# Build WAMR vmcore WAMR vmcore is the runtime library set that loads and runs Wasm modules. This guide walks you through building the WAMR vmcore. References: - [how to build iwasm](../product-mini/README.md): build for Linux, Windows, macOS, and more - [Blog: Introduction to WAMR running modes](https://bytecodealliance.github.io/wamr.dev/blog/introduction-to-wamr-running-modes/) ## building configurations Include the script `runtime_lib.cmake` from [build-scripts](../build-scripts) into your CMakeLists.txt to pull vmcore into your build. ```cmake # add this into your CMakeLists.txt include (${WAMR_ROOT_DIR}/build-scripts/runtime_lib.cmake) add_library(vmlib ${WAMR_RUNTIME_LIB_SOURCE}) ``` The `runtime_lib.cmake` script exposes variables that control WAMR runtime features. Set them in CMakeLists.txt or pass them on the cmake command line. ```cmake # Set flags in CMakeLists.txt set(WAMR_BUILD_AOT 1) set(WAMR_BUILD_JIT 0) set(WAMR_BUILD_LIBC_BUILTIN 1) set(WAMR_BUILD_LIBC_WASI 1) # Include the runtime lib script include (${WAMR_ROOT_DIR}/build-scripts/runtime_lib.cmake) add_library(vmlib ${WAMR_RUNTIME_LIB_SOURCE}) ``` ### All compilation flags | Compilation flags | Description | | -------------------------------------------------------------------------------------------------------- | ------------------------------------ | | [WAMR_APP_THREAD_STACK_SIZE_MAX](#set-maximum-app-thread-stack-size) | Maximum stack size for app threads | | [WAMR_BH_LOG](#host-defined-log) | Host defined logging | | [WAMR_BH_VPRINTF](#host-defined-vprintf) | Host defined vprintf | | [WAMR_BUILD_ALLOC_WITH_USAGE](#user-defined-linear-memory-allocator) | Allocation with usage tracking | | [WAMR_BUILD_ALLOC_WITH_USER_DATA](#user-defined-linear-memory-allocator) | Allocation with user data | | [WAMR_BUILD_AOT](#configure-aot) | AoT compilation(wamrc) | | [WAMR_BUILD_AOT](#configure-aot) | AoT runtime | | [WAMR_BUILD_AOT_INTRINSICS](#aot-intrinsics) | AoT intrinsics | | [WAMR_BUILD_AOT_STACK_FRAME](#aot-stack-frame-feature) | AoT stack frame | | [WAMR_BUILD_AOT_VALIDATOR](#aot-validator) | AoT validator | | [WAMR_BUILD_BULK_MEMORY](#bulk-memory-feature) | bulk memory | | [WAMR_BUILD_COPY_CALL_STACK](#copy-call-stack) | copy call stack | | [WAMR_BUILD_CUSTOM_NAME_SECTION](#name-section) | name section | | [WAMR_BUILD_DEBUG_AOT](#source-debugging-features) | debug AoT | | [WAMR_BUILD_DEBUG_INTERP](#source-debugging-features) | debug interpreter | | [WAMR_BUILD_DUMP_CALL_STACK](#dump-call-stack-feature) | dump call stack | | [WAMR_BUILD_DYNAMIC_AOT_DEBUG](#source-debugging-features) | dynamic AoT debugging | | [WAMR_BUILD_EXCE_HANDLING](#exception-handling) | exception handling | | [WAMR_BUILD_EXTENDED_CONST_EXPR](#extended-constant-expression) | extended constant expressions | | [WAMR_BUILD_FAST_INTERP](#configure-interpreters) | fast interpreter | | [WAMR_BUILD_FAST_JIT](#configure-fast-jit) | fast JIT | | [WAMR_BUILD_FAST_JIT_DUMP](#configure-fast-jit) | fast JIT dump | | [WAMR_BUILD_GC](#garbage-collection) | garbage collection | | [WAMR_BUILD_GC_HEAP_VERIFY](#garbage-collection) | garbage collection heap verification | | [WAMR_BUILD_GC_HEAP_SIZE_DEFAULT](garbage-collection) | default garbage collection heap size | | [WAMR_BUILD_GLOBAL_HEAP_POOL](#a-pre-allocation-for-runtime-and-wasm-apps) | global heap pool | | [WAMR_BUILD_GLOBAL_HEAP_SIZE](#a-pre-allocation-for-runtime-and-wasm-apps) | global heap size | | [WAMR_BUILD_INSTRUCTION_METERING](#instruction-metering) | instruction metering | | [WAMR_BUILD_INTERP](#configure-interpreters) | interpreter | | [WAMR_BUILD_INVOKE_NATIVE_GENERAL](#invoke-general-ffi) | FFI general | | [WAMR_BUILD_JIT](#configure-llvm-jit) | JIT compilation | | [WAMR_BUILD_LAZY_JIT](#configure-llvm-jit) | lazy JIT compilation | | [WAMR_BUILD_LIBC_BUILTIN](#configure-libc) | libc builtin functions | | [WAMR_BUILD_LIBC_EMCC](#configure-libc) | libc emcc compatibility | | [WAMR_BUILD_LIBC_UVWASI](#configure-libc) | libc uvwasi compatibility | | [WAMR_BUILD_LIBC_WASI](#configure-libc) | wasi libc | | [WAMR_BUILD_LIB_PTHREAD](#lib-pthread) | pthread library | | [WAMR_BUILD_LIB_PTHREAD_SEMAPHORE](#lib-pthread-semaphore) | pthread semaphore support | | [WAMR_BUILD_LIB_RATS](#librats) | RATS library | | [WAMR_BUILD_LIB_WASI_THREADS](#lib-wasi-threads) | wasi threads | | [WAMR_BUILD_LINUX_PERF](#linux-perf-support) | Linux performance counters | | [WAMR_BUILD_LIME1](#lime1-target) | LIME1 runtime | | [WAMR_BUILD_LOAD_CUSTOM_SECTION](#load-wasm-custom-sections) | loading custom sections | | [WAMR_BUILD_MEMORY64](#memory64-feature) | memory64 support | | [WAMR_BUILD_MEMORY_PROFILING](#memory-profiling-experiment) | memory profiling | | [WAMR_BUILD_MINI_LOADER](#wasm-mini-loader) :warning: :exclamation: | mini loader | | [WAMR_BUILD_MODULE_INST_CONTEXT](#module-instance-context-apis) | module instance context | | [WAMR_BUILD_MULTI_MEMORY](#multi-memory) | multi-memory support | | [WAMR_BUILD_MULTI_MODULE](#multi-module-feature) | multi-module support | | [WAMR_BUILD_PERF_PROFILING](#performance-profiling-experiment) | performance profiling | | [WAMR_BUILD_PLATFORM](#configure-platform-and-architecture) | Default platform | | [WAMR_BUILD_QUICK_AOT_ENTRY](#quick-aotjti-entries) | quick AOT entry | | [WAMR_BUILD_REF_TYPES](#reference-types-feature) | reference types | | [WAMR_BUILD_SANITIZER](#sanitizer) | sanitizer | | [WAMR_BUILD_SGX_IPFS](#intel-protected-file-system) | Intel Protected File System support | | [WAMR_BUILD_SHARED_HEAP](#shared-heap-among-wasm-apps-and-host-native) | shared heap | | [WAMR_BUILD_SHARED_MEMORY](shared-memory-feature) | shared memory | | [WAMR_BUILD_SHRUNK_MEMORY](#shrunk-the-memory-usage) | shrunk memory | | [WAMR_BUILD_SIMD](#128-bit-simd-feature) | SIMD support | | [WAMR_BUILD_SIMDE](#128-bit-simd-feature) | SIMD E extensions | | [WAMR_BUILD_SPEC_TEST](#support-spec-test) | spec test | | [WAMR_BUILD_STACK_GUARD_SIZE](#stack-guard-size) | Stack guard size | | [WAMR_BUILD_STATIC_PGO](running-pgoprofile-guided-optimization-instrumented-aot-file) | Static PGO | | [WAMR_BUILD_STRINGREF](#garbage-collection) | String reference support | | [WAMR_BUILD_TAIL_CALL](#tail-call-feature) | Tail call optimization | | [WAMR_BUILD_TARGET](#configure-platform-and-architecture) | Default target architecture | | [WAMR_BUILD_THREAD_MGR](#thread-manager) | Thread manager | | [WAMR_BUILD_WAMR_COMPILER](#configure-aot) | WAMR compiler | | [WAMR_BUILD_WASI_EPHEMERAL_NN](#lib-wasi-nn-with-wasi_ephemeral_nn-module-support) | WASI ephemeral NN | | [WAMR_BUILD_WASI_NN](#lib-wasi-nn) | WASI NN | | [WAMR_BUILD_WASI_NN_EXTERNAL_DELEGATE_PATH](#lib-wasi-nn-external-delegate-mode) | External delegate path for WASI NN | | [WAMR_BUILD_WASI_NN_ENABLE_GPU](#lib-wasi-nn-gpu-mode) | GPU support for WASI NN | | [WAMR_BUILD_WASI_NN_LLAMACPP](#lib-wasi-nn) | LLAMA CPP for WASI NN | | [WAMR_BUILD_WASI_NN_ONNX](#lib-wasi-nn) | ONNX for WASI NN | | [WAMR_BUILD_WASI_NN_OPENVINO](#lib-wasi-nn) | OpenVINO for WASI NN | | [WAMR_BUILD_WASI_NN_TFLITE](#lib-wasi-nn) | TFLite for WASI NN | | [WAMR_BUILD_WASM_CACHE](#wasm-cache) | WASM cache | | [WAMR_CONFIGURABLE_BOUNDS_CHECKS](#configurable-memory-access-boundary-check) :warning: :exclamation: | Configurable bounds checks | | [WAMR_DISABLE_APP_ENTRY](#exclude-wamr-application-entry-functions) | Disable app entry | | [WAMR_DISABLE_HW_BOUND_CHECK](#disable-boundary-check-with-hardware-trap) | Disable hardware bound check | | [WAMR_DISABLE_STACK_HW_BOUND_CHECK](#disable-native-stack-boundary-check-with-hardware-trap) | Disable stack hardware bound check | | [WAMR_DISABLE_WAKEUP_BLOCKING_OP](#disable-async-wakeup-of-blocking-operation) | Disable wakeup blocking operation | | [WAMR_DISABLE_WRITE_GS_BASE](#disable-writing-the-linear-memory-base-address-to-x86-gs-segment-register) | Disable write GS base | | [WAMR_TEST_GC](#test-garbage-collection) | Test garbage collection | ### **Privileged Features** :warning: :exclamation: _Privileged Features_ are powerful options that can boost performance or add capabilities but lower security by compromising isolation. Use them with care and test