain ... */
+
+    if (iSize >= 128) {  /* special header */
+        oSize = iSize - 127;
+        iSize = ((oSize+1)/2);
+        if (iSize+1 > srcSize) return ERROR(srcSize_wrong);
+        if (oSize >= hwSize) return ERROR(corruption_detected);
+        ip += 1;
+        {   U32 n;
+            for (n=0; n<oSize; n+=2) {
+                huffWeight[n]   = ip[n/2] >> 4;
+                huffWeight[n+1] = ip[n/2] & 15;
+    }   }   }
+    else  {   /* header compressed with FSE (normal case) */
+        if (iSize+1 > srcSize) return ERROR(srcSize_wrong);
+        /* max (hwSize-1) values decoded, as last one is implied */
+        oSize = FSE_decompress_wksp_bmi2(huffWeight, hwSize-1, ip+1, iSize, 6, workSpace, wkspSize, bmi2);
+        if (FSE_isError(oSize)) return oSize;
+    }
+
+    /* collect weight stats */
+    ZSTD_memset(rankStats, 0, (HUF_TABLELOG_MAX + 1) * sizeof(U32));
+    weightTotal = 0;
+    {   U32 n; for (n=0; n<oSize; n++) {
+            if (huffWeight[n] > HUF_TABLELOG_MAX) return ERROR(corruption_detected);
+            rankStats[huffWeight[n]]++;
+            weightTotal += (1 << huffWeight[n]) >> 1;
+    }   }
+    if (weightTotal == 0) return ERROR(corruption_detected);
+
+    /* get last non-null symbol weight (implied, total must be 2^n) */
+    {   U32 const tableLog = ZSTD_highbit32(weightTotal) + 1;
+        if (tableLog > HUF_TABLELOG_MAX) return ERROR(corruption_detected);
+        *tableLogPtr = tableLog;
+        /* determine last weight */
+        {   U32 const total = 1 << tableLog;
+            U32 const rest = total - weightTotal;
+            U32 const verif = 1 << ZSTD_highbit32(rest);
+            U32 const lastWeight = ZSTD_highbit32(rest) + 1;
+            if (verif != rest) return ERROR(corruption_detected);    /* last value must be a clean power of 2 */
+            huffWeight[oSize] = (BYTE)lastWeight;
+            rankStats[lastWeight]++;
+    }   }
+
+    /* check tree construction validity */
+    if ((rankStats[1] < 2) || (rankStats[1] & 1)) return ERROR(corruption_detected);   /* by construction : at least 2 elts of rank 1, must be even */
+
+    /* results */
+    *nbSymbolsPtr = (U32)(oSize+1);
+    return iSize+1;
+}
+
+/* Avoids the FORCE_INLINE of the _body() function. */
+static size_t HUF_readStats_body_default(BYTE* huffWeight, size_t hwSize, U32* rankStats,
+                     U32* nbSymbolsPtr, U32* tableLogPtr,
+                     const void* src, size_t srcSize,
+                     void* workSpace, size_t wkspSize)
+{
+    return HUF_readStats_body(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize, 0);
+}
+
+#if DYNAMIC_BMI2
+static BMI2_TARGET_ATTRIBUTE size_t HUF_readStats_body_bmi2(BYTE* huffWeight, size_t hwSize, U32* rankStats,
+                     U32* nbSymbolsPtr, U32* tableLogPtr,
+                     const void* src, size_t srcSize,
+                     void* workSpace, size_t wkspSize)
+{
+    return HUF_readStats_body(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize, 1);
+}
+#endif
+
+size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize, U32* rankStats,
+                     U32* nbSymbolsPtr, U32* tableLogPtr,
+                     const void* src, size_t srcSize,
+                     void* workSpace, size_t wkspSize,
+                     int bmi2)
+{
+#if DYNAMIC_BMI2
+    if (bmi2) {
+        return HUF_readStats_body_bmi2(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize);
+    }
+#endif
+    (void)bmi2;
+    return HUF_readStats_body_default(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize);
+}

external/zstd/common/error_private.c

@@ -0,0 +1,60 @@
+/*
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * All rights reserved.
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+ */
+
+/* The purpose of this file is to have a single list of error strings embedded in binary */
+
+#include "error_private.h"
+
+const char* ERR_getErrorString(ERR_enum code)
+{
+#ifdef ZSTD_STRIP_ERROR_STRINGS
+    (void)code;
+    return "Error strings stripped";
+#else
+    static const char* const notErrorCode = "Unspecified error code";
+    switch( code )
+    {
+    case PREFIX(no_error): return "No error detected";
+    case PREFIX(GENERIC):  return "Error (generic)";
+    case PREFIX(prefix_unknown): return "Unknown frame descriptor";
+    case PREFIX(version_unsupported): return "Version not supported";
+    case PREFIX(frameParameter_unsupported): return "Unsupported frame parameter";
+    case PREFIX(frameParameter_windowTooLarge): return "Frame requires too much memory for decoding";
+    case PREFIX(corruption_detected): return "Data corruption detected";
+    case PREFIX(checksum_wrong): return "Restored data doesn't match checksum";
+    case PREFIX(literals_headerWrong): return "Header of Literals' block doesn't respect format specification";
+    case PREFIX(parameter_unsupported): return "Unsupported parameter";
+    case PREFIX(parameter_combination_unsupported): return "Unsupported combination of parameters";
+    case PREFIX(parameter_outOfBound): return "Parameter is out of bound";
+    case PREFIX(init_missing): return "Context should be init first";
+    case PREFIX(memory_allocation): return "Allocation error : not enough memory";
+    case PREFIX(workSpace_tooSmall): return "workSpace buffer is not large enough";
+    case PREFIX(stage_wrong): return "Operation not authorized at current processing stage";
+    case PREFIX(tableLog_tooLarge): return "tableLog requires too much memory : unsupported";
+    case PREFIX(maxSymbolValue_tooLarge): return "Unsupported max Symbol Value : too large";
+    case PREFIX(maxSymbolValue_tooSmall): return "Specified maxSymbolValue is too small";
+    case PREFIX(stabilityCondition_notRespected): return "pledged buffer stability condition is not respected";
+    case PREFIX(dictionary_corrupted): return "Dictionary is corrupted";
+    case PREFIX(dictionary_wrong): return "Dictionary mismatch";
+    case PREFIX(dictionaryCreation_failed): return "Cannot create Dictionary from provided samples";
+    case PREFIX(dstSize_tooSmall): return "Destination buffer is too small";
+    case PREFIX(srcSize_wrong): return "Src size is incorrect";
+    case PREFIX(dstBuffer_null): return "Operation on NULL destination buffer";
+        /* following error codes are not stable and may be removed or changed in a future version */
+    case PREFIX(frameIndex_tooLarge): return "Frame index is too large";
+    case PREFIX(seekableIO): return "An I/O error occurred when reading/seeking";
+    case PREFIX(dstBuffer_wrong): return "Destination buffer is wrong";
+    case PREFIX(srcBuffer_wrong): return "Source buffer is wrong";
+    case PREFIX(externalMatchFinder_failed): return "External matchfinder returned an error code";
+    case PREFIX(maxCode):
+    default: return notErrorCode;
+    }
+#endif
+}

external/zstd/common/error_private.h

@@ -0,0 +1,159 @@
+/*
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * All rights reserved.
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+ */
+
+/* Note : this module is expected to remain private, do not expose it */
+
+#ifndef ERROR_H_MODULE
+#define ERROR_H_MODULE
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
+
+/* ****************************************
+*  Dependencies
+******************************************/
+#include "../zstd_errors.h"  /* enum list */
+#include "compiler.h"
+#include "debug.h"
+#include "zstd_deps.h"       /* size_t */
+
+
+/* ****************************************
+*  Compiler-specific
+******************************************/
+#if defined(__GNUC__)
+#  define ERR_STATIC static __attribute__((unused))
+#elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
+#  define ERR_STATIC static inline
+#elif defined(_MSC_VER)
+#  define ERR_STATIC static __inline
+#else
+#  define ERR_STATIC static  /* this version may generate warnings for unused static functions; disable the relevant warning */
+#endif
+
+
+/*-****************************************
+*  Customization (error_public.h)
+******************************************/
+typedef ZSTD_ErrorCode ERR_enum;
+#define PREFIX(name) ZSTD_error_##name
+
+
+/*-****************************************
+*  Error codes handling
+******************************************/
+#undef ERROR   /* already defined on Visual Studio */
+#define ERROR(name) ZSTD_ERROR(name)
+#define ZSTD_ERROR(name) ((size_t)-PREFIX(name))
+
+ERR_STATIC unsigned ERR_isError(size_t code) { return (code > ERROR(maxCode)); }
+
+ERR_STATIC ERR_enum ERR_getErrorCode(size_t code) { if (!ERR_isError(code)) return (ERR_enum)0; return (ERR_enum) (0-code); }
+
+/* check and forward error code */
+#define CHECK_V_F(e, f) size_t const e = f; if (ERR_isError(e)) return e
+#define CHECK_F(f)   { CHECK_V_F(_var_err__, f); }
+
+
+/*-****************************************
+*  Error Strings
+******************************************/
+
+const char* ERR_getErrorString(ERR_enum code);   /* error_private.c */
+
+ERR_STATIC const char* ERR_getErrorName(size_t code)
+{
+    return ERR_getErrorString(ERR_getErrorCode(code));
+}
+
+/**
+ * Ignore: this is an internal helper.
+ *
+ * This is a helper function to help force C99-correctness during compilation.
+ * Under strict compilation modes, variadic macro arguments can't be empty.
+ * However, variadic function arguments can be. Using a function therefore lets
+ * us statically check that at least one (string) argument was passed,
+ * independent of the compilation flags.
+ */
+static INLINE_KEYWORD UNUSED_ATTR
+void _force_has_format_string(const char *format, ...) {
+  (void)format;
+}
+
+/**
+ * Ignore: this is an internal helper.
+ *
+ * We want to force this function invocation to be syntactically correct, but
+ * we don't want to force runtime evaluation of its arguments.
+ */
+#define _FORCE_HAS_FORMAT_STRING(...) \
+  if (0) { \
+    _force_has_format_string(__VA_ARGS__); \
+  }
+
+#define ERR_QUOTE(str) #str
+
+/**
+ * Return the specified error if the condition evaluates to true.
+ *
+ * In debug modes, prints additional information.
+ * In order to do that (particularly, printing the conditional that failed),
+ * this can't just wrap RETURN_ERROR().
+ */
+#define RETURN_ERROR_IF(cond, err, ...) \
+  if (cond) { \
+    RAWLOG(3, "%s:%d: ERROR!: check %s failed, returning %s", \
+           __FILE__, __LINE__, ERR_QUOTE(cond), ERR_QUOTE(ERROR(err))); \
+    _FORCE_HAS_FORMAT_STRING(__VA_ARGS__); \
+    RAWLOG(3, ": " __VA_ARGS__); \
+    RAWLOG(3, "\n"); \
+    return ERROR(err); \
+  }
+
+/**
+ * Unconditionally return the specified error.
+ *
+ * In debug modes, prints additional information.
+ */
+#define RETURN_ERROR(err, ...) \
+  do { \
+    RAWLOG(3, "%s:%d: ERROR!: unconditional check failed, returning %s", \
+           __FILE__, __LINE__, ERR_QUOTE(ERROR(err))); \
+    _FORCE_HAS_FORMAT_STRING(__VA_ARGS__); \
+    RAWLOG(3, ": " __VA_ARGS__); \
+    RAWLOG(3, "\n"); \
+    return ERROR(err); \
+  } while(0);
+
+/**
+ * If the provided expression evaluates to an error code, returns that error code.
+ *
+ * In debug modes, prints additional information.
+ */
+#define FORWARD_IF_ERROR(err, ...) \
+  do { \
+    size_t const err_code = (err); \
+    if (ERR_isError(err_code)) { \
+      RAWLOG(3, "%s:%d: ERROR!: forwarding error in %s: %s", \
+             __FILE__, __LINE__, ERR_QUOTE(err), ERR_getErrorName(err_code)); \
+      _FORCE_HAS_FORMAT_STRING(__VA_ARGS__); \
+      RAWLOG(3, ": " __VA_ARGS__); \
+      RAWLOG(3, "\n"); \
+      return err_code; \
+    } \
+  } while(0);
+
+#if defined (__cplusplus)
+}
+#endif
+
+#endif /* ERROR_H_MODULE */

