diff --git a/.github/workflows/compilation_on_android_ubuntu.yml b/.github/workflows/compilation_on_android_ubuntu.yml
index 2f675235e2..06a921001d 100644
--- a/.github/workflows/compilation_on_android_ubuntu.yml
+++ b/.github/workflows/compilation_on_android_ubuntu.yml
@@ -47,12 +47,12 @@ concurrency:
 
 env:
   # For BUILD
-  AOT_BUILD_OPTIONS: "           -DWAMR_BUILD_AOT=1 -DWAMR_BUILD_FAST_INTERP=0 -DWAMR_BUILD_INTERP=0 _DWAMR_BUILD_FAST_JIT=0 -DWAMR_BUILD_JIT=0 -DWAMR_BUILD_LAZY_JIT=0"
-  CLASSIC_INTERP_BUILD_OPTIONS: "-DWAMR_BUILD_AOT=0 -DWAMR_BUILD_FAST_INTERP=0 -DWAMR_BUILD_INTERP=1 _DWAMR_BUILD_FAST_JIT=0 -DWAMR_BUILD_JIT=0 -DWAMR_BUILD_LAZY_JIT=0"
-  FAST_INTERP_BUILD_OPTIONS: "   -DWAMR_BUILD_AOT=0 -DWAMR_BUILD_FAST_INTERP=1 -DWAMR_BUILD_INTERP=1 _DWAMR_BUILD_FAST_JIT=0 -DWAMR_BUILD_JIT=0 -DWAMR_BUILD_LAZY_JIT=0"
-  FAST_JIT_BUILD_OPTIONS: "      -DWAMR_BUILD_AOT=1 -DWAMR_BUILD_FAST_INTERP=0 -DWAMR_BUILD_INTERP=0 _DWAMR_BUILD_FAST_JIT=1 -DWAMR_BUILD_JIT=0 -DWAMR_BUILD_LAZY_JIT=0"
-  LLVM_LAZY_JIT_BUILD_OPTIONS: " -DWAMR_BUILD_AOT=1 -DWAMR_BUILD_FAST_INTERP=0 -DWAMR_BUILD_INTERP=0 _DWAMR_BUILD_FAST_JIT=0 -DWAMR_BUILD_JIT=1 -DWAMR_BUILD_LAZY_JIT=1"
-  LLVM_EAGER_JIT_BUILD_OPTIONS: "-DWAMR_BUILD_AOT=1 -DWAMR_BUILD_FAST_INTERP=0 -DWAMR_BUILD_INTERP=0 _DWAMR_BUILD_FAST_JIT=0 -DWAMR_BUILD_JIT=1 -DWAMR_BUILD_LAZY_JIT=0"
+  AOT_BUILD_OPTIONS: "           -DWAMR_BUILD_AOT=1 -DWAMR_BUILD_FAST_INTERP=0 -DWAMR_BUILD_INTERP=0 -DWAMR_BUILD_FAST_JIT=0 -DWAMR_BUILD_JIT=0 -DWAMR_BUILD_LAZY_JIT=0"
+  CLASSIC_INTERP_BUILD_OPTIONS: "-DWAMR_BUILD_AOT=0 -DWAMR_BUILD_FAST_INTERP=0 -DWAMR_BUILD_INTERP=1 -DWAMR_BUILD_FAST_JIT=0 -DWAMR_BUILD_JIT=0 -DWAMR_BUILD_LAZY_JIT=0"
+  FAST_INTERP_BUILD_OPTIONS: "   -DWAMR_BUILD_AOT=0 -DWAMR_BUILD_FAST_INTERP=1 -DWAMR_BUILD_INTERP=1 -DWAMR_BUILD_FAST_JIT=0 -DWAMR_BUILD_JIT=0 -DWAMR_BUILD_LAZY_JIT=0"
+  FAST_JIT_BUILD_OPTIONS: "      -DWAMR_BUILD_AOT=1 -DWAMR_BUILD_FAST_INTERP=0 -DWAMR_BUILD_INTERP=0 -DWAMR_BUILD_FAST_JIT=1 -DWAMR_BUILD_JIT=0 -DWAMR_BUILD_LAZY_JIT=0"
+  LLVM_LAZY_JIT_BUILD_OPTIONS: " -DWAMR_BUILD_AOT=1 -DWAMR_BUILD_FAST_INTERP=0 -DWAMR_BUILD_INTERP=0 -DWAMR_BUILD_FAST_JIT=0 -DWAMR_BUILD_JIT=1 -DWAMR_BUILD_LAZY_JIT=1"
+  LLVM_EAGER_JIT_BUILD_OPTIONS: "-DWAMR_BUILD_AOT=1 -DWAMR_BUILD_FAST_INTERP=0 -DWAMR_BUILD_INTERP=0 -DWAMR_BUILD_FAST_JIT=0 -DWAMR_BUILD_JIT=1 -DWAMR_BUILD_LAZY_JIT=0"
   # LLVM
   LLVM_CACHE_SUFFIX: "build-llvm_libraries_ex"
   # For Spec Test
@@ -183,6 +183,9 @@ jobs:
             make_options_feature: "-DWAMR_BUILD_MINI_LOADER=1"
           - make_options_run_mode: $LLVM_EAGER_JIT_BUILD_OPTIONS
             make_options_feature: "-DWAMR_BUILD_MINI_LOADER=1"
+          # Fast-JIT mode doesn't support android(X86-32)
+          - make_options_run_mode: $FAST_JIT_BUILD_OPTIONS
+            platform: android
     steps:
       - name: checkout
         uses: actions/checkout@v3
diff --git a/build-scripts/config_common.cmake b/build-scripts/config_common.cmake
index 1f340fc7ef..f10b5b4eb7 100644
--- a/build-scripts/config_common.cmake
+++ b/build-scripts/config_common.cmake
@@ -311,3 +311,6 @@ endif ()
 if (WAMR_BUILD_WASI_NN EQUAL 1)
   message ("     WASI-NN enabled")
 endif ()
+if (WAMR_BUILD_ALLOC_WITH_USER_DATA EQUAL 1)
+  add_definitions(-DWASM_MEM_ALLOC_WITH_USER_DATA=1)
+endif()
diff --git a/core/config.h b/core/config.h
index 0ba617ce55..288160d020 100644
--- a/core/config.h
+++ b/core/config.h
@@ -426,4 +426,8 @@
 #define WASM_ENABLE_SGX_IPFS 0
 #endif
 