thoroughly. ### **[config.h](../core/config.h)** Above compilation flags map to macros in `config.h`. For example, `WAMR_BUILD_AOT` maps to `WAMR_BUILD_AOT` in `config.h`. The build system sets these macros automatically based on your CMake settings. If your build doesn't set those flags, default values in `config.h` apply. ### **Configure platform and architecture** - **WAMR_BUILD_PLATFORM**: set the target platform. Match the platform folder name under [core/shared/platform](../core/shared/platform). - **WAMR_BUILD_TARGET**: set the target CPU architecture. Supported targets: X86_64, X86_32, AARCH64, ARM, THUMB, XTENSA, ARC, RISCV32, RISCV64, and MIPS. - For ARM and THUMB, use `[][_VFP]`. `` is the ARM sub-architecture. `_VFP` means arguments and returns use VFP coprocessor registers s0-s15 (d0-d7). Both are optional, for example ARMV7, ARMV7_VFP, THUMBV7, or THUMBV7_VFP. - For AARCH64, use `[]`. VFP is on by default. `` is optional, for example AARCH64, AARCH64V8, or AARCH64V8.1. - For RISCV64, use `[_abi]`. `_abi` is optional. Supported: RISCV64, RISCV64_LP64D, and RISCV64_LP64. RISCV64 and RISCV64_LP64D both use [LP64D](https://github.com/riscv-non-isa/riscv-elf-psabi-doc/blob/master/riscv-cc.adoc#named-abis) (LP64 with hardware floating-point for FLEN=64). RISCV64_LP64 uses [LP64](https://github.com/riscv-non-isa/riscv-elf-psabi-doc/blob/master/riscv-cc.adoc#named-abis) (integer calling convention only; no hardware floating-point calling convention). - For RISCV32, use `[_abi]`. `_abi` is optional. Supported: RISCV32, RISCV32_ILP32D, RISCV32_ILP32F, and RISCV32_ILP32. RISCV32 and RISCV32_ILP32D both use [ILP32D](https://github.com/riscv-non-isa/riscv-elf-psabi-doc/blob/master/riscv-cc.adoc#named-abis) (ILP32 with hardware floating-point for FLEN=64). RISCV32_ILP32F uses [ILP32F](https://github.com/riscv-non-isa/riscv-elf-psabi-doc/blob/master/riscv-cc.adoc#named-abis) (ILP32 with hardware floating-point for FLEN=32). RISCV32_ILP32 uses [ILP32](https://github.com/riscv-non-isa/riscv-elf-psabi-doc/blob/master/riscv-cc.adoc#named-abis) (integer calling convention only). ```bash cmake -DWAMR_BUILD_PLATFORM=linux -DWAMR_BUILD_TARGET=ARM ``` ### **Configure interpreters** - **WAMR_BUILD_INTERP**=1/0: turn the WASM interpreter on or off. - **WAMR_BUILD_FAST_INTERP**=1/0: pick fast (default) or classic interpreter. > [!NOTE] > The fast interpreter runs ~2X faster than classic interpreter, but consumes about 2X memory to hold the pre-compiled code. ### **Configure AOT** - **WAMR_BUILD_AOT**=1/0: turn AOT on or off. Defaults to on. - **WAMR_BUILD_WAMR_COMPILER**=1/0. It is used to wasm loader and compilation to indictate compiler mode. ### **Configure LLVM JIT** Comparing with fast JIT, LLVM JIT covers more architectures and produces better optimized code, but takes longer on cold start. - **WAMR_BUILD_JIT**=1/0: turn LLVM JIT on or off. Defaults to off. - **WAMR_BUILD_LAZY_JIT**=1/0: turn lazy JIT on or off. Defaults to off. With lazy JIT, functions are compiled in background threads before they are called, which can reduce startup time for large modules. ### **Configure Fast JIT** The fast JIT is a lightweight JIT that emits code quickly and tunes hot functions. - **WAMR_BUILD_FAST_JIT**=1/0: turn Fast JIT on or off. Defaults to off. - **WAMR_BUILD_FAST_JIT_DUMP**=1/0: dump fast JIT compiled code to stdout for debugging. Defaults to off. > [!WARNING] > It currently covers only a few architectures (x86_64). ### **Configure Multi-tier JIT** Use fast jit as the first tier and LLVM JIT as the second tier. - With **WAMR_BUILD_FAST_JIT**=1 and **WAMR_BUILD_JIT**=1, you get multi-tier JIT. Defaults to off. > [!WARNING] > It currently covers only a few architectures (x86_64). ### **Configure LIBC** - **WAMR_BUILD_LIBC_BUILTIN**=1/0: build the built-in libc subset for WASM apps. Defaults to on. - **WAMR_BUILD_LIBC_WASI**=1/0: build the [WASI](https://github.com/WebAssembly/WASI) libc subset for WASM apps. Defaults to on. - **WAMR_BUILD_LIBC_UVWASI**=1/0 (Experiment): build the WASI libc subset for WASM apps using [uvwasi](https://github.com/nodejs/uvwasi). Defaults to off. - **WAMR_BUILD_LIBC_EMCC**=1/0: build the emcc-compatible libc subset for WASM apps. Defaults to off. > [!WARNING] > WAMR is not a secure sandbox on every platform. On platforms where **WAMR_BUILD_LIBC_WASI** is unsupported (for example Windows), you can try the uvwasi-based WASI via **WAMR_BUILD_LIBC_UVWASI**, but it is unsafe. ### **Multi-Module feature** - **WAMR_BUILD_MULTI_MODULE**=1/0, default to off. > [!NOTE] > See [Multiple Modules as Dependencies](./multi_module.md) for details. > [!WARNING] > The multi-module feature is not supported in fast-jit or llvm-jit modes. ### **WASM mini loader** - **WAMR_BUILD_MINI_LOADER**=1/0, default to off. > [!NOTE] > The mini loader skips integrity checks on the WASM binary. Make sure the file is valid yourself. > [!WARNING] > This is a [privileged feature](#privileged-features) that compromises security. Use it only when you trust the WASM binary source. ### **shared memory feature** - **WAMR_BUILD_SHARED_MEMORY**=1/0, default to off. ### **bulk memory feature** - **WAMR_BUILD_BULK_MEMORY**=1/0, default to off. ### **memory64 feature** - **WAMR_BUILD_MEMORY64**=1/0, default to off. > [!WARNING] > Supported only in classic interpreter mode and AOT mode. ### **thread manager** - **WAMR_BUILD_THREAD_MGR**=1/0, default to off. ### **lib-pthread** - **WAMR_BUILD_LIB_PTHREAD**=1/0, default to off. > [!NOTE] > When you enable lib pthread, required features such as `shared memory` and `thread manager` are enabled automatically. See [WAMR pthread library](./pthread_library.md) for details. ### **lib-pthread-semaphore** - **WAMR_BUILD_LIB_PTHREAD_SEMAPHORE**=1/0, default to off. > [!NOTE] > This depends on `lib-pthread` and turns it on automatically. ### **lib wasi-threads** - **WAMR_BUILD_LIB_WASI_THREADS**=1/0, default to off. > [!NOTE] > Enabling lib wasi-threads also enables its dependencies `shared memory` and `thread manager`. See [wasi-threads](./pthread_impls.md#wasi-threads-new) and [Introduction to WAMR WASI threads](https://bytecodealliance.github.io/wamr.dev/blog/introduction-to-wamr-wasi-threads) for details. ### **lib wasi-nn** - **WAMR_BUILD_WASI_NN**=1/0, default to off. - **WAMR_BUILD_WASI_NN_LLAMACPP**=1/0, default to off. - **WAMR_BUILD_WASI_NN_ONNX**=1/0, default to off. - **WAMR_BUILD_WASI_NN_OPENVINO**=1/0, default to off. - **WAMR_BUILD_WASI_NN_TFLITE**=1/0, default to off. > [!NOTE] > Using WAMR_BUILD_WASI_NN without WAMR_BUILD_WASI_EPHEMERAL_NN is deprecated and may be removed later. Please enable WAMR_BUILD_WASI_EPHEMERAL_NN too. See [WASI-NN](../core/iwasm/libraries/wasi-nn) for details. ### **lib wasi-nn GPU mode** - **WAMR_BUILD_WASI_NN_ENABLE_GPU**=1/0, default to off. ### **lib wasi-nn external delegate mode** - **WAMR_BUILD_WASI_NN_ENABLE_EXTERNAL_DELEGATE**=1/0, default to off. - **WAMR_BUILD_WASI_NN_EXTERNAL_DELEGATE_PATH**=Path to the external delegate shared library (for example `libedgetpu.so.1.0` for Coral USB). ### **lib wasi-nn with `wasi_ephemeral_nn` module support** - **WAMR_BUILD_WASI_EPHEMERAL_NN**=1/0, default to on. ### **Disable boundary check with hardware trap** - **WAMR_DISABLE_HW_BOUND_CHECK**=1/0, default to on if the platform supports it. Otherwise, software boundary checks are used. > [!NOTE] > By default only [linux/darwin/android/windows/vxworks 64-bit](https://github.com/bytecodealliance/wasm-micro-runtime/blob/5fb5119239220b0803e7045ca49b0a29fe65e70e/core/shared/platform/linux/platform_internal.h#L81) platforms enable this hardware trap boundary check. On 32-bit platforms it is off even if the flag is 0. The wamrc tool omits boundary check instructions in AOT code for all 64-bit targets except SGX to improve speed. The boundary check covers linear memory access and native stack access unless `WAMR_DISABLE_STACK_HW_BOUND_CHECK` is set. ### **Disable native stack boundary check with hardware trap** - **WAMR_DISABLE_STACK_HW_BOUND_CHECK**=1/0, default to on if the platform supports it; same rule as `WAMR_DISABLE_HW_BOUND_CHECK`. Otherwise, software