external/zstd/common/fse.h

@@ -0,0 +1,717 @@
+/* ******************************************************************
+ * FSE : Finite State Entropy codec
+ * Public Prototypes declaration
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ *
+ * You can contact the author at :
+ * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+****************************************************************** */
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
+#ifndef FSE_H
+#define FSE_H
+
+
+/*-*****************************************
+*  Dependencies
+******************************************/
+#include "zstd_deps.h"    /* size_t, ptrdiff_t */
+
+
+/*-*****************************************
+*  FSE_PUBLIC_API : control library symbols visibility
+******************************************/
+#if defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) && defined(__GNUC__) && (__GNUC__ >= 4)
+#  define FSE_PUBLIC_API __attribute__ ((visibility ("default")))
+#elif defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1)   /* Visual expected */
+#  define FSE_PUBLIC_API __declspec(dllexport)
+#elif defined(FSE_DLL_IMPORT) && (FSE_DLL_IMPORT==1)
+#  define FSE_PUBLIC_API __declspec(dllimport) /* It isn't required but allows to generate better code, saving a function pointer load from the IAT and an indirect jump.*/
+#else
+#  define FSE_PUBLIC_API
+#endif
+
+/*------   Version   ------*/
+#define FSE_VERSION_MAJOR    0
+#define FSE_VERSION_MINOR    9
+#define FSE_VERSION_RELEASE  0
+
+#define FSE_LIB_VERSION FSE_VERSION_MAJOR.FSE_VERSION_MINOR.FSE_VERSION_RELEASE
+#define FSE_QUOTE(str) #str
+#define FSE_EXPAND_AND_QUOTE(str) FSE_QUOTE(str)
+#define FSE_VERSION_STRING FSE_EXPAND_AND_QUOTE(FSE_LIB_VERSION)
+
+#define FSE_VERSION_NUMBER  (FSE_VERSION_MAJOR *100*100 + FSE_VERSION_MINOR *100 + FSE_VERSION_RELEASE)
+FSE_PUBLIC_API unsigned FSE_versionNumber(void);   /**< library version number; to be used when checking dll version */
+
+
+/*-****************************************
+*  FSE simple functions
+******************************************/
+/*! FSE_compress() :
+    Compress content of buffer 'src', of size 'srcSize', into destination buffer 'dst'.
+    'dst' buffer must be already allocated. Compression runs faster is dstCapacity >= FSE_compressBound(srcSize).
+    @return : size of compressed data (<= dstCapacity).
+    Special values : if return == 0, srcData is not compressible => Nothing is stored within dst !!!
+                     if return == 1, srcData is a single byte symbol * srcSize times. Use RLE compression instead.
+                     if FSE_isError(return), compression failed (more details using FSE_getErrorName())
+*/
+FSE_PUBLIC_API size_t FSE_compress(void* dst, size_t dstCapacity,
+                             const void* src, size_t srcSize);
+
+/*! FSE_decompress():
+    Decompress FSE data from buffer 'cSrc', of size 'cSrcSize',
+    into already allocated destination buffer 'dst', of size 'dstCapacity'.
+    @return : size of regenerated data (<= maxDstSize),
+              or an error code, which can be tested using FSE_isError() .
+
+    ** Important ** : FSE_decompress() does not decompress non-compressible nor RLE data !!!
+    Why ? : making this distinction requires a header.
+    Header management is intentionally delegated to the user layer, which can better manage special cases.
+*/
+FSE_PUBLIC_API size_t FSE_decompress(void* dst,  size_t dstCapacity,
+                               const void* cSrc, size_t cSrcSize);
+
+
+/*-*****************************************
+*  Tool functions
+******************************************/
+FSE_PUBLIC_API size_t FSE_compressBound(size_t size);       /* maximum compressed size */
+
+/* Error Management */
+FSE_PUBLIC_API unsigned    FSE_isError(size_t code);        /* tells if a return value is an error code */
+FSE_PUBLIC_API const char* FSE_getErrorName(size_t code);   /* provides error code string (useful for debugging) */
+
+
+/*-*****************************************
+*  FSE advanced functions
+******************************************/
+/*! FSE_compress2() :
+    Same as FSE_compress(), but allows the selection of 'maxSymbolValue' and 'tableLog'
+    Both parameters can be defined as '0' to mean : use default value
+    @return : size of compressed data
+    Special values : if return == 0, srcData is not compressible => Nothing is stored within cSrc !!!
+                     if return == 1, srcData is a single byte symbol * srcSize times. Use RLE compression.
+                     if FSE_isError(return), it's an error code.
+*/
+FSE_PUBLIC_API size_t FSE_compress2 (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog);
+
+
+/*-*****************************************
+*  FSE detailed API
+******************************************/
+/*!
+FSE_compress() does the following:
+1. count symbol occurrence from source[] into table count[] (see hist.h)
+2. normalize counters so that sum(count[]) == Power_of_2 (2^tableLog)
+3. save normalized counters to memory buffer using writeNCount()
+4. build encoding table 'CTable' from normalized counters
+5. encode the data stream using encoding table 'CTable'
+
+FSE_decompress() does the following:
+1. read normalized counters with readNCount()
+2. build decoding table 'DTable' from normalized counters
+3. decode the data stream using decoding table 'DTable'
+
+The following API allows targeting specific sub-functions for advanced tasks.
+For example, it's possible to compress several blocks using the same 'CTable',
+or to save and provide normalized distribution using external method.
+*/
+
+/* *** COMPRESSION *** */
+
+/*! FSE_optimalTableLog():
+    dynamically downsize 'tableLog' when conditions are met.
+    It saves CPU time, by using smaller tables, while preserving or even improving compression ratio.
+    @return : recommended tableLog (necessarily <= 'maxTableLog') */
+FSE_PUBLIC_API unsigned FSE_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue);
+
+/*! FSE_normalizeCount():
+    normalize counts so that sum(count[]) == Power_of_2 (2^tableLog)
+    'normalizedCounter' is a table of short, of minimum size (maxSymbolValue+1).
+    useLowProbCount is a boolean parameter which trades off compressed size for
+    faster header decoding. When it is set to 1, the compressed data will be slightly
+    smaller. And when it is set to 0, FSE_readNCount() and FSE_buildDTable() will be
+    faster. If you are compressing a small amount of data (< 2 KB) then useLowProbCount=0
+    is a good default, since header deserialization makes a big speed difference.
+    Otherwise, useLowProbCount=1 is a good default, since the speed difference is small.
+    @return : tableLog,
+              or an errorCode, which can be tested using FSE_isError() */
+FSE_PUBLIC_API size_t FSE_normalizeCount(short* normalizedCounter, unsigned tableLog,
+                    const unsigned* count, size_t srcSize, unsigned maxSymbolValue, unsigned useLowProbCount);
+
+/*! FSE_NCountWriteBound():
+    Provides the maximum possible size of an FSE normalized table, given 'maxSymbolValue' and 'tableLog'.
+    Typically useful for allocation purpose. */
+FSE_PUBLIC_API size_t FSE_NCountWriteBound(unsigned maxSymbolValue, unsigned tableLog);
+
+/*! FSE_writeNCount():
+    Compactly save 'normalizedCounter' into 'buffer'.
+    @return : size of the compressed table,
+              or an errorCode, which can be tested using FSE_isError(). */
+FSE_PUBLIC_API size_t FSE_writeNCount (void* buffer, size_t bufferSize,
+                                 const short* normalizedCounter,
+                                 unsigned maxSymbolValue, unsigned tableLog);
+
+/*! Constructor and Destructor of FSE_CTable.
+    Note that FSE_CTable size depends on 'tableLog' and 'maxSymbolValue' */
+typedef unsigned FSE_CTable;   /* don't allocate that. It's only meant to be more restrictive than void* */
+FSE_PUBLIC_API FSE_CTable* FSE_createCTable (unsigned maxSymbolValue, unsigned tableLog);
+FSE_PUBLIC_API void        FSE_freeCTable (FSE_CTable* ct);
+
+/*! FSE_buildCTable():
+    Builds `ct`, which must be already allocated, using FSE_createCTable().
+    @return : 0, or an errorCode, which can be tested using FSE_isError() */
+FSE_PUBLIC_API size_t FSE_buildCTable(FSE_CTable* ct, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog);
+
+/*! FSE_compress_usingCTable():
+    Compress `src` using `ct` into `dst` which must be already allocated.
+    @return : size of compressed data (<= `dstCapacity`),
+              or 0 if compressed data could not fit into `dst`,
+              or an errorCode, which can be tested using FSE_isError() */
+FSE_PUBLIC_API size_t FSE_compress_usingCTable (void* dst, size_t dstCapacity, const void* src, size_t srcSize, const FSE_CTable* ct);
+
+/*!
+Tutorial :
+----------
+The first step is to count all symbols. FSE_count() does this job very fast.
+Result will be saved into 'count', a table of unsigned int, which must be already allocated, and have 'maxSymbolValuePtr[0]+1' cells.
+'src' is a table of bytes of size 'srcSize'. All values within 'src' MUST be <= maxSymbolValuePtr[0]
+maxSymbolValuePtr[0] will be updated, with its real value (necessarily <= original value)
+FSE_count() will return the number of occurrence of the most frequent symbol.
+This can be used to know if there is a single symbol within 'src', and to quickly evaluate its compressibility.
+If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError()).
+
+The next step is to normalize the frequencies.
+FSE_normalizeCount() will ensure that sum of frequencies is == 2 ^'tableLog'.
+It also guarantees a minimum of 1 to any Symbol with frequency >= 1.
+You can use 'tableLog'==0 to mean "use default tableLog value".
+If you are unsure of which tableLog value to use, you can ask FSE_optimalTableLog(),
+which will provide the optimal valid tableLog given sourceSize, maxSymbolValue, and a user-defined maximum (0 means "default").
+
+The result of FSE_normalizeCount() will be saved into a table,
+called 'normalizedCounter', which is a table of signed short.
+'normalizedCounter' must be already allocated, and have at least 'maxSymbolValue+1' cells.
+The return value is tableLog if everything proceeded as expected.
+It is 0 if there is a single symbol within distribution.
+If there is an error (ex: invalid tableLog value), the function will return an ErrorCode (which can be tested using FSE_isError()).
+
+'normalizedCounter' can be saved in a compact manner to a memory area using FSE_writeNCount().
+'buffer' must be already allocated.
+For guaranteed success, buffer size must be at least FSE_headerBound().
+The result of the function is the number of bytes written into 'buffer'.
+If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError(); ex : buffer size too small).
+
+'normalizedCounter' can then be used to create the compression table 'CTable'.
+The space required by 'CTable' must be already allocated, using FSE_createCTable().
+You can then use FSE_buildCTable() to fill 'CTable'.
+If there is an error, both functions will return an ErrorCode (which can be tested using FSE_isError()).
+
+'CTable' can then be used to compress 'src', with FSE_compress_usingCTable().
+Similar to FSE_count(), the convention is that 'src' is assumed to be a table of char of size 'srcSize'
+The function returns the size of compressed data (without header), necessarily <= `dstCapacity`.
+If it returns '0', compressed data could not fit into 'dst'.
+If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError()).
+*/
+
+
+/* *** DECOMPRESSION *** */
+
+/*! FSE_readNCount():
+    Read compactly saved 'normalizedCounter' from 'rBuffer'.
+    @return : size read from 'rBuffer',
+              or an errorCode, which can be tested using FSE_isError().
+              maxSymbolValuePtr[0] and tableLogPtr[0] will also be updated with their respective values */
+FSE_PUBLIC_API size_t FSE_readNCount (short* normalizedCounter,
+                           unsigned* maxSymbolValuePtr, unsigned* tableLogPtr,
+                           const void* rBuffer, size_t rBuffSize);
+
+/*! FSE_readNCount_bmi2():
+ * Same as FSE_readNCount() but pass bmi2=1 when your CPU supports BMI2 and 0 otherwise.
+ */
+FSE_PUBLIC_API size_t FSE_readNCount_bmi2(short* normalizedCounter,
+                           unsigned* maxSymbolValuePtr, unsigned* tableLogPtr,
+                           const void* rBuffer, size_t rBuffSize, int bmi2);
+
+/*! Constructor and Destructor of FSE_DTable.
+    Note that its size depends on 'tableLog' */
+typedef unsigned FSE_DTable;   /* don't allocate that. It's just a way to be more restrictive than void* */
+FSE_PUBLIC_API FSE_DTable* FSE_createDTable(unsigned tableLog);
+FSE_PUBLIC_API void        FSE_freeDTable(FSE_DTable* dt);
+
+/*! FSE_buildDTable():
+    Builds 'dt', which must be already allocated, using FSE_createDTable().
+    return : 0, or an errorCode, which can be tested using FSE_isError() */
+FSE_PUBLIC_API size_t FSE_buildDTable (FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog);
+
+/*! FSE_decompress_usingDTable():
+    Decompress compressed source `cSrc` of size `cSrcSize` using `dt`
+    into `dst` which must be already allocated.
+    @return : size of regenerated data (necessarily <= `dstCapacity`),
+              or an errorCode, which can be tested using FSE_isError() */
+FSE_PUBLIC_API size_t FSE_decompress_usingDTable(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, const FSE_DTable* dt);
+
+/*!
+Tutorial :
+----------
+(Note : these functions only decompress FSE-compressed blocks.
+ If block is uncompressed, use memcpy() instead
+ If block is a single repeated byte, use memset() instead )
+
+The first step is to obtain the normalized frequencies of symbols.
+This can be performed by FSE_readNCount() if it was saved using FSE_writeNCount().
+'normalizedCounter' must be already allocated, and have at least 'maxSymbolValuePtr[0]+1' cells of signed short.
+In practice, that means it's necessary to know 'maxSymbolValue' beforehand,
+or size the table to handle worst case situations (typically 256).
+FSE_readNCount() will provide 'tableLog' and 'maxSymbolValue'.
+The result of FSE_readNCount() is the number of bytes read from 'rBuffer'.
+Note that 'rBufferSize' must be at least 4 bytes, even if useful information is less than that.
+If there is an error, the function will return an error code, which can be tested using FSE_isError().
+
+The next step is to build the decompression tables 'FSE_DTable' from 'normalizedCounter'.
+This is performed by the function FSE_buildDTable().
+The space required by 'FSE_DTable' must be already allocated using FSE_createDTable().
+If there is an error, the function will return an error code, which can be tested using FSE_isError().
+
+`FSE_DTable` can then be used to decompress `cSrc`, with FSE_decompress_usingDTable().
+`cSrcSize` must be strictly correct, otherwise decompression will fail.
+FSE_decompress_usingDTable() result will tell how many bytes were regenerated (<=`dstCapacity`).
+If there is an error, the function will return an error code, which can be tested using FSE_isError(). (ex: dst buffer too small)
+*/
+
+#endif  /* FSE_H */
+
+#if defined(FSE_STATIC_LINKING_ONLY) && !defined(FSE_H_FSE_STATIC_LINKING_ONLY)
+#define FSE_H_FSE_STATIC_LINKING_ONLY
+
+/* *** Dependency *** */
+#include "bitstream.h"
+
+
+/* *****************************************
+*  Static allocation
+*******************************************/
+/* FSE buffer bounds */
+#define FSE_NCOUNTBOUND 512
+#define FSE_BLOCKBOUND(size) ((size) + ((size)>>7) + 4 /* fse states */ + sizeof(size_t) /* bitContainer */)
+#define FSE_COMPRESSBOUND(size) (FSE_NCOUNTBOUND + FSE_BLOCKBOUND(size))   /* Macro version, useful for static allocation */
+
+/* It is possible to statically allocate FSE CTable/DTable as a table of FSE_CTable/FSE_DTable using below macros */
+#define FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue)   (1 + (1<<((maxTableLog)-1)) + (((maxSymbolValue)+1)*2))
+#define FSE_DTABLE_SIZE_U32(maxTableLog)                   (1 + (1<<(maxTableLog)))
+
+/* or use the size to malloc() space directly. Pay attention to alignment restrictions though */
+#define FSE_CTABLE_SIZE(maxTableLog, maxSymbolValue)   (FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(FSE_CTable))
+#define FSE_DTABLE_SIZE(maxTableLog)                   (FSE_DTABLE_SIZE_U32(maxTableLog) * sizeof(FSE_DTable))
+
+
+/* *****************************************
+ *  FSE advanced API
+ ***************************************** */
+
+unsigned FSE_optimalTableLog_internal(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, unsigned minus);
+/**< same as FSE_optimalTableLog(), which used `minus==2` */
+
+/* FSE_compress_wksp() :
+ * Same as FSE_compress2(), but using an externally allocated scratch buffer (`workSpace`).
+ * FSE_COMPRESS_WKSP_SIZE_U32() provides the minimum size required for `workSpace` as a table of FSE_CTable.
+ */
+#define FSE_COMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue)   ( FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) + ((maxTableLog > 12) ? (1 << (maxTableLog - 2)) : 1024) )
+size_t FSE_compress_wksp (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize);