+#ifndef WASM_MEM_ALLOC_WITH_USER_DATA
+#define WASM_MEM_ALLOC_WITH_USER_DATA 0
+#endif
+
 #endif /* end of _CONFIG_H_ */
diff --git a/core/iwasm/aot/aot_runtime.c b/core/iwasm/aot/aot_runtime.c
index 1c3f033c61..1077c3af82 100644
--- a/core/iwasm/aot/aot_runtime.c
+++ b/core/iwasm/aot/aot_runtime.c
@@ -1500,7 +1500,11 @@ aot_set_exception(AOTModuleInstance *module_inst, const char *exception)
 void
 aot_set_exception_with_id(AOTModuleInstance *module_inst, uint32 id)
 {
-    wasm_set_exception_with_id(module_inst, id);
+    if (id != EXCE_ALREADY_THROWN)
+        wasm_set_exception_with_id(module_inst, id);
+#ifdef OS_ENABLE_HW_BOUND_CHECK
+    wasm_runtime_access_exce_check_guard_page();
+#endif
 }
 
 const char *
@@ -1755,6 +1759,7 @@ aot_invoke_native(WASMExecEnv *exec_env, uint32 func_idx, uint32 argc,
     const char *signature;
     void *attachment;
     char buf[96];
+    bool ret = false;
 
     bh_assert(func_idx < aot_module->import_func_count);
 