boundary checks are used. > [!NOTE] > If hardware trap boundary checks are off (or `WAMR_DISABLE_HW_BOUND_CHECK` is 1), native stack boundary checks are also off regardless of `WAMR_DISABLE_STACK_HW_BOUND_CHECK`. If hardware trap boundary checks are on, this setting decides whether the native stack check is on. ### **Disable async wakeup of blocking operation** - **WAMR_DISABLE_WAKEUP_BLOCKING_OP**=1/0, default to on when the platform supports it. > [!NOTE] > This feature lets blocking threads terminate asynchronously. If you disable it, blocking threads may never finish when asked to exit. ### **tail call feature** - **WAMR_BUILD_TAIL_CALL**=1/0, default to off. ### **128-bit SIMD feature** - **WAMR_BUILD_SIMD**=1/0, default to on. - **WAMR_BUILD_SIMDE**=1/0, default to off. SIMDE (SIMD Everywhere) implements SIMD operations in fast interpreter mode. > [!WARNING] > Supported in AOT, JIT, and fast-interpreter modes with the SIMDe library. ### **SIMDe library for SIMD in fast interpreter** - **WAMR_BUILD_LIB_SIMDE**=1/0, default to off. > [!NOTE] > When enabled, SIMDe (SIMD Everywhere) implements SIMD operations in fast interpreter mode. ### **Exception Handling** - **WAMR_BUILD_EXCE_HANDLING**=1/0, default to off. > [!NOTE] > Current implementation supports only Legacy Wasm exception handling proposal, not the latest version. > [!WARNING] > Exception handling currently works only in classic interpreter mode. ### **Garbage Collection** - **WAMR_BUILD_GC**=1/0, default to off. - **WAMR_BUILD_GC_HEAP_VERIFY**=1/0, default to off. When enabled, verifies the heap during free. - **WAMR_BUILD_STRINGREF**=1/0, default to off. When enabled, need to set WAMR_STRINGREF_IMPL_SOURCE as well > [!WARNING] > Garbage collection is not supported in fast-jit mode and multi-tier-jit mode. ### **Set the Garbage Collection heap size** - **WAMR_BUILD_GC_HEAP_SIZE_DEFAULT**=n, default to 128 kB (131072). ### **Multi Memory** - **WAMR_BUIL_MULTI_MEMORY**=1/0, default to off. > [!WARNING] > Multi memory is supported only in classic interpreter mode. ### **Name Section** - **WAMR_BUILD_CUSTOM_NAME_SECTION**=1/0: load function names from the custom name section. Default is off. ### **AOT stack frame feature** - **WAMR_BUILD_AOT_STACK_FRAME**=1/0, default to off. > [!NOTE] > When enabled, AOT or JIT stack frames (similar to classic interpreter frames but storing only what is needed) are built during calls. Add `--enable-dump-call-stack` to wamrc when compiling AOT modules. ### **dump call stack feature** - **WAMR_BUILD_DUMP_CALL_STACK**=1/0, default to off. > [!NOTE] > When enabled, the runtime dumps the call stack on exceptions. > > - In interpreter mode, names come first from the custom name section. If that section is absent or disabled, names come from import/export sections. > - In AOT/JIT mode, names come from the import/export section. Export as many functions as possible (for `wasi-sdk` you can use `-Wl,--export-all`) when compiling the wasm module, and add `--enable-dump-call-stack --emit-custom-sections=name` to wamrc when compiling the AOT module. ### **memory profiling (Experiment)** - **WAMR_BUILD_MEMORY_PROFILING**=1/0, default to off. > [!NOTE] > When enabled, call `void wasm_runtime_dump_mem_consumption(wasm_exec_env_t exec_env)` to dump memory usage. Currently only module, module_instance, and exec_env memory are measured; other components such as `wasi-ctx`, `multi-module`, and `thread-manager` are not included. See [Memory usage estimation for a module](./memory_usage.md). ### **performance profiling (Experiment)** - **WAMR_BUILD_PERF_PROFILING**=1/0, default to off. > [!NOTE] > When enabled, call `void wasm_runtime_dump_perf_profiling(wasm_module_inst_t module_inst)` to dump per-function performance. Function name lookup follows the same order as the dump call stack feature. See [Tune the performance of running wasm/aot file](./perf_tune.md). ### **A pre-allocation for runtime and wasm apps** - **WAMR_BUILD_GLOBAL_HEAP_POOL**=1/0, default to off for _iwasm_ apps except on Alios and Zephyr. - **WAMR_BUILD_GLOBAL_HEAP_SIZE**=n, default