+
+size_t FSE_buildCTable_raw (FSE_CTable* ct, unsigned nbBits);
+/**< build a fake FSE_CTable, designed for a flat distribution, where each symbol uses nbBits */
+
+size_t FSE_buildCTable_rle (FSE_CTable* ct, unsigned char symbolValue);
+/**< build a fake FSE_CTable, designed to compress always the same symbolValue */
+
+/* FSE_buildCTable_wksp() :
+ * Same as FSE_buildCTable(), but using an externally allocated scratch buffer (`workSpace`).
+ * `wkspSize` must be >= `FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog)` of `unsigned`.
+ * See FSE_buildCTable_wksp() for breakdown of workspace usage.
+ */
+#define FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog) (((maxSymbolValue + 2) + (1ull << (tableLog)))/2 + sizeof(U64)/sizeof(U32) /* additional 8 bytes for potential table overwrite */)
+#define FSE_BUILD_CTABLE_WORKSPACE_SIZE(maxSymbolValue, tableLog) (sizeof(unsigned) * FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog))
+size_t FSE_buildCTable_wksp(FSE_CTable* ct, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize);
+
+#define FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) (sizeof(short) * (maxSymbolValue + 1) + (1ULL << maxTableLog) + 8)
+#define FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) ((FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) + sizeof(unsigned) - 1) / sizeof(unsigned))
+FSE_PUBLIC_API size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize);
+/**< Same as FSE_buildDTable(), using an externally allocated `workspace` produced with `FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxSymbolValue)` */
+
+size_t FSE_buildDTable_raw (FSE_DTable* dt, unsigned nbBits);
+/**< build a fake FSE_DTable, designed to read a flat distribution where each symbol uses nbBits */
+
+size_t FSE_buildDTable_rle (FSE_DTable* dt, unsigned char symbolValue);
+/**< build a fake FSE_DTable, designed to always generate the same symbolValue */
+
+#define FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) (FSE_DTABLE_SIZE_U32(maxTableLog) + 1 + FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) + (FSE_MAX_SYMBOL_VALUE + 1) / 2 + 1)
+#define FSE_DECOMPRESS_WKSP_SIZE(maxTableLog, maxSymbolValue) (FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(unsigned))
+size_t FSE_decompress_wksp(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize);
+/**< same as FSE_decompress(), using an externally allocated `workSpace` produced with `FSE_DECOMPRESS_WKSP_SIZE_U32(maxLog, maxSymbolValue)` */
+
+size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2);
+/**< Same as FSE_decompress_wksp() but with dynamic BMI2 support. Pass 1 if your CPU supports BMI2 or 0 if it doesn't. */
+
+typedef enum {
+   FSE_repeat_none,  /**< Cannot use the previous table */
+   FSE_repeat_check, /**< Can use the previous table but it must be checked */
+   FSE_repeat_valid  /**< Can use the previous table and it is assumed to be valid */
+ } FSE_repeat;
+
+/* *****************************************
+*  FSE symbol compression API
+*******************************************/
+/*!
+   This API consists of small unitary functions, which highly benefit from being inlined.
+   Hence their body are included in next section.
+*/
+typedef struct {
+    ptrdiff_t   value;
+    const void* stateTable;
+    const void* symbolTT;
+    unsigned    stateLog;
+} FSE_CState_t;
+
+static void FSE_initCState(FSE_CState_t* CStatePtr, const FSE_CTable* ct);
+
+static void FSE_encodeSymbol(BIT_CStream_t* bitC, FSE_CState_t* CStatePtr, unsigned symbol);
+
+static void FSE_flushCState(BIT_CStream_t* bitC, const FSE_CState_t* CStatePtr);
+
+/**<
+These functions are inner components of FSE_compress_usingCTable().
+They allow the creation of custom streams, mixing multiple tables and bit sources.
+
+A key property to keep in mind is that encoding and decoding are done **in reverse direction**.
+So the first symbol you will encode is the last you will decode, like a LIFO stack.
+
+You will need a few variables to track your CStream. They are :
+
+FSE_CTable    ct;         // Provided by FSE_buildCTable()
+BIT_CStream_t bitStream;  // bitStream tracking structure
+FSE_CState_t  state;      // State tracking structure (can have several)
+
+
+The first thing to do is to init bitStream and state.
+    size_t errorCode = BIT_initCStream(&bitStream, dstBuffer, maxDstSize);
+    FSE_initCState(&state, ct);
+
+Note that BIT_initCStream() can produce an error code, so its result should be tested, using FSE_isError();
+You can then encode your input data, byte after byte.
+FSE_encodeSymbol() outputs a maximum of 'tableLog' bits at a time.
+Remember decoding will be done in reverse direction.
+    FSE_encodeByte(&bitStream, &state, symbol);
+
+At any time, you can also add any bit sequence.
+Note : maximum allowed nbBits is 25, for compatibility with 32-bits decoders
+    BIT_addBits(&bitStream, bitField, nbBits);
+
+The above methods don't commit data to memory, they just store it into local register, for speed.
+Local register size is 64-bits on 64-bits systems, 32-bits on 32-bits systems (size_t).
+Writing data to memory is a manual operation, performed by the flushBits function.
+    BIT_flushBits(&bitStream);
+
+Your last FSE encoding operation shall be to flush your last state value(s).
+    FSE_flushState(&bitStream, &state);
+
+Finally, you must close the bitStream.
+The function returns the size of CStream in bytes.
+If data couldn't fit into dstBuffer, it will return a 0 ( == not compressible)
+If there is an error, it returns an errorCode (which can be tested using FSE_isError()).
+    size_t size = BIT_closeCStream(&bitStream);
+*/
+
+
+/* *****************************************
+*  FSE symbol decompression API
+*******************************************/
+typedef struct {
+    size_t      state;
+    const void* table;   /* precise table may vary, depending on U16 */
+} FSE_DState_t;
+
+
+static void     FSE_initDState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD, const FSE_DTable* dt);
+
+static unsigned char FSE_decodeSymbol(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD);
+
+static unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr);
+
+/**<
+Let's now decompose FSE_decompress_usingDTable() into its unitary components.
+You will decode FSE-encoded symbols from the bitStream,
+and also any other bitFields you put in, **in reverse order**.
+
+You will need a few variables to track your bitStream. They are :
+
+BIT_DStream_t DStream;    // Stream context
+FSE_DState_t  DState;     // State context. Multiple ones are possible
+FSE_DTable*   DTablePtr;  // Decoding table, provided by FSE_buildDTable()
+
+The first thing to do is to init the bitStream.
+    errorCode = BIT_initDStream(&DStream, srcBuffer, srcSize);
+
+You should then retrieve your initial state(s)
+(in reverse flushing order if you have several ones) :
+    errorCode = FSE_initDState(&DState, &DStream, DTablePtr);
+
+You can then decode your data, symbol after symbol.
+For information the maximum number of bits read by FSE_decodeSymbol() is 'tableLog'.
+Keep in mind that symbols are decoded in reverse order, like a LIFO stack (last in, first out).
+    unsigned char symbol = FSE_decodeSymbol(&DState, &DStream);
+
+You can retrieve any bitfield you eventually stored into the bitStream (in reverse order)
+Note : maximum allowed nbBits is 25, for 32-bits compatibility
+    size_t bitField = BIT_readBits(&DStream, nbBits);
+
+All above operations only read from local register (which size depends on size_t).
+Refueling the register from memory is manually performed by the reload method.
+    endSignal = FSE_reloadDStream(&DStream);
+
+BIT_reloadDStream() result tells if there is still some more data to read from DStream.
+BIT_DStream_unfinished : there is still some data left into the DStream.
+BIT_DStream_endOfBuffer : Dstream reached end of buffer. Its container may no longer be completely filled.
+BIT_DStream_completed : Dstream reached its exact end, corresponding in general to decompression completed.
+BIT_DStream_tooFar : Dstream went too far. Decompression result is corrupted.
+
+When reaching end of buffer (BIT_DStream_endOfBuffer), progress slowly, notably if you decode multiple symbols per loop,
+to properly detect the exact end of stream.
+After each decoded symbol, check if DStream is fully consumed using this simple test :
+    BIT_reloadDStream(&DStream) >= BIT_DStream_completed
+
+When it's done, verify decompression is fully completed, by checking both DStream and the relevant states.
+Checking if DStream has reached its end is performed by :
+    BIT_endOfDStream(&DStream);
+Check also the states. There might be some symbols left there, if some high probability ones (>50%) are possible.
+    FSE_endOfDState(&DState);
+*/
+
+
+/* *****************************************
+*  FSE unsafe API
+*******************************************/
+static unsigned char FSE_decodeSymbolFast(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD);
+/* faster, but works only if nbBits is always >= 1 (otherwise, result will be corrupted) */
+
+
+/* *****************************************
+*  Implementation of inlined functions
+*******************************************/
+typedef struct {
+    int deltaFindState;
+    U32 deltaNbBits;
+} FSE_symbolCompressionTransform; /* total 8 bytes */
+
+MEM_STATIC void FSE_initCState(FSE_CState_t* statePtr, const FSE_CTable* ct)
+{
+    const void* ptr = ct;
+    const U16* u16ptr = (const U16*) ptr;
+    const U32 tableLog = MEM_read16(ptr);
+    statePtr->value = (ptrdiff_t)1<<tableLog;
+    statePtr->stateTable = u16ptr+2;
+    statePtr->symbolTT = ct + 1 + (tableLog ? (1<<(tableLog-1)) : 1);
+    statePtr->stateLog = tableLog;
+}
+
+
+/*! FSE_initCState2() :
+*   Same as FSE_initCState(), but the first symbol to include (which will be the last to be read)
+*   uses the smallest state value possible, saving the cost of this symbol */
+MEM_STATIC void FSE_initCState2(FSE_CState_t* statePtr, const FSE_CTable* ct, U32 symbol)
+{
+    FSE_initCState(statePtr, ct);
+    {   const FSE_symbolCompressionTransform symbolTT = ((const FSE_symbolCompressionTransform*)(statePtr->symbolTT))[symbol];
+        const U16* stateTable = (const U16*)(statePtr->stateTable);
+        U32 nbBitsOut  = (U32)((symbolTT.deltaNbBits + (1<<15)) >> 16);
+        statePtr->value = (nbBitsOut << 16) - symbolTT.deltaNbBits;
+        statePtr->value = stateTable[(statePtr->value >> nbBitsOut) + symbolTT.deltaFindState];
+    }
+}
+
+MEM_STATIC void FSE_encodeSymbol(BIT_CStream_t* bitC, FSE_CState_t* statePtr, unsigned symbol)
+{
+    FSE_symbolCompressionTransform const symbolTT = ((const FSE_symbolCompressionTransform*)(statePtr->symbolTT))[symbol];
+    const U16* const stateTable = (const U16*)(statePtr->stateTable);
+    U32 const nbBitsOut  = (U32)((statePtr->value + symbolTT.deltaNbBits) >> 16);
+    BIT_addBits(bitC, statePtr->value, nbBitsOut);
+    statePtr->value = stateTable[ (statePtr->value >> nbBitsOut) + symbolTT.deltaFindState];
+}
+
+MEM_STATIC void FSE_flushCState(BIT_CStream_t* bitC, const FSE_CState_t* statePtr)
+{
+    BIT_addBits(bitC, statePtr->value, statePtr->stateLog);
+    BIT_flushBits(bitC);
+}
+
+
+/* FSE_getMaxNbBits() :
+ * Approximate maximum cost of a symbol, in bits.
+ * Fractional get rounded up (i.e. a symbol with a normalized frequency of 3 gives the same result as a frequency of 2)
+ * note 1 : assume symbolValue is valid (<= maxSymbolValue)
+ * note 2 : if freq[symbolValue]==0, @return a fake cost of tableLog+1 bits */
+MEM_STATIC U32 FSE_getMaxNbBits(const void* symbolTTPtr, U32 symbolValue)
+{
+    const FSE_symbolCompressionTransform* symbolTT = (const FSE_symbolCompressionTransform*) symbolTTPtr;
+    return (symbolTT[symbolValue].deltaNbBits + ((1<<16)-1)) >> 16;
+}
+
+/* FSE_bitCost() :
+ * Approximate symbol cost, as fractional value, using fixed-point format (accuracyLog fractional bits)
+ * note 1 : assume symbolValue is valid (<= maxSymbolValue)
+ * note 2 : if freq[symbolValue]==0, @return a fake cost of tableLog+1 bits */
+MEM_STATIC U32 FSE_bitCost(const void* symbolTTPtr, U32 tableLog, U32 symbolValue, U32 accuracyLog)
+{
+    const FSE_symbolCompressionTransform* symbolTT = (const FSE_symbolCompressionTransform*) symbolTTPtr;
+    U32 const minNbBits = symbolTT[symbolValue].deltaNbBits >> 16;
+    U32 const threshold = (minNbBits+1) << 16;
+    assert(tableLog < 16);
+    assert(accuracyLog < 31-tableLog);  /* ensure enough room for renormalization double shift */
+    {   U32 const tableSize = 1 << tableLog;
+        U32 const deltaFromThreshold = threshold - (symbolTT[symbolValue].deltaNbBits + tableSize);
+        U32 const normalizedDeltaFromThreshold = (deltaFromThreshold << accuracyLog) >> tableLog;   /* linear interpolation (very approximate) */
+        U32 const bitMultiplier = 1 << accuracyLog;
+        assert(symbolTT[symbolValue].deltaNbBits + tableSize <= threshold);
+        assert(normalizedDeltaFromThreshold <= bitMultiplier);
+        return (minNbBits+1)*bitMultiplier - normalizedDeltaFromThreshold;
+    }
+}
+
+
+/* ======    Decompression    ====== */
+
+typedef struct {
+    U16 tableLog;
+    U16 fastMode;
+} FSE_DTableHeader;   /* sizeof U32 */
+
+typedef struct
+{
+    unsigned short newState;
+    unsigned char  symbol;
+    unsigned char  nbBits;
+} FSE_decode_t;   /* size == U32 */
+
+MEM_STATIC void FSE_initDState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD, const FSE_DTable* dt)
+{
+    const void* ptr = dt;
+    const FSE_DTableHeader* const DTableH = (const FSE_DTableHeader*)ptr;
+    DStatePtr->state = BIT_readBits(bitD, DTableH->tableLog);
+    BIT_reloadDStream(bitD);
+    DStatePtr->table = dt + 1;
+}
+
+MEM_STATIC BYTE FSE_peekSymbol(const FSE_DState_t* DStatePtr)
+{
+    FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state];
+    return DInfo.symbol;
+}
+
+MEM_STATIC void FSE_updateState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD)
+{
+    FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state];
+    U32 const nbBits = DInfo.nbBits;
+    size_t const lowBits = BIT_readBits(bitD, nbBits);
+    DStatePtr->state = DInfo.newState + lowBits;
+}
+
+MEM_STATIC BYTE FSE_decodeSymbol(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD)
+{
+    FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state];
+    U32 const nbBits = DInfo.nbBits;
+    BYTE const symbol = DInfo.symbol;
+    size_t const lowBits = BIT_readBits(bitD, nbBits);
+
+    DStatePtr->state = DInfo.newState + lowBits;
+    return symbol;
+}
+
+/*! FSE_decodeSymbolFast() :
+    unsafe, only works if no symbol has a probability > 50% */
+MEM_STATIC BYTE FSE_decodeSymbolFast(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD)
+{
+    FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state];
+    U32 const nbBits = DInfo.nbBits;
+    BYTE const symbol = DInfo.symbol;
+    size_t const lowBits = BIT_readBitsFast(bitD, nbBits);
+
+    DStatePtr->state = DInfo.newState + lowBits;
+    return symbol;
+}
+
+MEM_STATIC unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr)
+{
+    return DStatePtr->state == 0;
+}
+
+
+
+#ifndef FSE_COMMONDEFS_ONLY
+
+/* **************************************************************
+*  Tuning parameters
+****************************************************************/
+/*!MEMORY_USAGE :
+*  Memory usage formula : N->2^N Bytes (examples : 10 -> 1KB; 12 -> 4KB ; 16 -> 64KB; 20 -> 1MB; etc.)
+*  Increasing memory usage improves compression ratio
+*  Reduced memory usage can improve speed, due to cache effect
+*  Recommended max value is 14, for 16KB, which nicely fits into Intel x86 L1 cache */
+#ifndef FSE_MAX_MEMORY_USAGE
+#  define FSE_MAX_MEMORY_USAGE 14
+#endif
+#ifndef FSE_DEFAULT_MEMORY_USAGE
+#  define FSE_DEFAULT_MEMORY_USAGE 13
+#endif
+#if (FSE_DEFAULT_MEMORY_USAGE > FSE_MAX_MEMORY_USAGE)
+#  error "FSE_DEFAULT_MEMORY_USAGE must be <= FSE_MAX_MEMORY_USAGE"
+#endif
+
+/*!FSE_MAX_SYMBOL_VALUE :
+*  Maximum symbol value authorized.
+*  Required for proper stack allocation */
+#ifndef FSE_MAX_SYMBOL_VALUE
+#  define FSE_MAX_SYMBOL_VALUE 255
+#endif
+
+/* **************************************************************
+*  template functions type & suffix
+****************************************************************/
+#define FSE_FUNCTION_TYPE BYTE
+#define FSE_FUNCTION_EXTENSION
+#define FSE_DECODE_TYPE FSE_decode_t
+
+
+#endif   /* !FSE_COMMONDEFS_ONLY */
+
+
+/* ***************************************************************
+*  Constants
+*****************************************************************/
+#define FSE_MAX_TABLELOG  (FSE_MAX_MEMORY_USAGE-2)
+#define FSE_MAX_TABLESIZE (1U<<FSE_MAX_TABLELOG)
+#define FSE_MAXTABLESIZE_MASK (FSE_MAX_TABLESIZE-1)
+#define FSE_DEFAULT_TABLELOG (FSE_DEFAULT_MEMORY_USAGE-2)
+#define FSE_MIN_TABLELOG 5
+
+#define FSE_TABLELOG_ABSOLUTE_MAX 15
+#if FSE_MAX_TABLELOG > FSE_TABLELOG_ABSOLUTE_MAX
+#  error "FSE_MAX_TABLELOG > FSE_TABLELOG_ABSOLUTE_MAX is not supported"
+#endif
+
+#define FSE_TABLESTEP(tableSize) (((tableSize)>>1) + ((tableSize)>>3) + 3)
+
+
+#endif /* FSE_STATIC_LINKING_ONLY */
+
+
+#if defined (__cplusplus)
+}
+#endif