@@ -1764,27 +1769,34 @@ aot_invoke_native(WASMExecEnv *exec_env, uint32 func_idx, uint32 argc,
                  "failed to call unlinked import function (%s, %s)",
                  import_func->module_name, import_func->func_name);
         aot_set_exception(module_inst, buf);
-        return false;
+        goto fail;
     }
 
     attachment = import_func->attachment;
     if (import_func->call_conv_wasm_c_api) {
-        return wasm_runtime_invoke_c_api_native(
+        ret = wasm_runtime_invoke_c_api_native(
             (WASMModuleInstanceCommon *)module_inst, func_ptr, func_type, argc,
             argv, import_func->wasm_c_api_with_env, attachment);
     }
     else if (!import_func->call_conv_raw) {
         signature = import_func->signature;
-        return wasm_runtime_invoke_native(exec_env, func_ptr, func_type,
-                                          signature, attachment, argv, argc,
-                                          argv);
+        ret =
+            wasm_runtime_invoke_native(exec_env, func_ptr, func_type, signature,
+                                       attachment, argv, argc, argv);
     }
     else {
         signature = import_func->signature;
-        return wasm_runtime_invoke_native_raw(exec_env, func_ptr, func_type,
-                                              signature, attachment, argv, argc,
-                                              argv);
+        ret = wasm_runtime_invoke_native_raw(exec_env, func_ptr, func_type,
+                                             signature, attachment, argv, argc,
+                                             argv);
     }
+
+fail:
+#ifdef OS_ENABLE_HW_BOUND_CHECK
+    if (!ret)
+        wasm_runtime_access_exce_check_guard_page();
+#endif
+    return ret;
 }
 
 bool
@@ -1811,7 +1823,7 @@ aot_call_indirect(WASMExecEnv *exec_env, uint32 tbl_idx, uint32 table_elem_idx,
 
     if ((uint8 *)&module_inst < exec_env->native_stack_boundary) {
         aot_set_exception_with_id(module_inst, EXCE_NATIVE_STACK_OVERFLOW);
-        return false;
+        goto fail;
     }
 
     tbl_inst = module_inst->tables[tbl_idx];
@@ -1819,13 +1831,13 @@ aot_call_indirect(WASMExecEnv *exec_env, uint32 tbl_idx, uint32 table_elem_idx,
 
     if (table_elem_idx >= tbl_inst->cur_size) {
         aot_set_exception_with_id(module_inst, EXCE_UNDEFINED_ELEMENT);
-        return false;
+        goto fail;
     }
 
     func_idx = tbl_inst->elems[table_elem_idx];
     if (func_idx == NULL_REF) {
         aot_set_exception_with_id(module_inst, EXCE_UNINITIALIZED_ELEMENT);
-        return false;
+        goto fail;
     }
 
     func_type_idx = func_type_indexes[func_idx];
@@ -1843,7 +1855,7 @@ aot_call_indirect(WASMExecEnv *exec_env, uint32 tbl_idx, uint32 table_elem_idx,
                  "failed to call unlinked import function (%s, %s)",
                  import_func->module_name, import_func->func_name);
         aot_set_exception(module_inst, buf);
-        return false;
+        goto fail;
     }
 
     if (func_idx < aot_module->import_func_count) {
@@ -1852,9 +1864,13 @@ aot_call_indirect(WASMExecEnv *exec_env, uint32 tbl_idx, uint32 table_elem_idx,
         signature = import_func->signature;
         if (import_func->call_conv_raw) {
             attachment = import_func->attachment;
-            return wasm_runtime_invoke_native_raw(exec_env, func_ptr, func_type,
-                                                  signature, attachment, argv,
-                                                  argc, argv);
+            ret = wasm_runtime_invoke_native_raw(exec_env, func_ptr, func_type,
+                                                 signature, attachment, argv,
+                                                 argc, argv);
+            if (!ret)
+                goto fail;
+
+            return true;
         }
     }
 