to 10 MB (10485760) for _iwasm_ apps, except Alios (256 kB), Riot (256 kB), and Zephyr (128 kB). > [!NOTE] > When enabled, WAMR uses a big global heap for runtime and wasm apps instead of allocating memory from the system directly. This can reduce memory fragmentation and improve performance when many small allocations happen. The global heap is allocated at startup. **WAMR_BUILD_GLOBAL_HEAP_POOL** applies to _iwasm_ apps in `product-mini`. For your own host app, set `mem_alloc_type` to `Alloc_With_Pool` if you want to use a global heap. The global heap is described in [Memory model and memory usage tunning](memory_tune.md). **WAMR_BUILD_GLOBAL_HEAP_SIZE** applies to _iwasm_ apps in `product-mini`. For your host app, set `mem_alloc_option.pool` with the size you want for the global heap. The global heap is described in [Memory model and memory usage tunning](memory_tune.md). ### **Set maximum app thread stack size** - **WAMR_APP_THREAD_STACK_SIZE_MAX**=n, default to 8 MB (8388608). > [!NOTE] > AOT boundary checks with hardware traps may use large stacks because the OS can grow stacks lazily when a guard page is hit. Use this setting to cap total stack use, for example `-DWAMR_APP_THREAD_STACK_SIZE_MAX=131072` (128 KB). ### **Host defined vprintf** - **WAMR_BH_VPRINTF**=, default to off. > [!NOTE] > If you provide `vprintf_callback`, `os_printf()` and `os_vprintf()` on Linux, Darwin, Windows, VxWorks, Android, and esp-idf, plus WASI libc output, call your callback instead of libc `vprintf()`. Example outside the runtime lib: > > ```C > int my_vprintf(const char *format, va_list ap) > { > /* output to pre-opened file stream */ > FILE *my_file = ...; > return vfprintf(my_file, format, ap); > /* or output to pre-opened file descriptor */ > int my_fd = ...; > return vdprintf(my_fd, format, ap); > /* or output to string buffer and print the string */ > char buf[128]; > vsnprintf(buf, sizeof(buf), format, ap); > return my_printf("%s", buf); > } > ``` > > Then run `cmake -DWAMR_BH_VPRINTF=my_vprintf ..`, or add the compiler macro `BH_VPRINTF=my_vprintf` (for example `add_definitions(-DBH_VPRINTF=my_vprintf)` in CMakeLists.txt). See [basic sample](../samples/basic/src/main.c) for an example. ### **WAMR_BH_LOG**=, default to off. > [!NOTE] > If you provide `log_callback`, WAMR logs go there. Example: > > ```C > void my_log(uint32 log_level, const char *file, int line, const char *fmt, ...) > { > /* Usage of custom logger */ > } > ``` > > See [basic sample](../samples/basic/src/main.c) for an example. ### **reference types feature** - **WAMR_BUILD_REF_TYPES**=1/0, default to on. ### **Exclude WAMR application entry functions** - **WAMR_DISABLE_APP_ENTRY**=1/0, default to off. > [!NOTE] > The WAMR application entry (`core/iwasm/common/wasm_application.c`) wraps common steps to instantiate and run wasm functions and print results. These use platform APIs. this flag to skip the file if your platform lacks those APIs. _Do not enable this flag when building `product-mini`._ ### **source debugging features** - **WAMR_BUILD_DEBUG_INTERP**=1/0, default to off. - **WAMR_BUILD_DEBUG_AOT**=1/0, default to off. - **WAMR_BUILD_DYNAMIC_AOT_DEBUG**=1/0, default to off. > [!NOTE] > Source debugging needs extra setup. See [source_debugging.md](./source_debugging.md) and [WAMR source debugging basic](https://bytecodealliance.github.io/wamr.dev/blog/wamr-source-debugging-basic). ### **load wasm custom sections** - **WAMR_BUILD_LOAD_CUSTOM_SECTION**=1/0, default to off. > [!NOTE] > By default, custom sections are ignored. `WAMR_BUILD_LOAD_CUSTOM_SECTION` so the embedder can read them via `wasm_runtime_get_custom_section`. If `WAMR_BUILD_CUSTOM_NAME_SECTION` is on, the custom name section is consumed by the runtime and unavailable to the embedder. For AoT files, pass `--emit-custom-sections` to wamrc to keep the sections; otherwise they are dropped. ### **Stack guard size** - **WAMR_BUILD_STACK_GUARD_SIZE**=n, default to N/A when not set. > [!NOTE] > By default, stack guard size is 1K (1024) or 24K when