external/zstd/common/fse_decompress.c

@@ -0,0 +1,405 @@
+/* ******************************************************************
+ * FSE : Finite State Entropy decoder
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ *
+ *  You can contact the author at :
+ *  - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
+ *  - Public forum : https://groups.google.com/forum/#!forum/lz4c
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+****************************************************************** */
+
+
+/* **************************************************************
+*  Includes
+****************************************************************/
+#include "debug.h"      /* assert */
+#include "bitstream.h"
+#include "compiler.h"
+#define FSE_STATIC_LINKING_ONLY
+#include "fse.h"
+#include "error_private.h"
+#define ZSTD_DEPS_NEED_MALLOC
+#include "zstd_deps.h"
+#include "bits.h"       /* ZSTD_highbit32 */
+
+
+/* **************************************************************
+*  Error Management
+****************************************************************/
+#define FSE_isError ERR_isError
+#define FSE_STATIC_ASSERT(c) DEBUG_STATIC_ASSERT(c)   /* use only *after* variable declarations */
+
+
+/* **************************************************************
+*  Templates
+****************************************************************/
+/*
+  designed to be included
+  for type-specific functions (template emulation in C)
+  Objective is to write these functions only once, for improved maintenance
+*/
+
+/* safety checks */
+#ifndef FSE_FUNCTION_EXTENSION
+#  error "FSE_FUNCTION_EXTENSION must be defined"
+#endif
+#ifndef FSE_FUNCTION_TYPE
+#  error "FSE_FUNCTION_TYPE must be defined"
+#endif
+
+/* Function names */
+#define FSE_CAT(X,Y) X##Y
+#define FSE_FUNCTION_NAME(X,Y) FSE_CAT(X,Y)
+#define FSE_TYPE_NAME(X,Y) FSE_CAT(X,Y)
+
+
+/* Function templates */
+FSE_DTable* FSE_createDTable (unsigned tableLog)
+{
+    if (tableLog > FSE_TABLELOG_ABSOLUTE_MAX) tableLog = FSE_TABLELOG_ABSOLUTE_MAX;
+    return (FSE_DTable*)ZSTD_malloc( FSE_DTABLE_SIZE_U32(tableLog) * sizeof (U32) );
+}
+
+void FSE_freeDTable (FSE_DTable* dt)
+{
+    ZSTD_free(dt);
+}
+
+static size_t FSE_buildDTable_internal(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize)
+{
+    void* const tdPtr = dt+1;   /* because *dt is unsigned, 32-bits aligned on 32-bits */
+    FSE_DECODE_TYPE* const tableDecode = (FSE_DECODE_TYPE*) (tdPtr);
+    U16* symbolNext = (U16*)workSpace;
+    BYTE* spread = (BYTE*)(symbolNext + maxSymbolValue + 1);
+
+    U32 const maxSV1 = maxSymbolValue + 1;
+    U32 const tableSize = 1 << tableLog;
+    U32 highThreshold = tableSize-1;
+
+    /* Sanity Checks */
+    if (FSE_BUILD_DTABLE_WKSP_SIZE(tableLog, maxSymbolValue) > wkspSize) return ERROR(maxSymbolValue_tooLarge);
+    if (maxSymbolValue > FSE_MAX_SYMBOL_VALUE) return ERROR(maxSymbolValue_tooLarge);
+    if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge);
+
+    /* Init, lay down lowprob symbols */
+    {   FSE_DTableHeader DTableH;
+        DTableH.tableLog = (U16)tableLog;
+        DTableH.fastMode = 1;
+        {   S16 const largeLimit= (S16)(1 << (tableLog-1));
+            U32 s;
+            for (s=0; s<maxSV1; s++) {
+                if (normalizedCounter[s]==-1) {
+                    tableDecode[highThreshold--].symbol = (FSE_FUNCTION_TYPE)s;
+                    symbolNext[s] = 1;
+                } else {
+                    if (normalizedCounter[s] >= largeLimit) DTableH.fastMode=0;
+                    symbolNext[s] = normalizedCounter[s];
+        }   }   }
+        ZSTD_memcpy(dt, &DTableH, sizeof(DTableH));
+    }
+
+    /* Spread symbols */
+    if (highThreshold == tableSize - 1) {
+        size_t const tableMask = tableSize-1;
+        size_t const step = FSE_TABLESTEP(tableSize);
+        /* First lay down the symbols in order.
+         * We use a uint64_t to lay down 8 bytes at a time. This reduces branch
+         * misses since small blocks generally have small table logs, so nearly
+         * all symbols have counts <= 8. We ensure we have 8 bytes at the end of
+         * our buffer to handle the over-write.
+         */
+        {
+            U64 const add = 0x0101010101010101ull;
+            size_t pos = 0;
+            U64 sv = 0;
+            U32 s;
+            for (s=0; s<maxSV1; ++s, sv += add) {
+                int i;
+                int const n = normalizedCounter[s];
+                MEM_write64(spread + pos, sv);
+                for (i = 8; i < n; i += 8) {
+                    MEM_write64(spread + pos + i, sv);
+                }
+                pos += n;
+            }
+        }
+        /* Now we spread those positions across the table.
+         * The benefit of doing it in two stages is that we avoid the
+         * variable size inner loop, which caused lots of branch misses.
+         * Now we can run through all the positions without any branch misses.
+         * We unroll the loop twice, since that is what empirically worked best.
+         */
+        {
+            size_t position = 0;
+            size_t s;
+            size_t const unroll = 2;
+            assert(tableSize % unroll == 0); /* FSE_MIN_TABLELOG is 5 */
+            for (s = 0; s < (size_t)tableSize; s += unroll) {
+                size_t u;
+                for (u = 0; u < unroll; ++u) {
+                    size_t const uPosition = (position + (u * step)) & tableMask;
+                    tableDecode[uPosition].symbol = spread[s + u];
+                }
+                position = (position + (unroll * step)) & tableMask;
+            }
+            assert(position == 0);
+        }
+    } else {
+        U32 const tableMask = tableSize-1;
+        U32 const step = FSE_TABLESTEP(tableSize);
+        U32 s, position = 0;
+        for (s=0; s<maxSV1; s++) {
+            int i;
+            for (i=0; i<normalizedCounter[s]; i++) {
+                tableDecode[position].symbol = (FSE_FUNCTION_TYPE)s;
+                position = (position + step) & tableMask;
+                while (position > highThreshold) position = (position + step) & tableMask;   /* lowprob area */
+        }   }
+        if (position!=0) return ERROR(GENERIC);   /* position must reach all cells once, otherwise normalizedCounter is incorrect */
+    }
+
+    /* Build Decoding table */
+    {   U32 u;
+        for (u=0; u<tableSize; u++) {
+            FSE_FUNCTION_TYPE const symbol = (FSE_FUNCTION_TYPE)(tableDecode[u].symbol);
+            U32 const nextState = symbolNext[symbol]++;
+            tableDecode[u].nbBits = (BYTE) (tableLog - ZSTD_highbit32(nextState) );
+            tableDecode[u].newState = (U16) ( (nextState << tableDecode[u].nbBits) - tableSize);
+    }   }
+
+    return 0;
+}
+
+size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize)
+{
+    return FSE_buildDTable_internal(dt, normalizedCounter, maxSymbolValue, tableLog, workSpace, wkspSize);
+}
+
+
+#ifndef FSE_COMMONDEFS_ONLY
+
+/*-*******************************************************
+*  Decompression (Byte symbols)
+*********************************************************/
+size_t FSE_buildDTable_rle (FSE_DTable* dt, BYTE symbolValue)
+{
+    void* ptr = dt;
+    FSE_DTableHeader* const DTableH = (FSE_DTableHeader*)ptr;
+    void* dPtr = dt + 1;
+    FSE_decode_t* const cell = (FSE_decode_t*)dPtr;
+
+    DTableH->tableLog = 0;
+    DTableH->fastMode = 0;
+
+    cell->newState = 0;
+    cell->symbol = symbolValue;
+    cell->nbBits = 0;
+
+    return 0;
+}
+
+
+size_t FSE_buildDTable_raw (FSE_DTable* dt, unsigned nbBits)
+{
+    void* ptr = dt;
+    FSE_DTableHeader* const DTableH = (FSE_DTableHeader*)ptr;
+    void* dPtr = dt + 1;
+    FSE_decode_t* const dinfo = (FSE_decode_t*)dPtr;
+    const unsigned tableSize = 1 << nbBits;
+    const unsigned tableMask = tableSize - 1;
+    const unsigned maxSV1 = tableMask+1;
+    unsigned s;
+
+    /* Sanity checks */
+    if (nbBits < 1) return ERROR(GENERIC);         /* min size */
+
+    /* Build Decoding Table */
+    DTableH->tableLog = (U16)nbBits;
+    DTableH->fastMode = 1;
+    for (s=0; s<maxSV1; s++) {
+        dinfo[s].newState = 0;
+        dinfo[s].symbol = (BYTE)s;
+        dinfo[s].nbBits = (BYTE)nbBits;
+    }
+
+    return 0;
+}
+
+FORCE_INLINE_TEMPLATE size_t FSE_decompress_usingDTable_generic(
+          void* dst, size_t maxDstSize,
+    const void* cSrc, size_t cSrcSize,
+    const FSE_DTable* dt, const unsigned fast)
+{
+    BYTE* const ostart = (BYTE*) dst;
+    BYTE* op = ostart;
+    BYTE* const omax = op + maxDstSize;
+    BYTE* const olimit = omax-3;
+
+    BIT_DStream_t bitD;
+    FSE_DState_t state1;
+    FSE_DState_t state2;
+
+    /* Init */
+    CHECK_F(BIT_initDStream(&bitD, cSrc, cSrcSize));
+
+    FSE_initDState(&state1, &bitD, dt);
+    FSE_initDState(&state2, &bitD, dt);
+
+#define FSE_GETSYMBOL(statePtr) fast ? FSE_decodeSymbolFast(statePtr, &bitD) : FSE_decodeSymbol(statePtr, &bitD)
+
+    /* 4 symbols per loop */
+    for ( ; (BIT_reloadDStream(&bitD)==BIT_DStream_unfinished) & (op<olimit) ; op+=4) {
+        op[0] = FSE_GETSYMBOL(&state1);
+
+        if (FSE_MAX_TABLELOG*2+7 > sizeof(bitD.bitContainer)*8)    /* This test must be static */
+            BIT_reloadDStream(&bitD);
+
+        op[1] = FSE_GETSYMBOL(&state2);
+
+        if (FSE_MAX_TABLELOG*4+7 > sizeof(bitD.bitContainer)*8)    /* This test must be static */
+            { if (BIT_reloadDStream(&bitD) > BIT_DStream_unfinished) { op+=2; break; } }
+
+        op[2] = FSE_GETSYMBOL(&state1);
+
+        if (FSE_MAX_TABLELOG*2+7 > sizeof(bitD.bitContainer)*8)    /* This test must be static */
+            BIT_reloadDStream(&bitD);
+
+        op[3] = FSE_GETSYMBOL(&state2);
+    }
+
+    /* tail */
+    /* note : BIT_reloadDStream(&bitD) >= FSE_DStream_partiallyFilled; Ends at exactly BIT_DStream_completed */
+    while (1) {
+        if (op>(omax-2)) return ERROR(dstSize_tooSmall);
+        *op++ = FSE_GETSYMBOL(&state1);
+        if (BIT_reloadDStream(&bitD)==BIT_DStream_overflow) {
+            *op++ = FSE_GETSYMBOL(&state2);
+            break;
+        }
+
+        if (op>(omax-2)) return ERROR(dstSize_tooSmall);
+        *op++ = FSE_GETSYMBOL(&state2);
+        if (BIT_reloadDStream(&bitD)==BIT_DStream_overflow) {
+            *op++ = FSE_GETSYMBOL(&state1);
+            break;
+    }   }
+
+    return op-ostart;
+}
+
+
+size_t FSE_decompress_usingDTable(void* dst, size_t originalSize,
+                            const void* cSrc, size_t cSrcSize,
+                            const FSE_DTable* dt)
+{
+    const void* ptr = dt;
+    const FSE_DTableHeader* DTableH = (const FSE_DTableHeader*)ptr;
+    const U32 fastMode = DTableH->fastMode;
+
+    /* select fast mode (static) */
+    if (fastMode) return FSE_decompress_usingDTable_generic(dst, originalSize, cSrc, cSrcSize, dt, 1);
+    return FSE_decompress_usingDTable_generic(dst, originalSize, cSrc, cSrcSize, dt, 0);
+}
+
+
+size_t FSE_decompress_wksp(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize)
+{
+    return FSE_decompress_wksp_bmi2(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, /* bmi2 */ 0);
+}
+
+typedef struct {
+    short ncount[FSE_MAX_SYMBOL_VALUE + 1];
+    FSE_DTable dtable[1]; /* Dynamically sized */
+} FSE_DecompressWksp;
+
+
+FORCE_INLINE_TEMPLATE size_t FSE_decompress_wksp_body(
+        void* dst, size_t dstCapacity,
+        const void* cSrc, size_t cSrcSize,
+        unsigned maxLog, void* workSpace, size_t wkspSize,
+        int bmi2)
+{
+    const BYTE* const istart = (const BYTE*)cSrc;
+    const BYTE* ip = istart;
+    unsigned tableLog;
+    unsigned maxSymbolValue = FSE_MAX_SYMBOL_VALUE;
+    FSE_DecompressWksp* const wksp = (FSE_DecompressWksp*)workSpace;
+
+    DEBUG_STATIC_ASSERT((FSE_MAX_SYMBOL_VALUE + 1) % 2 == 0);
+    if (wkspSize < sizeof(*wksp)) return ERROR(GENERIC);
+
+    /* normal FSE decoding mode */
+    {
+        size_t const NCountLength = FSE_readNCount_bmi2(wksp->ncount, &maxSymbolValue, &tableLog, istart, cSrcSize, bmi2);
+        if (FSE_isError(NCountLength)) return NCountLength;
+        if (tableLog > maxLog) return ERROR(tableLog_tooLarge);
+        assert(NCountLength <= cSrcSize);
+        ip += NCountLength;
+        cSrcSize -= NCountLength;
+    }
+
+    if (FSE_DECOMPRESS_WKSP_SIZE(tableLog, maxSymbolValue) > wkspSize) return ERROR(tableLog_tooLarge);
+    assert(sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog) <= wkspSize);
+    workSpace = (BYTE*)workSpace + sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog);
+    wkspSize -= sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog);
+
+    CHECK_F( FSE_buildDTable_internal(wksp->dtable, wksp->ncount, maxSymbolValue, tableLog, workSpace, wkspSize) );
+
+    {
+        const void* ptr = wksp->dtable;
+        const FSE_DTableHeader* DTableH = (const FSE_DTableHeader*)ptr;
+        const U32 fastMode = DTableH->fastMode;
+
+        /* select fast mode (static) */
+        if (fastMode) return FSE_decompress_usingDTable_generic(dst, dstCapacity, ip, cSrcSize, wksp->dtable, 1);
+        return FSE_decompress_usingDTable_generic(dst, dstCapacity, ip, cSrcSize, wksp->dtable, 0);
+    }
+}
+
+/* Avoids the FORCE_INLINE of the _body() function. */
+static size_t FSE_decompress_wksp_body_default(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize)
+{
+    return FSE_decompress_wksp_body(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, 0);
+}
+
+#if DYNAMIC_BMI2
+BMI2_TARGET_ATTRIBUTE static size_t FSE_decompress_wksp_body_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize)
+{
+    return FSE_decompress_wksp_body(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, 1);
+}
+#endif
+
+size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2)
+{
+#if DYNAMIC_BMI2
+    if (bmi2) {
+        return FSE_decompress_wksp_body_bmi2(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize);
+    }
+#endif
+    (void)bmi2;
+    return FSE_decompress_wksp_body_default(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize);
+}
+
+
+typedef FSE_DTable DTable_max_t[FSE_DTABLE_SIZE_U32(FSE_MAX_TABLELOG)];
+
+#ifndef ZSTD_NO_UNUSED_FUNCTIONS
+size_t FSE_buildDTable(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog) {
+    U32 wksp[FSE_BUILD_DTABLE_WKSP_SIZE_U32(FSE_TABLELOG_ABSOLUTE_MAX, FSE_MAX_SYMBOL_VALUE)];
+    return FSE_buildDTable_wksp(dt, normalizedCounter, maxSymbolValue, tableLog, wksp, sizeof(wksp));
+}
+
+size_t FSE_decompress(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize)
+{
+    /* Static analyzer seems unable to understand this table will be properly initialized later */
+    U32 wksp[FSE_DECOMPRESS_WKSP_SIZE_U32(FSE_MAX_TABLELOG, FSE_MAX_SYMBOL_VALUE)];
+    return FSE_decompress_wksp(dst, dstCapacity, cSrc, cSrcSize, FSE_MAX_TABLELOG, wksp, sizeof(wksp));
+}
+#endif
+
+
+#endif   /* FSE_COMMONDEFS_ONLY */