@@ -1878,7 +1894,7 @@ aot_call_indirect(WASMExecEnv *exec_env, uint32 tbl_idx, uint32 table_elem_idx,
             && !(argv1 = runtime_malloc(size, module_inst->cur_exception,
                                         sizeof(module_inst->cur_exception)))) {
             aot_set_exception_with_id(module_inst, EXCE_OUT_OF_MEMORY);
-            return false;
+            goto fail;
         }
 
         /* Copy original arguments */
@@ -1897,12 +1913,10 @@ aot_call_indirect(WASMExecEnv *exec_env, uint32 tbl_idx, uint32 table_elem_idx,
 
         ret = invoke_native_internal(exec_env, func_ptr, func_type, signature,
                                      attachment, argv1, argc, argv);
-        if (!ret || aot_get_exception(module_inst)) {
+        if (!ret) {
             if (argv1 != argv1_buf)
                 wasm_runtime_free(argv1);
-            if (clear_wasi_proc_exit_exception(module_inst))
-                return true;
-            return false;
+            goto fail;
         }
 
         /* Get extra result values */
@@ -1941,10 +1955,20 @@ aot_call_indirect(WASMExecEnv *exec_env, uint32 tbl_idx, uint32 table_elem_idx,
     else {
         ret = invoke_native_internal(exec_env, func_ptr, func_type, signature,
                                      attachment, argv, argc, argv);
-        if (clear_wasi_proc_exit_exception(module_inst))
-            return true;
-        return ret;
+        if (!ret)
+            goto fail;
+
+        return true;
     }
+
+fail:
+    if (clear_wasi_proc_exit_exception(module_inst))
+        return true;
+
+#ifdef OS_ENABLE_HW_BOUND_CHECK
+    wasm_runtime_access_exce_check_guard_page();
+#endif
+    return false;
 }
 
 bool
@@ -1952,8 +1976,17 @@ aot_check_app_addr_and_convert(AOTModuleInstance *module_inst, bool is_str,
                                uint32 app_buf_addr, uint32 app_buf_size,
                                void **p_native_addr)
 {
-    return wasm_check_app_addr_and_convert(module_inst, is_str, app_buf_addr,
-                                           app_buf_size, p_native_addr);
+    bool ret;
+
+    ret = wasm_check_app_addr_and_convert(module_inst, is_str, app_buf_addr,
+                                          app_buf_size, p_native_addr);
+
+#ifdef OS_ENABLE_HW_BOUND_CHECK
+    if (!ret)
+        wasm_runtime_access_exce_check_guard_page();
+#endif
+
+    return ret;
 }
 
 void *
diff --git a/core/iwasm/common/wasm_c_api.c b/core/iwasm/common/wasm_c_api.c
index 7a24b17a41..212a778019 100644
--- a/core/iwasm/common/wasm_c_api.c
+++ b/core/iwasm/common/wasm_c_api.c
@@ -340,6 +340,10 @@ wasm_engine_new_internal(mem_alloc_type_t type, const MemAllocOption *opts)
             opts->allocator.free_func;
         init_args.mem_alloc_option.allocator.realloc_func =
             opts->allocator.realloc_func;
+#if WASM_MEM_ALLOC_WITH_USER_DATA != 0
+        init_args.mem_alloc_option.allocator.user_data =
+            opts->allocator.user_data;
+#endif
     }
     else {
         init_args.mem_alloc_option.pool.heap_buf = NULL;
diff --git a/core/iwasm/common/wasm_exec_env.c b/core/iwasm/common/wasm_exec_env.c
index cd8b570893..515a60b8e9 100644
--- a/core/iwasm/common/wasm_exec_env.c
+++ b/core/iwasm/common/wasm_exec_env.c
@@ -56,6 +56,12 @@ wasm_exec_env_create_internal(struct WASMModuleInstanceCommon *module_inst,
 #endif
 #endif
 
+#ifdef OS_ENABLE_HW_BOUND_CHECK
+    if (!(exec_env->exce_check_guard_page =
+              os_mmap(NULL, os_getpagesize(), MMAP_PROT_NONE, MMAP_MAP_NONE)))
+        goto fail5;
+#endif
+
     exec_env->module_inst = module_inst;
     exec_env->wasm_stack_size = stack_size;
     exec_env->wasm_stack.s.top_boundary =
@@ -76,6 +82,12 @@ wasm_exec_env_create_internal(struct WASMModuleInstanceCommon *module_inst,
 
     return exec_env;
 
