Shared heap enhancements for Interpreter and AOT (#4400)

Propose two enhancements:

- Shared heap created from preallocated memory buffer: The user can create a shared heap from a pre-allocated buffer and see that memory region as one large chunk; there's no need to dynamically manage it(malloc/free). The user needs to make sure the native address and size of that memory region are valid.
- Introduce shared heap chain: The user can create a shared heap chain, from the wasm app point of view, it's still a continuous memory region in wasm app's point of view while in the native it can consist of multiple shared heaps (each of which is a continuous memory region). For example, one 500MB shared heap 1 and one 500 MB shared heap 2 form a chain, in Wasm's point of view, it's one 1GB shared heap.

After these enhancements, the data sharing between wasm apps, and between hosts can be more efficient and flexible. Admittedly shared heap management can be more complex for users, but it's similar to the zero-overhead principle. No overhead will be imposed for the users who don't use the shared heap enhancement or don't use the shared heap at all.

Author: TianlongLiang

core/iwasm/aot/aot_reloc.h

@@ -185,6 +185,13 @@ typedef struct {
 #define REG_STRINGREF_SYM()
 #endif
 
+#if WASM_ENABLE_SHARED_HEAP != 0
+#define REG_SHARED_HEAP_SYM()                 \
+    REG_SYM(wasm_runtime_check_and_update_last_used_shared_heap),
+#else
+#define REG_SHARED_HEAP_SYM()
+#endif
+
 #define REG_COMMON_SYMBOLS                \
     REG_SYM(aot_set_exception_with_id),   \
     REG_SYM(aot_invoke_native),           \
@@ -218,6 +225,7 @@ typedef struct {
     REG_LLVM_PGO_SYM()                    \
     REG_GC_SYM()                          \
     REG_STRINGREF_SYM()                   \
+    REG_SHARED_HEAP_SYM()                 \
 
 #define CHECK_RELOC_OFFSET(data_size) do {              \
     if (!check_reloc_offset(target_section_size,        \

core/iwasm/aot/aot_runtime.c

@@ -60,6 +60,16 @@ bh_static_assert(offsetof(AOTModuleInstanceExtra, stack_sizes) == 0);
 bh_static_assert(offsetof(AOTModuleInstanceExtra, shared_heap_base_addr_adj)
                  == 8);
 bh_static_assert(offsetof(AOTModuleInstanceExtra, shared_heap_start_off) == 16);
+bh_static_assert(offsetof(AOTModuleInstanceExtra, shared_heap_end_off) == 24);
+bh_static_assert(offsetof(AOTModuleInstanceExtra, shared_heap) == 32);
+
+bh_static_assert(offsetof(WASMSharedHeap, next) == 0);
+bh_static_assert(offsetof(WASMSharedHeap, chain_next) == 8);
+bh_static_assert(offsetof(WASMSharedHeap, heap_handle) == 16);
+bh_static_assert(offsetof(WASMSharedHeap, base_addr) == 24);
+bh_static_assert(offsetof(WASMSharedHeap, size) == 32);
+bh_static_assert(offsetof(WASMSharedHeap, start_off_mem64) == 40);
+bh_static_assert(offsetof(WASMSharedHeap, start_off_mem32) == 48);
 
 bh_static_assert(sizeof(CApiFuncImport) == sizeof(uintptr_t) * 3);
 
@@ -1989,6 +1999,8 @@ aot_instantiate(AOTModule *module, AOTModuleInstance *parent,
 #else
     extra->shared_heap_start_off.u32[0] = UINT32_MAX;
 #endif
+    /* After shared heap chain, will early stop if shared heap is NULL */
+    extra->shared_heap = NULL;
 
 #if WASM_ENABLE_PERF_PROFILING != 0
     total_size = sizeof(AOTFuncPerfProfInfo)

core/iwasm/aot/aot_runtime.h

@@ -125,6 +125,8 @@ typedef struct AOTModuleInstanceExtra {
      */
     DefPointer(uint8 *, shared_heap_base_addr_adj);
     MemBound shared_heap_start_off;
+    MemBound shared_heap_end_off;
+    DefPointer(WASMSharedHeap *, shared_heap);
 
     WASMModuleInstanceExtraCommon common;
 
@@ -142,9 +144,6 @@ typedef struct AOTModuleInstanceExtra {
     WASMModuleInstanceCommon **import_func_module_insts;
 #endif
 
-#if WASM_ENABLE_SHARED_HEAP != 0
-    WASMSharedHeap *shared_heap;
-#endif
 } AOTModuleInstanceExtra;
 
 #if defined(BUILD_TARGET_X86_64) || defined(BUILD_TARGET_AMD_64)