external/zstd/common/huf.h

@@ -0,0 +1,377 @@
+/* ******************************************************************
+ * huff0 huffman codec,
+ * part of Finite State Entropy library
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ *
+ * You can contact the author at :
+ * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+****************************************************************** */
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
+#ifndef HUF_H_298734234
+#define HUF_H_298734234
+
+/* *** Dependencies *** */
+#include "zstd_deps.h"    /* size_t */
+
+
+/* *** library symbols visibility *** */
+/* Note : when linking with -fvisibility=hidden on gcc, or by default on Visual,
+ *        HUF symbols remain "private" (internal symbols for library only).
+ *        Set macro FSE_DLL_EXPORT to 1 if you want HUF symbols visible on DLL interface */
+#if defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) && defined(__GNUC__) && (__GNUC__ >= 4)
+#  define HUF_PUBLIC_API __attribute__ ((visibility ("default")))
+#elif defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1)   /* Visual expected */
+#  define HUF_PUBLIC_API __declspec(dllexport)
+#elif defined(FSE_DLL_IMPORT) && (FSE_DLL_IMPORT==1)
+#  define HUF_PUBLIC_API __declspec(dllimport)  /* not required, just to generate faster code (saves a function pointer load from IAT and an indirect jump) */
+#else
+#  define HUF_PUBLIC_API
+#endif
+
+
+/* ========================== */
+/* ***  simple functions  *** */
+/* ========================== */
+
+/** HUF_compress() :
+ *  Compress content from buffer 'src', of size 'srcSize', into buffer 'dst'.
+ * 'dst' buffer must be already allocated.
+ *  Compression runs faster if `dstCapacity` >= HUF_compressBound(srcSize).
+ * `srcSize` must be <= `HUF_BLOCKSIZE_MAX` == 128 KB.
+ * @return : size of compressed data (<= `dstCapacity`).
+ *  Special values : if return == 0, srcData is not compressible => Nothing is stored within dst !!!
+ *                   if HUF_isError(return), compression failed (more details using HUF_getErrorName())
+ */
+HUF_PUBLIC_API size_t HUF_compress(void* dst, size_t dstCapacity,
+                             const void* src, size_t srcSize);
+
+/** HUF_decompress() :
+ *  Decompress HUF data from buffer 'cSrc', of size 'cSrcSize',
+ *  into already allocated buffer 'dst', of minimum size 'dstSize'.
+ * `originalSize` : **must** be the ***exact*** size of original (uncompressed) data.
+ *  Note : in contrast with FSE, HUF_decompress can regenerate
+ *         RLE (cSrcSize==1) and uncompressed (cSrcSize==dstSize) data,
+ *         because it knows size to regenerate (originalSize).
+ * @return : size of regenerated data (== originalSize),
+ *           or an error code, which can be tested using HUF_isError()
+ */
+HUF_PUBLIC_API size_t HUF_decompress(void* dst,  size_t originalSize,
+                               const void* cSrc, size_t cSrcSize);
+
+
+/* ***   Tool functions *** */
+#define HUF_BLOCKSIZE_MAX (128 * 1024)                  /**< maximum input size for a single block compressed with HUF_compress */
+HUF_PUBLIC_API size_t HUF_compressBound(size_t size);   /**< maximum compressed size (worst case) */
+
+/* Error Management */
+HUF_PUBLIC_API unsigned    HUF_isError(size_t code);       /**< tells if a return value is an error code */
+HUF_PUBLIC_API const char* HUF_getErrorName(size_t code);  /**< provides error code string (useful for debugging) */
+
+
+/* ***   Advanced function   *** */
+
+/** HUF_compress2() :
+ *  Same as HUF_compress(), but offers control over `maxSymbolValue` and `tableLog`.
+ * `maxSymbolValue` must be <= HUF_SYMBOLVALUE_MAX .
+ * `tableLog` must be `<= HUF_TABLELOG_MAX` . */
+HUF_PUBLIC_API size_t HUF_compress2 (void* dst, size_t dstCapacity,
+                               const void* src, size_t srcSize,
+                               unsigned maxSymbolValue, unsigned tableLog);
+
+/** HUF_compress4X_wksp() :
+ *  Same as HUF_compress2(), but uses externally allocated @workSpace.
+ * @workSpace's size, aka @wkspSize, must be >= HUF_WORKSPACE_SIZE
+ * @srcSize must be >= 6
+ */
+#define HUF_WORKSPACE_SIZE ((8 << 10) + 512 /* sorting scratch space */)
+#define HUF_WORKSPACE_SIZE_U64 (HUF_WORKSPACE_SIZE / sizeof(U64))
+HUF_PUBLIC_API size_t HUF_compress4X_wksp (void* dst, size_t dstCapacity,
+                                     const void* src, size_t srcSize,
+                                     unsigned maxSymbolValue, unsigned tableLog,
+                                     void* workSpace, size_t wkspSize);
+
+#endif   /* HUF_H_298734234 */
+
+/* ******************************************************************
+ *  WARNING !!
+ *  The following section contains advanced and experimental definitions
+ *  which shall never be used in the context of a dynamic library,
+ *  because they are not guaranteed to remain stable in the future.
+ *  Only consider them in association with static linking.
+ * *****************************************************************/
+#if defined(HUF_STATIC_LINKING_ONLY) && !defined(HUF_H_HUF_STATIC_LINKING_ONLY)
+#define HUF_H_HUF_STATIC_LINKING_ONLY
+
+/* *** Dependencies *** */
+#include "mem.h"   /* U32 */
+#define FSE_STATIC_LINKING_ONLY
+#include "fse.h"
+
+
+/* *** Constants *** */
+#define HUF_TABLELOG_MAX      12      /* max runtime value of tableLog (due to static allocation); can be modified up to HUF_TABLELOG_ABSOLUTEMAX */
+#define HUF_TABLELOG_DEFAULT  11      /* default tableLog value when none specified */
+#define HUF_SYMBOLVALUE_MAX  255
+
+#define HUF_TABLELOG_ABSOLUTEMAX  12  /* absolute limit of HUF_MAX_TABLELOG. Beyond that value, code does not work */
+#if (HUF_TABLELOG_MAX > HUF_TABLELOG_ABSOLUTEMAX)
+#  error "HUF_TABLELOG_MAX is too large !"
+#endif
+
+
+/* ****************************************
+*  Static allocation
+******************************************/
+/* HUF buffer bounds */
+#define HUF_CTABLEBOUND 129
+#define HUF_BLOCKBOUND(size) (size + (size>>8) + 8)   /* only true when incompressible is pre-filtered with fast heuristic */
+#define HUF_COMPRESSBOUND(size) (HUF_CTABLEBOUND + HUF_BLOCKBOUND(size))   /* Macro version, useful for static allocation */
+
+/* static allocation of HUF's Compression Table */
+/* this is a private definition, just exposed for allocation and strict aliasing purpose. never EVER access its members directly */
+typedef size_t HUF_CElt;   /* consider it an incomplete type */
+#define HUF_CTABLE_SIZE_ST(maxSymbolValue)   ((maxSymbolValue)+2)   /* Use tables of size_t, for proper alignment */
+#define HUF_CTABLE_SIZE(maxSymbolValue)       (HUF_CTABLE_SIZE_ST(maxSymbolValue) * sizeof(size_t))
+#define HUF_CREATE_STATIC_CTABLE(name, maxSymbolValue) \
+    HUF_CElt name[HUF_CTABLE_SIZE_ST(maxSymbolValue)] /* no final ; */
+
+/* static allocation of HUF's DTable */
+typedef U32 HUF_DTable;
+#define HUF_DTABLE_SIZE(maxTableLog)   (1 + (1<<(maxTableLog)))
+#define HUF_CREATE_STATIC_DTABLEX1(DTable, maxTableLog) \
+        HUF_DTable DTable[HUF_DTABLE_SIZE((maxTableLog)-1)] = { ((U32)((maxTableLog)-1) * 0x01000001) }
+#define HUF_CREATE_STATIC_DTABLEX2(DTable, maxTableLog) \
+        HUF_DTable DTable[HUF_DTABLE_SIZE(maxTableLog)] = { ((U32)(maxTableLog) * 0x01000001) }
+
+
+/* ****************************************
+*  Advanced decompression functions
+******************************************/
+size_t HUF_decompress4X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< single-symbol decoder */
+#ifndef HUF_FORCE_DECOMPRESS_X1
+size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< double-symbols decoder */
+#endif
+
+size_t HUF_decompress4X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< decodes RLE and uncompressed */
+size_t HUF_decompress4X_hufOnly(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< considers RLE and uncompressed as errors */
+size_t HUF_decompress4X_hufOnly_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< considers RLE and uncompressed as errors */
+size_t HUF_decompress4X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< single-symbol decoder */
+size_t HUF_decompress4X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize);   /**< single-symbol decoder */
+#ifndef HUF_FORCE_DECOMPRESS_X1
+size_t HUF_decompress4X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< double-symbols decoder */
+size_t HUF_decompress4X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize);   /**< double-symbols decoder */
+#endif
+
+
+/* ****************************************
+ *  HUF detailed API
+ * ****************************************/
+#define HUF_OPTIMAL_DEPTH_THRESHOLD ZSTD_btultra
+typedef enum {
+   HUF_depth_fast, /** Use heuristic to find the table depth**/
+   HUF_depth_optimal /** Test possible table depths to find the one that produces the smallest header + encoded size**/
+ } HUF_depth_mode;
+
+/*! HUF_compress() does the following:
+ *  1. count symbol occurrence from source[] into table count[] using FSE_count() (exposed within "fse.h")
+ *  2. (optional) refine tableLog using HUF_optimalTableLog()
+ *  3. build Huffman table from count using HUF_buildCTable()
+ *  4. save Huffman table to memory buffer using HUF_writeCTable()
+ *  5. encode the data stream using HUF_compress4X_usingCTable()
+ *
+ *  The following API allows targeting specific sub-functions for advanced tasks.
+ *  For example, it's possible to compress several blocks using the same 'CTable',
+ *  or to save and regenerate 'CTable' using external methods.
+ */
+unsigned HUF_minTableLog(unsigned symbolCardinality);
+unsigned HUF_cardinality(const unsigned* count, unsigned maxSymbolValue);
+unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, void* workSpace,
+ size_t wkspSize, HUF_CElt* table, const unsigned* count, HUF_depth_mode depthMode); /* table is used as scratch space for building and testing tables, not a return value */
+size_t HUF_buildCTable (HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue, unsigned maxNbBits);   /* @return : maxNbBits; CTable and count can overlap. In which case, CTable will overwrite count content */
+size_t HUF_writeCTable (void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog);
+size_t HUF_writeCTable_wksp(void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog, void* workspace, size_t workspaceSize);
+size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable);
+size_t HUF_compress4X_usingCTable_bmi2(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int bmi2);
+size_t HUF_estimateCompressedSize(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
+int HUF_validateCTable(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
+
+typedef enum {
+   HUF_repeat_none,  /**< Cannot use the previous table */
+   HUF_repeat_check, /**< Can use the previous table but it must be checked. Note : The previous table must have been constructed by HUF_compress{1, 4}X_repeat */
+   HUF_repeat_valid  /**< Can use the previous table and it is assumed to be valid */
+ } HUF_repeat;
+
+/** HUF_compress4X_repeat() :
+ *  Same as HUF_compress4X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
+ *  If it uses hufTable it does not modify hufTable or repeat.
+ *  If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used.
+ *  If preferRepeat then the old table will always be used if valid.
+ *  If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding */
+size_t HUF_compress4X_repeat(void* dst, size_t dstSize,
+                       const void* src, size_t srcSize,
+                       unsigned maxSymbolValue, unsigned tableLog,
+                       void* workSpace, size_t wkspSize,    /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
+                       HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2,
+                       int suspectUncompressible, HUF_depth_mode depthMode);
+
+/** HUF_buildCTable_wksp() :
+ *  Same as HUF_buildCTable(), but using externally allocated scratch buffer.
+ * `workSpace` must be aligned on 4-bytes boundaries, and its size must be >= HUF_CTABLE_WORKSPACE_SIZE.
+ */
+#define HUF_CTABLE_WORKSPACE_SIZE_U32 (2*HUF_SYMBOLVALUE_MAX +1 +1)
+#define HUF_CTABLE_WORKSPACE_SIZE (HUF_CTABLE_WORKSPACE_SIZE_U32 * sizeof(unsigned))
+size_t HUF_buildCTable_wksp (HUF_CElt* tree,
+                       const unsigned* count, U32 maxSymbolValue, U32 maxNbBits,
+                             void* workSpace, size_t wkspSize);
+
+/*! HUF_readStats() :
+ *  Read compact Huffman tree, saved by HUF_writeCTable().
+ * `huffWeight` is destination buffer.
+ * @return : size read from `src` , or an error Code .
+ *  Note : Needed by HUF_readCTable() and HUF_readDTableXn() . */
+size_t HUF_readStats(BYTE* huffWeight, size_t hwSize,
+                     U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr,
+                     const void* src, size_t srcSize);
+
+/*! HUF_readStats_wksp() :
+ * Same as HUF_readStats() but takes an external workspace which must be
+ * 4-byte aligned and its size must be >= HUF_READ_STATS_WORKSPACE_SIZE.
+ * If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0.
+ */
+#define HUF_READ_STATS_WORKSPACE_SIZE_U32 FSE_DECOMPRESS_WKSP_SIZE_U32(6, HUF_TABLELOG_MAX-1)
+#define HUF_READ_STATS_WORKSPACE_SIZE (HUF_READ_STATS_WORKSPACE_SIZE_U32 * sizeof(unsigned))
+size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize,
+                          U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr,
+                          const void* src, size_t srcSize,
+                          void* workspace, size_t wkspSize,
+                          int bmi2);
+
+/** HUF_readCTable() :
+ *  Loading a CTable saved with HUF_writeCTable() */
+size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize, unsigned *hasZeroWeights);
+
+/** HUF_getNbBitsFromCTable() :
+ *  Read nbBits from CTable symbolTable, for symbol `symbolValue` presumed <= HUF_SYMBOLVALUE_MAX
+ *  Note 1 : is not inlined, as HUF_CElt definition is private */
+U32 HUF_getNbBitsFromCTable(const HUF_CElt* symbolTable, U32 symbolValue);
+
+/*
+ * HUF_decompress() does the following:
+ * 1. select the decompression algorithm (X1, X2) based on pre-computed heuristics
+ * 2. build Huffman table from save, using HUF_readDTableX?()
+ * 3. decode 1 or 4 segments in parallel using HUF_decompress?X?_usingDTable()
+ */
+
+/** HUF_selectDecoder() :
+ *  Tells which decoder is likely to decode faster,
+ *  based on a set of pre-computed metrics.
+ * @return : 0==HUF_decompress4X1, 1==HUF_decompress4X2 .
+ *  Assumption : 0 < dstSize <= 128 KB */
+U32 HUF_selectDecoder (size_t dstSize, size_t cSrcSize);
+
+/**
+ *  The minimum workspace size for the `workSpace` used in
+ *  HUF_readDTableX1_wksp() and HUF_readDTableX2_wksp().
+ *
+ *  The space used depends on HUF_TABLELOG_MAX, ranging from ~1500 bytes when
+ *  HUF_TABLE_LOG_MAX=12 to ~1850 bytes when HUF_TABLE_LOG_MAX=15.
+ *  Buffer overflow errors may potentially occur if code modifications result in
+ *  a required workspace size greater than that specified in the following
+ *  macro.
+ */
+#define HUF_DECOMPRESS_WORKSPACE_SIZE ((2 << 10) + (1 << 9))
+#define HUF_DECOMPRESS_WORKSPACE_SIZE_U32 (HUF_DECOMPRESS_WORKSPACE_SIZE / sizeof(U32))
+
+#ifndef HUF_FORCE_DECOMPRESS_X2
+size_t HUF_readDTableX1 (HUF_DTable* DTable, const void* src, size_t srcSize);
+size_t HUF_readDTableX1_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize);
+#endif
+#ifndef HUF_FORCE_DECOMPRESS_X1
+size_t HUF_readDTableX2 (HUF_DTable* DTable, const void* src, size_t srcSize);
+size_t HUF_readDTableX2_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize);
+#endif
+
+size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
+#ifndef HUF_FORCE_DECOMPRESS_X2
+size_t HUF_decompress4X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
+#endif
+#ifndef HUF_FORCE_DECOMPRESS_X1
+size_t HUF_decompress4X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
+#endif
+
+
+/* ====================== */
+/* single stream variants */
+/* ====================== */
+
+size_t HUF_compress1X (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog);
+size_t HUF_compress1X_wksp (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize);  /**< `workSpace` must be a table of at least HUF_WORKSPACE_SIZE_U64 U64 */
+size_t HUF_compress1X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable);
+size_t HUF_compress1X_usingCTable_bmi2(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int bmi2);
+/** HUF_compress1X_repeat() :
+ *  Same as HUF_compress1X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
+ *  If it uses hufTable it does not modify hufTable or repeat.
+ *  If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used.
+ *  If preferRepeat then the old table will always be used if valid.
+ *  If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding */
+size_t HUF_compress1X_repeat(void* dst, size_t dstSize,
+                       const void* src, size_t srcSize,
+                       unsigned maxSymbolValue, unsigned tableLog,
+                       void* workSpace, size_t wkspSize,   /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
+                       HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2,
+                       int suspectUncompressible, HUF_depth_mode depthMode);
+
+size_t HUF_decompress1X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /* single-symbol decoder */
+#ifndef HUF_FORCE_DECOMPRESS_X1
+size_t HUF_decompress1X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /* double-symbol decoder */
+#endif
+
+size_t HUF_decompress1X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);
+size_t HUF_decompress1X_DCtx_wksp (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize);
+#ifndef HUF_FORCE_DECOMPRESS_X2
+size_t HUF_decompress1X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< single-symbol decoder */
+size_t HUF_decompress1X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize);   /**< single-symbol decoder */
+#endif
+#ifndef HUF_FORCE_DECOMPRESS_X1
+size_t HUF_decompress1X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< double-symbols decoder */
+size_t HUF_decompress1X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize);   /**< double-symbols decoder */
+#endif
+
+size_t HUF_decompress1X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);   /**< automatic selection of sing or double symbol decoder, based on DTable */
+#ifndef HUF_FORCE_DECOMPRESS_X2
+size_t HUF_decompress1X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
+#endif
+#ifndef HUF_FORCE_DECOMPRESS_X1
+size_t HUF_decompress1X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
+#endif
+
+/* BMI2 variants.
+ * If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0.
+ */
+size_t HUF_decompress1X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2);
+#ifndef HUF_FORCE_DECOMPRESS_X2
+size_t HUF_decompress1X1_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2);
+#endif
+size_t HUF_decompress4X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2);
+size_t HUF_decompress4X_hufOnly_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2);
+#ifndef HUF_FORCE_DECOMPRESS_X2
+size_t HUF_readDTableX1_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2);
+#endif
+#ifndef HUF_FORCE_DECOMPRESS_X1
+size_t HUF_readDTableX2_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2);
+#endif
+
+#endif /* HUF_STATIC_LINKING_ONLY */
+
+#if defined (__cplusplus)
+}
+#endif