uvwasi is enabled. ### **Disable writing the linear memory base address to x86 GS segment register** - **WAMR_DISABLE_WRITE_GS_BASE**=1/0, default to on if the platform supports it. > [!NOTE] > By default only [linux x86-64](https://github.com/bytecodealliance/wasm-micro-runtime/blob/5fb5119239220b0803e7045ca49b0a29fe65e70e/core/shared/platform/linux/platform_internal.h#L67) enables this. On 32-bit platforms it stays off even if set to 0. On linux x86-64, writing the linear memory base to the GS segment can speed up linear memory access for LLVM AOT/JIT when `--enable-segue=[]` is passed to `wamrc` or `iwasm`. > > See [segue optimization for wamrc when generating the aot file](./perf_tune.md#3-enable-segue-optimization-for-wamrc-when-generating-the-aot-file) for details. ### **User defined linear memory allocator** - **WAMR_BUILD_ALLOC_WITH_USAGE**=1/0, default to off. - **WAMR_BUILD_ALLOC_WITH_USER_DATA**=1/0, default to off. > [!NOTE] > By default, the system allocates linear memory. With this on and `Alloc_With_Allocator` selected, you can provide your own allocator. ### **running PGO(Profile-Guided Optimization) instrumented AOT file** - **WAMR_BUILD_STATIC_PGO**=1/0, default to off. > [!NOTE] > See [Use the AOT static PGO method](./perf_tune.md#5-use-the-aot-static-pgo-method). ### **linux perf support** - **WAMR_BUILD_LINUX_PERF**=1/0: enable linux perf support to generate flamegraphs for wasm app performance. Default is off. > [!NOTE] > See [Use linux-perf](./perf_tune.md#7-use-linux-perf). ### **module instance context APIs** - **WAMR_BUILD_MODULE_INST_CONTEXT**=1/0: enable module instance context APIs so the embedder can set one or more contexts for a wasm module instance. Default is on. ```C wasm_runtime_create_context_key wasm_runtime_destroy_context_key wasm_runtime_set_context wasm_runtime_set_context_spread wasm_runtime_get_context ``` > [!NOTE] > See [wasm_export.h](../core/iwasm/include/wasm_export.h) for details. ### **quick AOT/JTI entries** - **WAMR_BUILD_QUICK_AOT_ENTRY**=1/0: register quick call entries to speed up AOT/JIT function calls. Default is on. > [!NOTE] > See [Refine callings to AOT/JIT functions from host native](./perf_tune.md#83-refine-callings-to-aotjit-functions-from-host-native). ### **AOT intrinsics** - **WAMR_BUILD_AOT_INTRINSICS**=1/0: turn on AOT intrinsic functions. Default is on. AOT code can call these when wamrc uses `--disable-llvm-intrinsics` or `--enable-builtin-intrinsics=`. > [!NOTE] > See [Tuning the XIP intrinsic functions](./xip.md#tuning-the-xip-intrinsic-functions). ### **extended constant expression** - **WAMR_BUILD_EXTENDED_CONST_EXPR**=1/0, default to off. > [!NOTE] > See [Extended Constant Expressions](https://github.com/WebAssembly/extended-const/blob/main/proposals/extended-const/Overview.md). ### **bulk-memory-opt** - **WAMR_BUILD_BULK_MEMORY_OPT**=1/0, default to off. > [!NOTE] > See [bulk-memory-opt](https://github.com/WebAssembly/tool-conventions/blob/main/Lime.md#bulk-memory-opt). ### **call-indirect-overlong** - **WAMR_BUILD_CALL_INDIRECT_OVERLONG**=1/0, default to off. > [!NOTE] > See [call-indirect-overlong](https://github.com/WebAssembly/tool-conventions/blob/main/Lime.md#call-indirect-overlong). ### **Lime1 target** - **WAMR_BUILD_LIME1**=1/0, default to off. > [!NOTE] > See [Lime1](https://github.com/WebAssembly/tool-conventions/blob/main/Lime.md#lime1). ### **Configurable memory access boundary check** - **WAMR_CONFIGURABLE_BOUNDS_CHECKS**=1/0, default to off. > [!WARNING] > When enabled, you can run `iwasm --disable-bounds-checks` to turn off memory access boundary checks in interpreter mode. This is a [privileged feature](#privileged-features); use it carefully. ### **Shared heap among wasm apps and host native** - **WAMR_BUILD_SHARED_HEAP**=1/0, default to off. > [!NOTE] > When enabled, you can create and attach shared heaps, and the following APIs become available: > > ```C > wasm_runtime_create_shared_heap > wasm_runtime_attach_shared_heap > wasm_runtime_detach_shared_heap > wasm_runtime_shared_heap_malloc > wasm_runtime_shared_heap_free > ``` > > A wasm app can call these to use the shared heap attached to its module instance: > > ```C > void *shared_heap_malloc(); > void shared_heap_free(void *ptr); > ``` > [!WARNING] > The shared-heap feature is not supported in fast-jit mode. ### **Shrunk the memory usage** - **WAMR_BUILD_SHRUNK_MEMORY**=1/0, default to on. > [!NOTE] > When enabled, this reduces memory by shrinking linear memory, especially when `memory.grow` is unused and memory needs are predictable. ### **Instruction metering** - **WAMR_BUILD_INSTRUCTION_METERING**=1/0, default to off. > [!NOTE] > This limits the number of instructions a wasm module instance can run. Call `wasm_runtime_set_instruction_count_limit(...)` before `wasm_runtime_call_*(...)` to enforce the cap. > [!WARNING] > This is only supported in classic interpreter mode. ### **Invoke general FFI** - **WAMR_BUILD_INVOKE_NATIVE_GENERAL**=1/0, default to off. By default, WAMR uses architecture-specific calling conventions to call native functions from WASM modules. When this feature is enabled, WAMR uses a general calling convention that works on all architectures but is slower. The details are in [iwasm_common.cmake](../core/iwasm/common/iwasm_common.cmake) ### **Host defined log** - **WAMR_BH_LOG**=, default to off. ### **AOT Validator** - **WAMR_BUILD_AOT_VALIDATOR**=1/0, default to off. > [!NOTE] > By default, WAMR believes AOT files are valid and unforged. ### **Copy Call Stack** - **WAMR_BUILD_COPY_CALL_STACK**=1/0, default to off. > [!NOTE] > Unlike [dump call stack](dump-call-stack-feature), which prints the call stack on exceptions, this feature lets the embedder copy the call stack programmatically via `wasm_runtime_dump_call_stack_to_buf()`. ### **Librats** - **WAMR_BUILD_LIB_RATS**=1/0, default to off. > [librats](https://github.com/inclavare-containers/librats) is a C library designed to facilitate remote attestation for secure computing environments. It provides a framework for attesting the integrity of computing environments remotely, enabling trust establishment between different Trusted Execution Environments (TEEs). > [!WARNING] > This is for Intel SGX platforms only. ### **Sanitizer** - **WAMR_BUILD_SANITIZER**=[ubsan|asan|tsan|posan], default is empty Use one or more of the following sanitizers when building WAMR with sanitizer support: AddressSanitizer, UndefinedBehaviorSanitizer, ThreadSanitizer, or Pointer-Overflow Sanitizer. ### **Intel Protected File System** - **WAMR_BUILD_SGX_IPFS**=1/0, default to off. > [!WARNING] > This is for Intel SGX platforms only. ### **Support spec test** - **WAMR_BUILD_SPEC_TEST**=1/0, default to off. ### **WASM Cache** - **WAMR_BUILD_WASM_CACHE**=1/0, default to off. ### **Test Garbage Collection** - **WAMR_TEST_GC**=1/0, default to off. It is used to test garbage collection related APIs and features. Refer to [iwasm_gc.cmake](../core/iwasm/common/gc/iwasm_gc.cmake) for details. It is used to cache loaded wasm modules in memory to speed up module instantiation only in wasm-c-api. ## **Combination of configurations** You can mix settings. For example, to disable the interpreter, enable AOT and WASI, run: ```Bash cmake .. -DWAMR_BUILD_INTERP=0 -DWAMR_BUILD_AOT=1 -DWAMR_BUILD_LIBC_WASI=1 -DWAMR_BUILD_PLATFORM=linux ``` To enable the interpreter, disable AOT and WASI, and target X86_32, run: ```Bash cmake .. -DWAMR_BUILD_INTERP=1 -DWAMR_BUILD_AOT=0 -DWAMR_BUILD_LIBC_WASI=0 -DWAMR_BUILD_TARGET=X86_32 ``` When enabling SIMD for fast interpreter mode, turn on both SIMD and the SIMDe library: ```Bash cmake .. -DWAMR_BUILD_INTERP=1 -DWAMR_BUILD_FAST_INTERP=1 -DWAMR_BUILD_SIMD=1 -DWAMR_BUILD_LIB_SIMDE=1 ``` For Valgrind, start with these and add more as needed: ```Bash #... -DCMAKE_BUILD_TYPE=Debug \ -DWAMR_DISABLE_HW_BOUND_CHECK=0 \ -DWAMR_DISABLE_WRITE_GS_BASE=0 #... ``` To enable the minimal Lime1 feature set, turn off features that are on by default such as bulk memory and reference types: ```Bash cmake .. -DWAMR_BUILD_LIME1=1 -DWAMR_BUILD_BULK_MEMORY=0 -DWAMR_BUILD_REF_TYPES=0 -DDWAMR_BUILD_SIMD=0 ```