+#ifdef OS_ENABLE_HW_BOUND_CHECK
+fail5:
+#if WASM_ENABLE_THREAD_MGR != 0 && WASM_ENABLE_DEBUG_INTERP != 0
+    wasm_cluster_destroy_exenv_status(exec_env->current_status);
+#endif
+#endif
 #if WASM_ENABLE_THREAD_MGR != 0
 #if WASM_ENABLE_DEBUG_INTERP != 0
 fail4:
@@ -96,6 +108,9 @@ wasm_exec_env_create_internal(struct WASMModuleInstanceCommon *module_inst,
 void
 wasm_exec_env_destroy_internal(WASMExecEnv *exec_env)
 {
+#ifdef OS_ENABLE_HW_BOUND_CHECK
+    os_munmap(exec_env->exce_check_guard_page, os_getpagesize());
+#endif
 #if WASM_ENABLE_THREAD_MGR != 0
     os_mutex_destroy(&exec_env->wait_lock);
     os_cond_destroy(&exec_env->wait_cond);
diff --git a/core/iwasm/common/wasm_exec_env.h b/core/iwasm/common/wasm_exec_env.h
index d4d7ce00a1..398292079a 100644
--- a/core/iwasm/common/wasm_exec_env.h
+++ b/core/iwasm/common/wasm_exec_env.h
@@ -137,6 +137,8 @@ typedef struct WASMExecEnv {
 
 #ifdef OS_ENABLE_HW_BOUND_CHECK
     WASMJmpBuf *jmpbuf_stack_top;
+    /* One guard page for the exception check */
+    uint8 *exce_check_guard_page;
 #endif
 
 #if WASM_ENABLE_MEMORY_PROFILING != 0
@@ -199,7 +201,8 @@ wasm_exec_env_alloc_wasm_frame(WASMExecEnv *exec_env, unsigned size)
        the outs area contains const cells, its size may be larger than current
        frame size, we should check again before putting the function arguments
        into the outs area. */
-    if (addr + size * 2 > exec_env->wasm_stack.s.top_boundary) {
+    if (size * 2
+        > (uint32)(uintptr_t)(exec_env->wasm_stack.s.top_boundary - addr)) {
         /* WASM stack overflow. */
         return NULL;
     }
diff --git a/core/iwasm/common/wasm_memory.c b/core/iwasm/common/wasm_memory.c
index ffe21857da..0aa2d6415c 100644
--- a/core/iwasm/common/wasm_memory.c
+++ b/core/iwasm/common/wasm_memory.c
@@ -11,16 +11,25 @@
 typedef enum Memory_Mode {
     MEMORY_MODE_UNKNOWN = 0,
     MEMORY_MODE_POOL,
-    MEMORY_MODE_ALLOCATOR
+    MEMORY_MODE_ALLOCATOR,
+    MEMORY_MODE_SYSTEM_ALLOCATOR
 } Memory_Mode;
 
 static Memory_Mode memory_mode = MEMORY_MODE_UNKNOWN;
 
 static mem_allocator_t pool_allocator = NULL;
 
+#if WASM_MEM_ALLOC_WITH_USER_DATA != 0
+static void *allocator_user_data = NULL;
+static void *(*malloc_func)(void *user_data, unsigned int size) = NULL;
+static void *(*realloc_func)(void *user_data, void *ptr,
+                             unsigned int size) = NULL;
+static void (*free_func)(void *user_data, void *ptr) = NULL;
+#else
 static void *(*malloc_func)(unsigned int size) = NULL;
 static void *(*realloc_func)(void *ptr, unsigned int size) = NULL;
 static void (*free_func)(void *ptr) = NULL;
+#endif
 
 static unsigned int global_pool_size;
 