external/zstd/common/mem.h

@@ -0,0 +1,435 @@
+/*
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * All rights reserved.
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+ */
+
+#ifndef MEM_H_MODULE
+#define MEM_H_MODULE
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
+/*-****************************************
+*  Dependencies
+******************************************/
+#include <stddef.h>  /* size_t, ptrdiff_t */
+#include "compiler.h"  /* __has_builtin */
+#include "debug.h"  /* DEBUG_STATIC_ASSERT */
+#include "zstd_deps.h"  /* ZSTD_memcpy */
+
+
+/*-****************************************
+*  Compiler specifics
+******************************************/
+#if defined(_MSC_VER)   /* Visual Studio */
+#   include <stdlib.h>  /* _byteswap_ulong */
+#   include <intrin.h>  /* _byteswap_* */
+#endif
+#if defined(__GNUC__)
+#  define MEM_STATIC static __inline __attribute__((unused))
+#elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
+#  define MEM_STATIC static inline
+#elif defined(_MSC_VER)
+#  define MEM_STATIC static __inline
+#else
+#  define MEM_STATIC static  /* this version may generate warnings for unused static functions; disable the relevant warning */
+#endif
+
+/*-**************************************************************
+*  Basic Types
+*****************************************************************/
+#if  !defined (__VMS) && (defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) )
+#  if defined(_AIX)
+#    include <inttypes.h>
+#  else
+#    include <stdint.h> /* intptr_t */
+#  endif
+  typedef   uint8_t BYTE;
+  typedef   uint8_t U8;
+  typedef    int8_t S8;
+  typedef  uint16_t U16;
+  typedef   int16_t S16;
+  typedef  uint32_t U32;
+  typedef   int32_t S32;
+  typedef  uint64_t U64;
+  typedef   int64_t S64;
+#else
+# include <limits.h>
+#if CHAR_BIT != 8
+#  error "this implementation requires char to be exactly 8-bit type"
+#endif
+  typedef unsigned char      BYTE;
+  typedef unsigned char      U8;
+  typedef   signed char      S8;
+#if USHRT_MAX != 65535
+#  error "this implementation requires short to be exactly 16-bit type"
+#endif
+  typedef unsigned short      U16;
+  typedef   signed short      S16;
+#if UINT_MAX != 4294967295
+#  error "this implementation requires int to be exactly 32-bit type"
+#endif
+  typedef unsigned int        U32;
+  typedef   signed int        S32;
+/* note : there are no limits defined for long long type in C90.
+ * limits exist in C99, however, in such case, <stdint.h> is preferred */
+  typedef unsigned long long  U64;
+  typedef   signed long long  S64;
+#endif
+
+
+/*-**************************************************************
+*  Memory I/O API
+*****************************************************************/
+/*=== Static platform detection ===*/
+MEM_STATIC unsigned MEM_32bits(void);
+MEM_STATIC unsigned MEM_64bits(void);
+MEM_STATIC unsigned MEM_isLittleEndian(void);
+
+/*=== Native unaligned read/write ===*/
+MEM_STATIC U16 MEM_read16(const void* memPtr);
+MEM_STATIC U32 MEM_read32(const void* memPtr);
+MEM_STATIC U64 MEM_read64(const void* memPtr);
+MEM_STATIC size_t MEM_readST(const void* memPtr);
+
+MEM_STATIC void MEM_write16(void* memPtr, U16 value);
+MEM_STATIC void MEM_write32(void* memPtr, U32 value);
+MEM_STATIC void MEM_write64(void* memPtr, U64 value);
+
+/*=== Little endian unaligned read/write ===*/
+MEM_STATIC U16 MEM_readLE16(const void* memPtr);
+MEM_STATIC U32 MEM_readLE24(const void* memPtr);
+MEM_STATIC U32 MEM_readLE32(const void* memPtr);
+MEM_STATIC U64 MEM_readLE64(const void* memPtr);
+MEM_STATIC size_t MEM_readLEST(const void* memPtr);
+
+MEM_STATIC void MEM_writeLE16(void* memPtr, U16 val);
+MEM_STATIC void MEM_writeLE24(void* memPtr, U32 val);
+MEM_STATIC void MEM_writeLE32(void* memPtr, U32 val32);
+MEM_STATIC void MEM_writeLE64(void* memPtr, U64 val64);
+MEM_STATIC void MEM_writeLEST(void* memPtr, size_t val);
+
+/*=== Big endian unaligned read/write ===*/
+MEM_STATIC U32 MEM_readBE32(const void* memPtr);
+MEM_STATIC U64 MEM_readBE64(const void* memPtr);
+MEM_STATIC size_t MEM_readBEST(const void* memPtr);
+
+MEM_STATIC void MEM_writeBE32(void* memPtr, U32 val32);
+MEM_STATIC void MEM_writeBE64(void* memPtr, U64 val64);
+MEM_STATIC void MEM_writeBEST(void* memPtr, size_t val);
+
+/*=== Byteswap ===*/
+MEM_STATIC U32 MEM_swap32(U32 in);
+MEM_STATIC U64 MEM_swap64(U64 in);
+MEM_STATIC size_t MEM_swapST(size_t in);
+
+
+/*-**************************************************************
+*  Memory I/O Implementation
+*****************************************************************/
+/* MEM_FORCE_MEMORY_ACCESS : For accessing unaligned memory:
+ * Method 0 : always use `memcpy()`. Safe and portable.
+ * Method 1 : Use compiler extension to set unaligned access.
+ * Method 2 : direct access. This method is portable but violate C standard.
+ *            It can generate buggy code on targets depending on alignment.
+ * Default  : method 1 if supported, else method 0
+ */
+#ifndef MEM_FORCE_MEMORY_ACCESS   /* can be defined externally, on command line for example */
+#  ifdef __GNUC__
+#    define MEM_FORCE_MEMORY_ACCESS 1
+#  endif
+#endif
+
+MEM_STATIC unsigned MEM_32bits(void) { return sizeof(size_t)==4; }
+MEM_STATIC unsigned MEM_64bits(void) { return sizeof(size_t)==8; }
+
+MEM_STATIC unsigned MEM_isLittleEndian(void)
+{
+#if defined(__BYTE_ORDER__) && defined(__ORDER_LITTLE_ENDIAN__) && (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
+    return 1;
+#elif defined(__BYTE_ORDER__) && defined(__ORDER_BIG_ENDIAN__) && (__BYTE_ORDER__ == __ORDER_BIG_ENDIAN__)
+    return 0;
+#elif defined(__clang__) && __LITTLE_ENDIAN__
+    return 1;
+#elif defined(__clang__) && __BIG_ENDIAN__
+    return 0;
+#elif defined(_MSC_VER) && (_M_AMD64 || _M_IX86)
+    return 1;
+#elif defined(__DMC__) && defined(_M_IX86)
+    return 1;
+#else
+    const union { U32 u; BYTE c[4]; } one = { 1 };   /* don't use static : performance detrimental  */
+    return one.c[0];
+#endif
+}
+
+#if defined(MEM_FORCE_MEMORY_ACCESS) && (MEM_FORCE_MEMORY_ACCESS==2)
+
+/* violates C standard, by lying on structure alignment.
+Only use if no other choice to achieve best performance on target platform */
+MEM_STATIC U16 MEM_read16(const void* memPtr) { return *(const U16*) memPtr; }
+MEM_STATIC U32 MEM_read32(const void* memPtr) { return *(const U32*) memPtr; }
+MEM_STATIC U64 MEM_read64(const void* memPtr) { return *(const U64*) memPtr; }
+MEM_STATIC size_t MEM_readST(const void* memPtr) { return *(const size_t*) memPtr; }
+
+MEM_STATIC void MEM_write16(void* memPtr, U16 value) { *(U16*)memPtr = value; }
+MEM_STATIC void MEM_write32(void* memPtr, U32 value) { *(U32*)memPtr = value; }
+MEM_STATIC void MEM_write64(void* memPtr, U64 value) { *(U64*)memPtr = value; }
+
+#elif defined(MEM_FORCE_MEMORY_ACCESS) && (MEM_FORCE_MEMORY_ACCESS==1)
+
+typedef __attribute__((aligned(1))) U16 unalign16;
+typedef __attribute__((aligned(1))) U32 unalign32;
+typedef __attribute__((aligned(1))) U64 unalign64;
+typedef __attribute__((aligned(1))) size_t unalignArch;
+
+MEM_STATIC U16 MEM_read16(const void* ptr) { return *(const unalign16*)ptr; }
+MEM_STATIC U32 MEM_read32(const void* ptr) { return *(const unalign32*)ptr; }
+MEM_STATIC U64 MEM_read64(const void* ptr) { return *(const unalign64*)ptr; }
+MEM_STATIC size_t MEM_readST(const void* ptr) { return *(const unalignArch*)ptr; }
+
+MEM_STATIC void MEM_write16(void* memPtr, U16 value) { *(unalign16*)memPtr = value; }
+MEM_STATIC void MEM_write32(void* memPtr, U32 value) { *(unalign32*)memPtr = value; }
+MEM_STATIC void MEM_write64(void* memPtr, U64 value) { *(unalign64*)memPtr = value; }
+
+#else
+
+/* default method, safe and standard.
+   can sometimes prove slower */
+
+MEM_STATIC U16 MEM_read16(const void* memPtr)
+{
+    U16 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
+}
+
+MEM_STATIC U32 MEM_read32(const void* memPtr)
+{
+    U32 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
+}
+
+MEM_STATIC U64 MEM_read64(const void* memPtr)
+{
+    U64 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
+}
+
+MEM_STATIC size_t MEM_readST(const void* memPtr)
+{
+    size_t val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
+}
+
+MEM_STATIC void MEM_write16(void* memPtr, U16 value)
+{
+    ZSTD_memcpy(memPtr, &value, sizeof(value));
+}
+
+MEM_STATIC void MEM_write32(void* memPtr, U32 value)
+{
+    ZSTD_memcpy(memPtr, &value, sizeof(value));
+}
+
+MEM_STATIC void MEM_write64(void* memPtr, U64 value)
+{
+    ZSTD_memcpy(memPtr, &value, sizeof(value));
+}
+
+#endif /* MEM_FORCE_MEMORY_ACCESS */
+
+MEM_STATIC U32 MEM_swap32_fallback(U32 in)
+{
+    return  ((in << 24) & 0xff000000 ) |
+            ((in <<  8) & 0x00ff0000 ) |
+            ((in >>  8) & 0x0000ff00 ) |
+            ((in >> 24) & 0x000000ff );
+}
+
+MEM_STATIC U32 MEM_swap32(U32 in)
+{
+#if defined(_MSC_VER)     /* Visual Studio */
+    return _byteswap_ulong(in);
+#elif (defined (__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 403)) \
+  || (defined(__clang__) && __has_builtin(__builtin_bswap32))
+    return __builtin_bswap32(in);
+#else
+    return MEM_swap32_fallback(in);
+#endif
+}
+
+MEM_STATIC U64 MEM_swap64_fallback(U64 in)
+{
+     return  ((in << 56) & 0xff00000000000000ULL) |
+            ((in << 40) & 0x00ff000000000000ULL) |
+            ((in << 24) & 0x0000ff0000000000ULL) |
+            ((in << 8)  & 0x000000ff00000000ULL) |
+            ((in >> 8)  & 0x00000000ff000000ULL) |
+            ((in >> 24) & 0x0000000000ff0000ULL) |
+            ((in >> 40) & 0x000000000000ff00ULL) |
+            ((in >> 56) & 0x00000000000000ffULL);
+}
+
+MEM_STATIC U64 MEM_swap64(U64 in)
+{
+#if defined(_MSC_VER)     /* Visual Studio */
+    return _byteswap_uint64(in);
+#elif (defined (__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 403)) \
+  || (defined(__clang__) && __has_builtin(__builtin_bswap64))
+    return __builtin_bswap64(in);
+#else
+    return MEM_swap64_fallback(in);
+#endif
+}
+
+MEM_STATIC size_t MEM_swapST(size_t in)
+{
+    if (MEM_32bits())
+        return (size_t)MEM_swap32((U32)in);
+    else
+        return (size_t)MEM_swap64((U64)in);
+}
+
+/*=== Little endian r/w ===*/
+
+MEM_STATIC U16 MEM_readLE16(const void* memPtr)
+{
+    if (MEM_isLittleEndian())
+        return MEM_read16(memPtr);
+    else {
+        const BYTE* p = (const BYTE*)memPtr;
+        return (U16)(p[0] + (p[1]<<8));
+    }
+}
+
+MEM_STATIC void MEM_writeLE16(void* memPtr, U16 val)
+{
+    if (MEM_isLittleEndian()) {
+        MEM_write16(memPtr, val);
+    } else {
+        BYTE* p = (BYTE*)memPtr;
+        p[0] = (BYTE)val;
+        p[1] = (BYTE)(val>>8);
+    }
+}
+
+MEM_STATIC U32 MEM_readLE24(const void* memPtr)
+{
+    return (U32)MEM_readLE16(memPtr) + ((U32)(((const BYTE*)memPtr)[2]) << 16);
+}
+
+MEM_STATIC void MEM_writeLE24(void* memPtr, U32 val)
+{
+    MEM_writeLE16(memPtr, (U16)val);
+    ((BYTE*)memPtr)[2] = (BYTE)(val>>16);
+}
+
+MEM_STATIC U32 MEM_readLE32(const void* memPtr)
+{
+    if (MEM_isLittleEndian())
+        return MEM_read32(memPtr);
+    else
+        return MEM_swap32(MEM_read32(memPtr));
+}
+
+MEM_STATIC void MEM_writeLE32(void* memPtr, U32 val32)
+{
+    if (MEM_isLittleEndian())
+        MEM_write32(memPtr, val32);
+    else
+        MEM_write32(memPtr, MEM_swap32(val32));
+}
+
+MEM_STATIC U64 MEM_readLE64(const void* memPtr)
+{
+    if (MEM_isLittleEndian())
+        return MEM_read64(memPtr);
+    else
+        return MEM_swap64(MEM_read64(memPtr));
+}
+
+MEM_STATIC void MEM_writeLE64(void* memPtr, U64 val64)
+{
+    if (MEM_isLittleEndian())
+        MEM_write64(memPtr, val64);
+    else
+        MEM_write64(memPtr, MEM_swap64(val64));
+}
+
+MEM_STATIC size_t MEM_readLEST(const void* memPtr)
+{
+    if (MEM_32bits())
+        return (size_t)MEM_readLE32(memPtr);
+    else
+        return (size_t)MEM_readLE64(memPtr);
+}
+
+MEM_STATIC void MEM_writeLEST(void* memPtr, size_t val)
+{
+    if (MEM_32bits())
+        MEM_writeLE32(memPtr, (U32)val);
+    else
+        MEM_writeLE64(memPtr, (U64)val);
+}
+
+/*=== Big endian r/w ===*/
+
+MEM_STATIC U32 MEM_readBE32(const void* memPtr)
+{
+    if (MEM_isLittleEndian())
+        return MEM_swap32(MEM_read32(memPtr));
+    else
+        return MEM_read32(memPtr);
+}
+
+MEM_STATIC void MEM_writeBE32(void* memPtr, U32 val32)
+{
+    if (MEM_isLittleEndian())
+        MEM_write32(memPtr, MEM_swap32(val32));
+    else
+        MEM_write32(memPtr, val32);
+}
+
+MEM_STATIC U64 MEM_readBE64(const void* memPtr)
+{
+    if (MEM_isLittleEndian())
+        return MEM_swap64(MEM_read64(memPtr));
+    else
+        return MEM_read64(memPtr);
+}
+
+MEM_STATIC void MEM_writeBE64(void* memPtr, U64 val64)
+{
+    if (MEM_isLittleEndian())
+        MEM_write64(memPtr, MEM_swap64(val64));
+    else
+        MEM_write64(memPtr, val64);
+}
+
+MEM_STATIC size_t MEM_readBEST(const void* memPtr)
+{
+    if (MEM_32bits())
+        return (size_t)MEM_readBE32(memPtr);
+    else
+        return (size_t)MEM_readBE64(memPtr);
+}
+
+MEM_STATIC void MEM_writeBEST(void* memPtr, size_t val)
+{
+    if (MEM_32bits())
+        MEM_writeBE32(memPtr, (U32)val);
+    else
+        MEM_writeBE64(memPtr, (U64)val);
+}
+
+/* code only tested on 32 and 64 bits systems */
+MEM_STATIC void MEM_check(void) { DEBUG_STATIC_ASSERT((sizeof(size_t)==4) || (sizeof(size_t)==8)); }
+
+
+#if defined (__cplusplus)
+}
+#endif
+
+#endif /* MEM_H_MODULE */

external/zstd/common/pool.c

@@ -0,0 +1,369 @@
+/*
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * All rights reserved.
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+ */
+
+
+/* ======   Dependencies   ======= */
+#include "zstd_deps.h" /* size_t */
+#include "debug.h"     /* assert */
+#include "zstd_internal.h"  /* ZSTD_customCalloc, ZSTD_customFree */
+#include "pool.h"
+
+/* ======   Compiler specifics   ====== */
+#if defined(_MSC_VER)
+#  pragma warning(disable : 4204)        /* disable: C4204: non-constant aggregate initializer */
+#endif
+
+
+#ifdef ZSTD_MULTITHREAD
+
+#include "threading.h"   /* pthread adaptation */
+
+/* A job is a function and an opaque argument */
+typedef struct POOL_job_s {
+    POOL_function function;
+    void *opaque;
+} POOL_job;
+
+struct POOL_ctx_s {
+    ZSTD_customMem customMem;
+    /* Keep track of the threads */
+    ZSTD_pthread_t* threads;
+    size_t threadCapacity;
+    size_t threadLimit;
+
+    /* The queue is a circular buffer */
+    POOL_job *queue;
+    size_t queueHead;
+    size_t queueTail;
+    size_t queueSize;
+
+    /* The number of threads working on jobs */
+    size_t numThreadsBusy;
+    /* Indicates if the queue is empty */
+    int queueEmpty;
+
+    /* The mutex protects the queue */
+    ZSTD_pthread_mutex_t queueMutex;
+    /* Condition variable for pushers to wait on when the queue is full */
+    ZSTD_pthread_cond_t queuePushCond;
+    /* Condition variables for poppers to wait on when the queue is empty */
+    ZSTD_pthread_cond_t queuePopCond;
+    /* Indicates if the queue is shutting down */
+    int shutdown;
+};
+
+/* POOL_thread() :
+ * Work thread for the thread pool.
+ * Waits for jobs and executes them.
+ * @returns : NULL on failure else non-null.
+ */
+static void* POOL_thread(void* opaque) {
+    POOL_ctx* const ctx = (POOL_ctx*)opaque;
+    if (!ctx) { return NULL; }
+    for (;;) {
+        /* Lock the mutex and wait for a non-empty queue or until shutdown */
+        ZSTD_pthread_mutex_lock(&ctx->queueMutex);
+
+        while ( ctx->queueEmpty
+            || (ctx->numThreadsBusy >= ctx->threadLimit) ) {
+            if (ctx->shutdown) {
+                /* even if !queueEmpty, (possible if numThreadsBusy >= threadLimit),
+                 * a few threads will be shutdown while !queueEmpty,
+                 * but enough threads will remain active to finish the queue */
+                ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
+                return opaque;
+            }
+            ZSTD_pthread_cond_wait(&ctx->queuePopCond, &ctx->queueMutex);
+        }
+        /* Pop a job off the queue */
+        {   POOL_job const job = ctx->queue[ctx->queueHead];
+            ctx->queueHead = (ctx->queueHead + 1) % ctx->queueSize;
+            ctx->numThreadsBusy++;
+            ctx->queueEmpty = (ctx->queueHead == ctx->queueTail);
+            /* Unlock the mutex, signal a pusher, and run the job */
+            ZSTD_pthread_cond_signal(&ctx->queuePushCond);
+            ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
+
+            job.function(job.opaque);
+
+            /* If the intended queue size was 0, signal after finishing job */
+            ZSTD_pthread_mutex_lock(&ctx->queueMutex);
+            ctx->numThreadsBusy--;
+            ZSTD_pthread_cond_signal(&ctx->queuePushCond);
+            ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
+        }
+    }  /* for (;;) */
+    assert(0);  /* Unreachable */
+}
+
+/* ZSTD_createThreadPool() : public access point */
+POOL_ctx* ZSTD_createThreadPool(size_t numThreads) {
+    return POOL_create (numThreads, 0);
+}
+
+POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) {
+    return POOL_create_advanced(numThreads, queueSize, ZSTD_defaultCMem);
+}
+
+POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
+                               ZSTD_customMem customMem)
+{
+    POOL_ctx* ctx;
+    /* Check parameters */
+    if (!numThreads) { return NULL; }
+    /* Allocate the context and zero initialize */
+    ctx = (POOL_ctx*)ZSTD_customCalloc(sizeof(POOL_ctx), customMem);
+    if (!ctx) { return NULL; }
+    /* Initialize the job queue.
+     * It needs one extra space since one space is wasted to differentiate
+     * empty and full queues.
+     */
+    ctx->queueSize = queueSize + 1;
+    ctx->queue = (POOL_job*)ZSTD_customCalloc(ctx->queueSize * sizeof(POOL_job), customMem);
+    ctx->queueHead = 0;
+    ctx->queueTail = 0;
+    ctx->numThreadsBusy = 0;
+    ctx->queueEmpty = 1;
+    {
+        int error = 0;
+        error |= ZSTD_pthread_mutex_init(&ctx->queueMutex, NULL);
+        error |= ZSTD_pthread_cond_init(&ctx->queuePushCond, NULL);
+        error |= ZSTD_pthread_cond_init(&ctx->queuePopCond, NULL);
+        if (error) { POOL_free(ctx); return NULL; }
+    }
+    ctx->shutdown = 0;
+    /* Allocate space for the thread handles */
+    ctx->threads = (ZSTD_pthread_t*)ZSTD_customCalloc(numThreads * sizeof(ZSTD_pthread_t), customMem);
+    ctx->threadCapacity = 0;
+    ctx->customMem = customMem;
+    /* Check for errors */
+    if (!ctx->threads || !ctx->queue) { POOL_free(ctx); return NULL; }
+    /* Initialize the threads */
+    {   size_t i;
+        for (i = 0; i < numThreads; ++i) {
+            if (ZSTD_pthread_create(&ctx->threads[i], NULL, &POOL_thread, ctx)) {
+                ctx->threadCapacity = i;
+                POOL_free(ctx);
+                return NULL;
+        }   }
+        ctx->threadCapacity = numThreads;
+        ctx->threadLimit = numThreads;
+    }
+    return ctx;
+}
+
+/*! POOL_join() :
+    Shutdown the queue, wake any sleeping threads, and join all of the threads.
+*/
+static void POOL_join(POOL_ctx* ctx) {
+    /* Shut down the queue */
+    ZSTD_pthread_mutex_lock(&ctx->queueMutex);
+    ctx->shutdown = 1;
+    ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
+    /* Wake up sleeping threads */
+    ZSTD_pthread_cond_broadcast(&ctx->queuePushCond);
+    ZSTD_pthread_cond_broadcast(&ctx->queuePopCond);
+    /* Join all of the threads */
+    {   size_t i;
+        for (i = 0; i < ctx->threadCapacity; ++i) {
+            ZSTD_pthread_join(ctx->threads[i]);  /* note : could fail */
+    }   }
+}
+
+void POOL_free(POOL_ctx *ctx) {
+    if (!ctx) { return; }
+    POOL_join(ctx);
+    ZSTD_pthread_mutex_destroy(&ctx->queueMutex);
+    ZSTD_pthread_cond_destroy(&ctx->queuePushCond);
+    ZSTD_pthread_cond_destroy(&ctx->queuePopCond);
+    ZSTD_customFree(ctx->queue, ctx->customMem);
+    ZSTD_customFree(ctx->threads, ctx->customMem);
+    ZSTD_customFree(ctx, ctx->customMem);
+}
+
+/*! POOL_joinJobs() :
+ *  Waits for all queued jobs to finish executing.
+ */
+void POOL_joinJobs(POOL_ctx* ctx) {
+    ZSTD_pthread_mutex_lock(&ctx->queueMutex);
+    while(!ctx->queueEmpty || ctx->numThreadsBusy > 0) {
+        ZSTD_pthread_cond_wait(&ctx->queuePushCond, &ctx->queueMutex);
+    }
+    ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
+}
+
+void ZSTD_freeThreadPool (ZSTD_threadPool* pool) {
+  POOL_free (pool);
+}
+
+size_t POOL_sizeof(const POOL_ctx* ctx) {
+    if (ctx==NULL) return 0;  /* supports sizeof NULL */
+    return sizeof(*ctx)
+        + ctx->queueSize * sizeof(POOL_job)
+        + ctx->threadCapacity * sizeof(ZSTD_pthread_t);
+}
+
+
+/* @return : 0 on success, 1 on error */
+static int POOL_resize_internal(POOL_ctx* ctx, size_t numThreads)
+{
+    if (numThreads <= ctx->threadCapacity) {
+        if (!numThreads) return 1;
+        ctx->threadLimit = numThreads;
+        return 0;
+    }
+    /* numThreads > threadCapacity */
+    {   ZSTD_pthread_t* const threadPool = (ZSTD_pthread_t*)ZSTD_customCalloc(numThreads * sizeof(ZSTD_pthread_t), ctx->customMem);
+        if (!threadPool) return 1;
+        /* replace existing thread pool */
+        ZSTD_memcpy(threadPool, ctx->threads, ctx->threadCapacity * sizeof(*threadPool));
+        ZSTD_customFree(ctx->threads, ctx->customMem);
+        ctx->threads = threadPool;
+        /* Initialize additional threads */
+        {   size_t threadId;
+            for (threadId = ctx->threadCapacity; threadId < numThreads; ++threadId) {
+                if (ZSTD_pthread_create(&threadPool[threadId], NULL, &POOL_thread, ctx)) {
+                    ctx->threadCapacity = threadId;
+                    return 1;
+            }   }
+    }   }
+    /* successfully expanded */
+    ctx->threadCapacity = numThreads;
+    ctx->threadLimit = numThreads;
+    return 0;
+}
+
+/* @return : 0 on success, 1 on error */
+int POOL_resize(POOL_ctx* ctx, size_t numThreads)
+{
+    int result;
+    if (ctx==NULL) return 1;
+    ZSTD_pthread_mutex_lock(&ctx->queueMutex);
+    result = POOL_resize_internal(ctx, numThreads);
+    ZSTD_pthread_cond_broadcast(&ctx->queuePopCond);
+    ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
+    return result;
+}
+
+/**
+ * Returns 1 if the queue is full and 0 otherwise.
+ *
+ * When queueSize is 1 (pool was created with an intended queueSize of 0),
+ * then a queue is empty if there is a thread free _and_ no job is waiting.
+ */
+static int isQueueFull(POOL_ctx const* ctx) {
+    if (ctx->queueSize > 1) {
+        return ctx->queueHead == ((ctx->queueTail + 1) % ctx->queueSize);
+    } else {
+        return (ctx->numThreadsBusy == ctx->threadLimit) ||
+               !ctx->queueEmpty;
+    }
+}
+
+
+static void
+POOL_add_internal(POOL_ctx* ctx, POOL_function function, void *opaque)
+{
+    POOL_job const job = {function, opaque};
+    assert(ctx != NULL);
+    if (ctx->shutdown) return;
+
+    ctx->queueEmpty = 0;
+    ctx->queue[ctx->queueTail] = job;
+    ctx->queueTail = (ctx->queueTail + 1) % ctx->queueSize;
+    ZSTD_pthread_cond_signal(&ctx->queuePopCond);
+}
+
+void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque)
+{
+    assert(ctx != NULL);
+    ZSTD_pthread_mutex_lock(&ctx->queueMutex);
+    /* Wait until there is space in the queue for the new job */
+    while (isQueueFull(ctx) && (!ctx->shutdown)) {
+        ZSTD_pthread_cond_wait(&ctx->queuePushCond, &ctx->queueMutex);
+    }
+    POOL_add_internal(ctx, function, opaque);
+    ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
+}
+
+
+int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque)
+{
+    assert(ctx != NULL);
+    ZSTD_pthread_mutex_lock(&ctx->queueMutex);
+    if (isQueueFull(ctx)) {
+        ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
+        return 0;
+    }
+    POOL_add_internal(ctx, function, opaque);
+    ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
+    return 1;
+}
+
+
+#else  /* ZSTD_MULTITHREAD  not defined */
+
+/* ========================== */
+/* No multi-threading support */
+/* ========================== */
+
+
+/* We don't need any data, but if it is empty, malloc() might return NULL. */
+struct POOL_ctx_s {
+    int dummy;
+};
+static POOL_ctx g_poolCtx;
+
+POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) {
+    return POOL_create_advanced(numThreads, queueSize, ZSTD_defaultCMem);
+}
+
+POOL_ctx*
+POOL_create_advanced(size_t numThreads, size_t queueSize, ZSTD_customMem customMem)
+{
+    (void)numThreads;
+    (void)queueSize;
+    (void)customMem;
+    return &g_poolCtx;
+}
+
+void POOL_free(POOL_ctx* ctx) {
+    assert(!ctx || ctx == &g_poolCtx);
+    (void)ctx;
+}
+
+void POOL_joinJobs(POOL_ctx* ctx){
+    assert(!ctx || ctx == &g_poolCtx);
+    (void)ctx;
+}
+
+int POOL_resize(POOL_ctx* ctx, size_t numThreads) {
+    (void)ctx; (void)numThreads;
+    return 0;
+}
+
+void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque) {
+    (void)ctx;
+    function(opaque);
+}
+
+int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque) {
+    (void)ctx;
+    function(opaque);
+    return 1;
+}
+
+size_t POOL_sizeof(const POOL_ctx* ctx) {
+    if (ctx==NULL) return 0;  /* supports sizeof NULL */
+    assert(ctx == &g_poolCtx);
+    return sizeof(*ctx);
+}
+
+#endif  /* ZSTD_MULTITHREAD */

external/zstd/common/pool.h

@@ -0,0 +1,90 @@
+/*
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * All rights reserved.
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+ */
+
+#ifndef POOL_H
+#define POOL_H
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
+
+#include "zstd_deps.h"
+#define ZSTD_STATIC_LINKING_ONLY   /* ZSTD_customMem */
+#include "../zstd.h"
+
+typedef struct POOL_ctx_s POOL_ctx;
+
+/*! POOL_create() :
+ *  Create a thread pool with at most `numThreads` threads.
+ * `numThreads` must be at least 1.
+ *  The maximum number of queued jobs before blocking is `queueSize`.
+ * @return : POOL_ctx pointer on success, else NULL.
+*/
+POOL_ctx* POOL_create(size_t numThreads, size_t queueSize);
+
+POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
+                               ZSTD_customMem customMem);
+
+/*! POOL_free() :
+ *  Free a thread pool returned by POOL_create().
+ */
+void POOL_free(POOL_ctx* ctx);
+
+
+/*! POOL_joinJobs() :
+ *  Waits for all queued jobs to finish executing.
+ */
+void POOL_joinJobs(POOL_ctx* ctx);
+
+/*! POOL_resize() :
+ *  Expands or shrinks pool's number of threads.
+ *  This is more efficient than releasing + creating a new context,
+ *  since it tries to preserve and re-use existing threads.
+ * `numThreads` must be at least 1.
+ * @return : 0 when resize was successful,
+ *           !0 (typically 1) if there is an error.
+ *    note : only numThreads can be resized, queueSize remains unchanged.
+ */
+int POOL_resize(POOL_ctx* ctx, size_t numThreads);
+
+/*! POOL_sizeof() :
+ * @return threadpool memory usage
+ *  note : compatible with NULL (returns 0 in this case)
+ */
+size_t POOL_sizeof(const POOL_ctx* ctx);
+
+/*! POOL_function :
+ *  The function type that can be added to a thread pool.
+ */
+typedef void (*POOL_function)(void*);
+
+/*! POOL_add() :
+ *  Add the job `function(opaque)` to the thread pool. `ctx` must be valid.
+ *  Possibly blocks until there is room in the queue.
+ *  Note : The function may be executed asynchronously,
+ *         therefore, `opaque` must live until function has been completed.
+ */
+void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque);
+
+
+/*! POOL_tryAdd() :
+ *  Add the job `function(opaque)` to thread pool _if_ a queue slot is available.
+ *  Returns immediately even if not (does not block).
+ * @return : 1 if successful, 0 if not.
+ */
+int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque);
+
+
+#if defined (__cplusplus)
+}
+#endif
+
+#endif