@@ -39,6 +48,24 @@ wasm_memory_init_with_pool(void *mem, unsigned int bytes)
     return false;
 }
 
+#if WASM_MEM_ALLOC_WITH_USER_DATA != 0
+static bool
+wasm_memory_init_with_allocator(void *_user_data, void *_malloc_func,
+                                void *_realloc_func, void *_free_func)
+{
+    if (_malloc_func && _free_func && _malloc_func != _free_func) {
+        memory_mode = MEMORY_MODE_ALLOCATOR;
+        allocator_user_data = _user_data;
+        malloc_func = _malloc_func;
+        realloc_func = _realloc_func;
+        free_func = _free_func;
+        return true;
+    }
+    LOG_ERROR("Init memory with allocator (%p, %p, %p, %p) failed.\n",
+              _user_data, _malloc_func, _realloc_func, _free_func);
+    return false;
+}
+#else
 static bool
 wasm_memory_init_with_allocator(void *_malloc_func, void *_realloc_func,
                                 void *_free_func)
@@ -54,23 +81,37 @@ wasm_memory_init_with_allocator(void *_malloc_func, void *_realloc_func,
               _realloc_func, _free_func);
     return false;
 }
+#endif
 
 bool
 wasm_runtime_memory_init(mem_alloc_type_t mem_alloc_type,
                          const MemAllocOption *alloc_option)
 {
-    if (mem_alloc_type == Alloc_With_Pool)
+    if (mem_alloc_type == Alloc_With_Pool) {
         return wasm_memory_init_with_pool(alloc_option->pool.heap_buf,
                                           alloc_option->pool.heap_size);
-    else if (mem_alloc_type == Alloc_With_Allocator)
+    }
+    else if (mem_alloc_type == Alloc_With_Allocator) {
+#if WASM_MEM_ALLOC_WITH_USER_DATA != 0
         return wasm_memory_init_with_allocator(
+            alloc_option->allocator.user_data,
             alloc_option->allocator.malloc_func,
             alloc_option->allocator.realloc_func,
             alloc_option->allocator.free_func);
-    else if (mem_alloc_type == Alloc_With_System_Allocator)
-        return wasm_memory_init_with_allocator(os_malloc, os_realloc, os_free);
-    else
+#else
+        return wasm_memory_init_with_allocator(
+            alloc_option->allocator.malloc_func,
+            alloc_option->allocator.realloc_func,
+            alloc_option->allocator.free_func);
+#endif
+    }
+    else if (mem_alloc_type == Alloc_With_System_Allocator) {
+        memory_mode = MEMORY_MODE_SYSTEM_ALLOCATOR;
+        return true;
+    }
+    else {
         return false;
+    }
 }
 
 void
@@ -110,8 +151,15 @@ wasm_runtime_malloc_internal(unsigned int size)
     else if (memory_mode == MEMORY_MODE_POOL) {
         return mem_allocator_malloc(pool_allocator, size);
     }
-    else {
+    else if (memory_mode == MEMORY_MODE_ALLOCATOR) {
+#if WASM_MEM_ALLOC_WITH_USER_DATA != 0
+        return malloc_func(allocator_user_data, size);
+#else
         return malloc_func(size);
+#endif
+    }
+    else {
+        return os_malloc(size);
     }
 }
 
@@ -126,12 +174,19 @@ wasm_runtime_realloc_internal(void *ptr, unsigned int size)
     else if (memory_mode == MEMORY_MODE_POOL) {
         return mem_allocator_realloc(pool_allocator, ptr, size);
     }
-    else {
+    else if (memory_mode == MEMORY_MODE_ALLOCATOR) {
         if (realloc_func)
+#if WASM_MEM_ALLOC_WITH_USER_DATA != 0
+            return realloc_func(allocator_user_data, ptr, size);
+#else
             return realloc_func(ptr, size);
+#endif
         else
             return NULL;
     }
+    else {
+        return os_realloc(ptr, size);
+    }
 }
 
 static inline void
@@ -152,8 +207,15 @@ wasm_runtime_free_internal(void *ptr)
     else if (memory_mode == MEMORY_MODE_POOL) {
         mem_allocator_free(pool_allocator, ptr);
     }
-    else {
+    else if (memory_mode == MEMORY_MODE_ALLOCATOR) {
+#if WASM_MEM_ALLOC_WITH_USER_DATA != 0
+        free_func(allocator_user_data, ptr);
+#else
         free_func(ptr);
+#endif
+    }
+    else {
+        os_free(ptr);
     }
 }
 
diff --git a/core/iwasm/common/wasm_runtime_common.c b/core/iwasm/common/wasm_runtime_common.c
index fbcdfcd009..17cf9da31b 100644
--- a/core/iwasm/common/wasm_runtime_common.c
+++ b/core/iwasm/common/wasm_runtime_common.c
@@ -185,6 +185,12 @@ runtime_signal_handler(void *sig_addr)
             os_longjmp(jmpbuf_node->jmpbuf, 1);
         }
 #endif
+        else if (exec_env_tls->exce_check_guard_page <= (uint8 *)sig_addr
+                 && (uint8 *)sig_addr
+                        < exec_env_tls->exce_check_guard_page + page_size) {
+            bh_assert(wasm_get_exception(module_inst));
+            os_longjmp(jmpbuf_node->jmpbuf, 1);
+        }
     }
 }
 #else
@@ -1435,6 +1441,17 @@ wasm_runtime_get_user_data(WASMExecEnv *exec_env)
     return exec_env->user_data;
 }
 
+#ifdef OS_ENABLE_HW_BOUND_CHECK
+void
+wasm_runtime_access_exce_check_guard_page()
+{
+    if (exec_env_tls && exec_env_tls->handle == os_self_thread()) {
+        uint32 page_size = os_getpagesize();
+        memset(exec_env_tls->exce_check_guard_page, 0, page_size);
+    }
+}
+#endif
+
 WASMType *
 wasm_runtime_get_function_type(const WASMFunctionInstanceCommon *function,
                                uint32 module_type)
diff --git a/core/iwasm/common/wasm_runtime_common.h b/core/iwasm/common/wasm_runtime_common.h
index cc311ba730..3e7f3725d2 100644
--- a/core/iwasm/common/wasm_runtime_common.h
+++ b/core/iwasm/common/wasm_runtime_common.h
@@ -554,6 +554,12 @@ wasm_runtime_set_user_data(WASMExecEnv *exec_env, void *user_data);
 WASM_RUNTIME_API_EXTERN void *
 wasm_runtime_get_user_data(WASMExecEnv *exec_env);
 
+#ifdef OS_ENABLE_HW_BOUND_CHECK
+/* Access exception check guard page to trigger the signal handler */
+void
+wasm_runtime_access_exce_check_guard_page();
+#endif
+
 /* See wasm_export.h for description */
 WASM_RUNTIME_API_EXTERN bool
 wasm_runtime_call_wasm(WASMExecEnv *exec_env,
diff --git a/core/iwasm/common/wasm_shared_memory.c b/core/iwasm/common/wasm_shared_memory.c
index 862bbe964c..b2b63de5c5 100644
--- a/core/iwasm/common/wasm_shared_memory.c
+++ b/core/iwasm/common/wasm_shared_memory.c
@@ -386,7 +386,8 @@ wasm_runtime_atomic_wait(WASMModuleInstanceCommon *module, void *address,
     os_mutex_lock(&wait_node->wait_lock);
 
     os_cond_reltimedwait(&wait_node->wait_cond, &wait_node->wait_lock,
-                         timeout < 0 ? BHT_WAIT_FOREVER : timeout / 1000);
+                         timeout < 0 ? BHT_WAIT_FOREVER
+                                     : (uint64)timeout / 1000);
 
     os_mutex_unlock(&wait_node->wait_lock);
 