external/zstd/common/portability_macros.h

@@ -0,0 +1,148 @@
+/*
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * All rights reserved.
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+ */
+
+#ifndef ZSTD_PORTABILITY_MACROS_H
+#define ZSTD_PORTABILITY_MACROS_H
+
+/**
+ * This header file contains macro definitions to support portability.
+ * This header is shared between C and ASM code, so it MUST only
+ * contain macro definitions. It MUST not contain any C code.
+ *
+ * This header ONLY defines macros to detect platforms/feature support.
+ *
+ */
+
+
+/* compat. with non-clang compilers */
+#ifndef __has_attribute
+  #define __has_attribute(x) 0
+#endif
+
+/* compat. with non-clang compilers */
+#ifndef __has_builtin
+#  define __has_builtin(x) 0
+#endif
+
+/* compat. with non-clang compilers */
+#ifndef __has_feature
+#  define __has_feature(x) 0
+#endif
+
+/* detects whether we are being compiled under msan */
+#ifndef ZSTD_MEMORY_SANITIZER
+#  if __has_feature(memory_sanitizer)
+#    define ZSTD_MEMORY_SANITIZER 1
+#  else
+#    define ZSTD_MEMORY_SANITIZER 0
+#  endif
+#endif
+
+/* detects whether we are being compiled under asan */
+#ifndef ZSTD_ADDRESS_SANITIZER
+#  if __has_feature(address_sanitizer)
+#    define ZSTD_ADDRESS_SANITIZER 1
+#  elif defined(__SANITIZE_ADDRESS__)
+#    define ZSTD_ADDRESS_SANITIZER 1
+#  else
+#    define ZSTD_ADDRESS_SANITIZER 0
+#  endif
+#endif
+
+/* detects whether we are being compiled under dfsan */
+#ifndef ZSTD_DATAFLOW_SANITIZER
+# if __has_feature(dataflow_sanitizer)
+#  define ZSTD_DATAFLOW_SANITIZER 1
+# else
+#  define ZSTD_DATAFLOW_SANITIZER 0
+# endif
+#endif
+
+/* Mark the internal assembly functions as hidden  */
+#ifdef __ELF__
+# define ZSTD_HIDE_ASM_FUNCTION(func) .hidden func
+#else
+# define ZSTD_HIDE_ASM_FUNCTION(func)
+#endif
+
+/* Enable runtime BMI2 dispatch based on the CPU.
+ * Enabled for clang & gcc >=4.8 on x86 when BMI2 isn't enabled by default.
+ */
+#ifndef DYNAMIC_BMI2
+  #if ((defined(__clang__) && __has_attribute(__target__)) \
+      || (defined(__GNUC__) \
+          && (__GNUC__ >= 5 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 8)))) \
+      && (defined(__x86_64__) || defined(_M_X64)) \
+      && !defined(__BMI2__)
+  #  define DYNAMIC_BMI2 1
+  #else
+  #  define DYNAMIC_BMI2 0
+  #endif
+#endif
+
+/**
+ * Only enable assembly for GNUC compatible compilers,
+ * because other platforms may not support GAS assembly syntax.
+ *
+ * Only enable assembly for Linux / MacOS, other platforms may
+ * work, but they haven't been tested. This could likely be
+ * extended to BSD systems.
+ *
+ * Disable assembly when MSAN is enabled, because MSAN requires
+ * 100% of code to be instrumented to work.
+ */
+#if defined(__GNUC__)
+#  if defined(__linux__) || defined(__linux) || defined(__APPLE__)
+#    if ZSTD_MEMORY_SANITIZER
+#      define ZSTD_ASM_SUPPORTED 0
+#    elif ZSTD_DATAFLOW_SANITIZER
+#      define ZSTD_ASM_SUPPORTED 0
+#    else
+#      define ZSTD_ASM_SUPPORTED 1
+#    endif
+#  else
+#    define ZSTD_ASM_SUPPORTED 0
+#  endif
+#else
+#  define ZSTD_ASM_SUPPORTED 0
+#endif
+
+/**
+ * Determines whether we should enable assembly for x86-64
+ * with BMI2.
+ *
+ * Enable if all of the following conditions hold:
+ * - ASM hasn't been explicitly disabled by defining ZSTD_DISABLE_ASM
+ * - Assembly is supported
+ * - We are compiling for x86-64 and either:
+ *   - DYNAMIC_BMI2 is enabled
+ *   - BMI2 is supported at compile time
+ */
+#if !defined(ZSTD_DISABLE_ASM) &&                                 \
+    ZSTD_ASM_SUPPORTED &&                                         \
+    defined(__x86_64__) &&                                        \
+    (DYNAMIC_BMI2 || defined(__BMI2__))
+# define ZSTD_ENABLE_ASM_X86_64_BMI2 1
+#else
+# define ZSTD_ENABLE_ASM_X86_64_BMI2 0
+#endif
+
+/*
+ * For x86 ELF targets, add .note.gnu.property section for Intel CET in
+ * assembly sources when CET is enabled.
+ */
+#if defined(__ELF__) && (defined(__x86_64__) || defined(__i386__)) \
+    && defined(__has_include)
+# if __has_include(<cet.h>)
+#  include <cet.h>
+# endif
+#endif
+
+#endif /* ZSTD_PORTABILITY_MACROS_H */

external/zstd/common/threading.c

@@ -0,0 +1,176 @@
+/**
+ * Copyright (c) 2016 Tino Reichardt
+ * All rights reserved.
+ *
+ * You can contact the author at:
+ * - zstdmt source repository: https://github.com/mcmilk/zstdmt
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+ */
+
+/**
+ * This file will hold wrapper for systems, which do not support pthreads
+ */
+
+#include "threading.h"
+
+/* create fake symbol to avoid empty translation unit warning */
+int g_ZSTD_threading_useless_symbol;
+
+#if defined(ZSTD_MULTITHREAD) && defined(_WIN32)
+
+/**
+ * Windows minimalist Pthread Wrapper
+ */
+
+
+/* ===  Dependencies  === */
+#include <process.h>
+#include <errno.h>
+
+
+/* ===  Implementation  === */
+
+typedef struct {
+    void* (*start_routine)(void*);
+    void* arg;
+    int initialized;
+    ZSTD_pthread_cond_t initialized_cond;
+    ZSTD_pthread_mutex_t initialized_mutex;
+} ZSTD_thread_params_t;
+
+static unsigned __stdcall worker(void *arg)
+{
+    void* (*start_routine)(void*);
+    void* thread_arg;
+
+    /* Inialized thread_arg and start_routine and signal main thread that we don't need it
+     * to wait any longer.
+     */
+    {
+        ZSTD_thread_params_t*  thread_param = (ZSTD_thread_params_t*)arg;
+        thread_arg = thread_param->arg;
+        start_routine = thread_param->start_routine;
+
+        /* Signal main thread that we are running and do not depend on its memory anymore */
+        ZSTD_pthread_mutex_lock(&thread_param->initialized_mutex);
+        thread_param->initialized = 1;
+        ZSTD_pthread_cond_signal(&thread_param->initialized_cond);
+        ZSTD_pthread_mutex_unlock(&thread_param->initialized_mutex);
+    }
+
+    start_routine(thread_arg);
+
+    return 0;
+}
+
+int ZSTD_pthread_create(ZSTD_pthread_t* thread, const void* unused,
+            void* (*start_routine) (void*), void* arg)
+{
+    ZSTD_thread_params_t thread_param;
+    (void)unused;
+
+    thread_param.start_routine = start_routine;
+    thread_param.arg = arg;
+    thread_param.initialized = 0;
+    *thread = NULL;
+
+    /* Setup thread initialization synchronization */
+    if(ZSTD_pthread_cond_init(&thread_param.initialized_cond, NULL)) {
+        /* Should never happen on Windows */
+        return -1;
+    }
+    if(ZSTD_pthread_mutex_init(&thread_param.initialized_mutex, NULL)) {
+        /* Should never happen on Windows */
+        ZSTD_pthread_cond_destroy(&thread_param.initialized_cond);
+        return -1;
+    }
+
+    /* Spawn thread */
+    *thread = (HANDLE)_beginthreadex(NULL, 0, worker, &thread_param, 0, NULL);
+    if (!thread) {
+        ZSTD_pthread_mutex_destroy(&thread_param.initialized_mutex);
+        ZSTD_pthread_cond_destroy(&thread_param.initialized_cond);
+        return errno;
+    }
+
+    /* Wait for thread to be initialized */
+    ZSTD_pthread_mutex_lock(&thread_param.initialized_mutex);
+    while(!thread_param.initialized) {
+        ZSTD_pthread_cond_wait(&thread_param.initialized_cond, &thread_param.initialized_mutex);
+    }
+    ZSTD_pthread_mutex_unlock(&thread_param.initialized_mutex);
+    ZSTD_pthread_mutex_destroy(&thread_param.initialized_mutex);
+    ZSTD_pthread_cond_destroy(&thread_param.initialized_cond);
+
+    return 0;
+}
+
+int ZSTD_pthread_join(ZSTD_pthread_t thread)
+{
+    DWORD result;
+
+    if (!thread) return 0;
+
+    result = WaitForSingleObject(thread, INFINITE);
+    CloseHandle(thread);
+
+    switch (result) {
+    case WAIT_OBJECT_0:
+        return 0;
+    case WAIT_ABANDONED:
+        return EINVAL;
+    default:
+        return GetLastError();
+    }
+}
+
+#endif   /* ZSTD_MULTITHREAD */
+
+#if defined(ZSTD_MULTITHREAD) && DEBUGLEVEL >= 1 && !defined(_WIN32)
+
+#define ZSTD_DEPS_NEED_MALLOC
+#include "zstd_deps.h"
+
+int ZSTD_pthread_mutex_init(ZSTD_pthread_mutex_t* mutex, pthread_mutexattr_t const* attr)
+{
+    *mutex = (pthread_mutex_t*)ZSTD_malloc(sizeof(pthread_mutex_t));
+    if (!*mutex)
+        return 1;
+    return pthread_mutex_init(*mutex, attr);
+}
+
+int ZSTD_pthread_mutex_destroy(ZSTD_pthread_mutex_t* mutex)
+{
+    if (!*mutex)
+        return 0;
+    {
+        int const ret = pthread_mutex_destroy(*mutex);
+        ZSTD_free(*mutex);
+        return ret;
+    }
+}
+
+int ZSTD_pthread_cond_init(ZSTD_pthread_cond_t* cond, pthread_condattr_t const* attr)
+{
+    *cond = (pthread_cond_t*)ZSTD_malloc(sizeof(pthread_cond_t));
+    if (!*cond)
+        return 1;
+    return pthread_cond_init(*cond, attr);
+}
+
+int ZSTD_pthread_cond_destroy(ZSTD_pthread_cond_t* cond)
+{
+    if (!*cond)
+        return 0;
+    {
+        int const ret = pthread_cond_destroy(*cond);
+        ZSTD_free(*cond);
+        return ret;
+    }
+}
+
+#endif

external/zstd/common/threading.h

@@ -0,0 +1,150 @@
+/**
+ * Copyright (c) 2016 Tino Reichardt
+ * All rights reserved.
+ *
+ * You can contact the author at:
+ * - zstdmt source repository: https://github.com/mcmilk/zstdmt
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+ */
+
+#ifndef THREADING_H_938743
+#define THREADING_H_938743
+
+#include "debug.h"
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
+#if defined(ZSTD_MULTITHREAD) && defined(_WIN32)
+
+/**
+ * Windows minimalist Pthread Wrapper
+ */
+#ifdef WINVER
+#  undef WINVER
+#endif
+#define WINVER       0x0600
+
+#ifdef _WIN32_WINNT
+#  undef _WIN32_WINNT
+#endif
+#define _WIN32_WINNT 0x0600
+
+#ifndef WIN32_LEAN_AND_MEAN
+#  define WIN32_LEAN_AND_MEAN
+#endif
+
+#undef ERROR   /* reported already defined on VS 2015 (Rich Geldreich) */
+#include <windows.h>
+#undef ERROR
+#define ERROR(name) ZSTD_ERROR(name)
+
+
+/* mutex */
+#define ZSTD_pthread_mutex_t           CRITICAL_SECTION
+#define ZSTD_pthread_mutex_init(a, b)  ((void)(b), InitializeCriticalSection((a)), 0)
+#define ZSTD_pthread_mutex_destroy(a)  DeleteCriticalSection((a))
+#define ZSTD_pthread_mutex_lock(a)     EnterCriticalSection((a))
+#define ZSTD_pthread_mutex_unlock(a)   LeaveCriticalSection((a))
+
+/* condition variable */
+#define ZSTD_pthread_cond_t             CONDITION_VARIABLE
+#define ZSTD_pthread_cond_init(a, b)    ((void)(b), InitializeConditionVariable((a)), 0)
+#define ZSTD_pthread_cond_destroy(a)    ((void)(a))
+#define ZSTD_pthread_cond_wait(a, b)    SleepConditionVariableCS((a), (b), INFINITE)
+#define ZSTD_pthread_cond_signal(a)     WakeConditionVariable((a))
+#define ZSTD_pthread_cond_broadcast(a)  WakeAllConditionVariable((a))
+
+/* ZSTD_pthread_create() and ZSTD_pthread_join() */
+typedef HANDLE ZSTD_pthread_t;
+
+int ZSTD_pthread_create(ZSTD_pthread_t* thread, const void* unused,
+                   void* (*start_routine) (void*), void* arg);
+
+int ZSTD_pthread_join(ZSTD_pthread_t thread);
+
+/**
+ * add here more wrappers as required
+ */
+
+
+#elif defined(ZSTD_MULTITHREAD)    /* posix assumed ; need a better detection method */
+/* ===   POSIX Systems   === */
+#  include <pthread.h>
+
+#if DEBUGLEVEL < 1
+
+#define ZSTD_pthread_mutex_t            pthread_mutex_t
+#define ZSTD_pthread_mutex_init(a, b)   pthread_mutex_init((a), (b))
+#define ZSTD_pthread_mutex_destroy(a)   pthread_mutex_destroy((a))
+#define ZSTD_pthread_mutex_lock(a)      pthread_mutex_lock((a))
+#define ZSTD_pthread_mutex_unlock(a)    pthread_mutex_unlock((a))
+
+#define ZSTD_pthread_cond_t             pthread_cond_t
+#define ZSTD_pthread_cond_init(a, b)    pthread_cond_init((a), (b))
+#define ZSTD_pthread_cond_destroy(a)    pthread_cond_destroy((a))
+#define ZSTD_pthread_cond_wait(a, b)    pthread_cond_wait((a), (b))
+#define ZSTD_pthread_cond_signal(a)     pthread_cond_signal((a))
+#define ZSTD_pthread_cond_broadcast(a)  pthread_cond_broadcast((a))
+
+#define ZSTD_pthread_t                  pthread_t
+#define ZSTD_pthread_create(a, b, c, d) pthread_create((a), (b), (c), (d))
+#define ZSTD_pthread_join(a)         pthread_join((a),NULL)
+
+#else /* DEBUGLEVEL >= 1 */
+
+/* Debug implementation of threading.
+ * In this implementation we use pointers for mutexes and condition variables.
+ * This way, if we forget to init/destroy them the program will crash or ASAN
+ * will report leaks.
+ */
+
+#define ZSTD_pthread_mutex_t            pthread_mutex_t*
+int ZSTD_pthread_mutex_init(ZSTD_pthread_mutex_t* mutex, pthread_mutexattr_t const* attr);
+int ZSTD_pthread_mutex_destroy(ZSTD_pthread_mutex_t* mutex);
+#define ZSTD_pthread_mutex_lock(a)      pthread_mutex_lock(*(a))
+#define ZSTD_pthread_mutex_unlock(a)    pthread_mutex_unlock(*(a))
+
+#define ZSTD_pthread_cond_t             pthread_cond_t*
+int ZSTD_pthread_cond_init(ZSTD_pthread_cond_t* cond, pthread_condattr_t const* attr);
+int ZSTD_pthread_cond_destroy(ZSTD_pthread_cond_t* cond);
+#define ZSTD_pthread_cond_wait(a, b)    pthread_cond_wait(*(a), *(b))
+#define ZSTD_pthread_cond_signal(a)     pthread_cond_signal(*(a))
+#define ZSTD_pthread_cond_broadcast(a)  pthread_cond_broadcast(*(a))
+
+#define ZSTD_pthread_t                  pthread_t
+#define ZSTD_pthread_create(a, b, c, d) pthread_create((a), (b), (c), (d))
+#define ZSTD_pthread_join(a)         pthread_join((a),NULL)
+
+#endif
+
+#else  /* ZSTD_MULTITHREAD not defined */
+/* No multithreading support */
+
+typedef int ZSTD_pthread_mutex_t;
+#define ZSTD_pthread_mutex_init(a, b)   ((void)(a), (void)(b), 0)
+#define ZSTD_pthread_mutex_destroy(a)   ((void)(a))
+#define ZSTD_pthread_mutex_lock(a)      ((void)(a))
+#define ZSTD_pthread_mutex_unlock(a)    ((void)(a))
+
+typedef int ZSTD_pthread_cond_t;
+#define ZSTD_pthread_cond_init(a, b)    ((void)(a), (void)(b), 0)
+#define ZSTD_pthread_cond_destroy(a)    ((void)(a))
+#define ZSTD_pthread_cond_wait(a, b)    ((void)(a), (void)(b))
+#define ZSTD_pthread_cond_signal(a)     ((void)(a))
+#define ZSTD_pthread_cond_broadcast(a)  ((void)(a))
+
+/* do not use ZSTD_pthread_t */
+
+#endif /* ZSTD_MULTITHREAD */
+
+#if defined (__cplusplus)
+}
+#endif
+
+#endif /* THREADING_H_938743 */

external/zstd/common/xxhash.c

@@ -0,0 +1,24 @@
+/*
+ *  xxHash - Fast Hash algorithm
+ *  Copyright (c) Meta Platforms, Inc. and affiliates.
+ *
+ *  You can contact the author at :
+ *  - xxHash homepage: https://cyan4973.github.io/xxHash/
+ *  - xxHash source repository : https://github.com/Cyan4973/xxHash
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+*/
+
+
+
+/*
+ * xxhash.c instantiates functions defined in xxhash.h
+ */
+
+#define XXH_STATIC_LINKING_ONLY   /* access advanced declarations */
+#define XXH_IMPLEMENTATION   /* access definitions */
+
+#include "xxhash.h"