diff --git a/core/iwasm/compilation/aot_emit_function.c b/core/iwasm/compilation/aot_emit_function.c
index cbcdebffc1..fb2fca5656 100644
--- a/core/iwasm/compilation/aot_emit_function.c
+++ b/core/iwasm/compilation/aot_emit_function.c
@@ -35,7 +35,14 @@ create_func_return_block(AOTCompContext *comp_ctx, AOTFuncContext *func_ctx)
         /* Create return IR */
         LLVMPositionBuilderAtEnd(comp_ctx->builder,
                                  func_ctx->func_return_block);
-        if (!aot_build_zero_function_ret(comp_ctx, func_ctx, aot_func_type)) {
+        if (!comp_ctx->enable_bound_check) {
+            if (!aot_emit_exception(comp_ctx, func_ctx, EXCE_ALREADY_THROWN,
+                                    false, NULL, NULL)) {
+                return false;
+            }
+        }
+        else if (!aot_build_zero_function_ret(comp_ctx, func_ctx,
+                                              aot_func_type)) {
             return false;
         }
     }
@@ -494,7 +501,8 @@ check_app_addr_and_convert(AOTCompContext *comp_ctx, AOTFuncContext *func_ctx,
     }
 
     /* Check whether exception was thrown when executing the function */
-    if (!check_call_return(comp_ctx, func_ctx, res)) {
+    if (comp_ctx->enable_bound_check
+        && !check_call_return(comp_ctx, func_ctx, res)) {
         return false;
     }
 
@@ -707,7 +715,8 @@ aot_compile_op_call(AOTCompContext *comp_ctx, AOTFuncContext *func_ctx,
                 goto fail;
             /* Check whether there was exception thrown when executing
                the function */
-            if (!check_call_return(comp_ctx, func_ctx, res))
+            if (comp_ctx->enable_bound_check
+                && !check_call_return(comp_ctx, func_ctx, res))
                 goto fail;
         }
         else { /* call native func directly */
@@ -823,7 +832,7 @@ aot_compile_op_call(AOTCompContext *comp_ctx, AOTFuncContext *func_ctx,
 
         /* Check whether there was exception thrown when executing
            the function */
-        if (!tail_call && !recursive_call
+        if (!tail_call && !recursive_call && comp_ctx->enable_bound_check
             && !check_exception_thrown(comp_ctx, func_ctx))
             goto fail;
     }
@@ -1395,7 +1404,8 @@ aot_compile_op_call_indirect(AOTCompContext *comp_ctx, AOTFuncContext *func_ctx,
         goto fail;
 
     /* Check whether exception was thrown when executing the function */
-    if (!check_call_return(comp_ctx, func_ctx, res))
+    if (comp_ctx->enable_bound_check
+        && !check_call_return(comp_ctx, func_ctx, res))
         goto fail;
 
     block_curr = LLVMGetInsertBlock(comp_ctx->builder);
@@ -1454,7 +1464,8 @@ aot_compile_op_call_indirect(AOTCompContext *comp_ctx, AOTFuncContext *func_ctx,
     }
 
     /* Check whether exception was thrown when executing the function */
-    if (!check_exception_thrown(comp_ctx, func_ctx))
+    if (comp_ctx->enable_bound_check
+        && !check_exception_thrown(comp_ctx, func_ctx))
         goto fail;
 
     if (func_result_count > 0) {
diff --git a/core/iwasm/compilation/aot_llvm_extra.cpp b/core/iwasm/compilation/aot_llvm_extra.cpp
index 52e219b2d2..9b77f5e6a1 100644
--- a/core/iwasm/compilation/aot_llvm_extra.cpp
+++ b/core/iwasm/compilation/aot_llvm_extra.cpp
@@ -344,11 +344,12 @@ aot_apply_llvm_new_pass_manager(AOTCompContext *comp_ctx, LLVMModuleRef module)
 
         if (!disable_llvm_lto) {
             /* Apply LTO for AOT mode */
-#if LLVM_VERSION_MAJOR < 14
-            MPM.addPass(PB.buildLTODefaultPipeline(OL, NULL));
-#else
-            MPM.addPass(PB.buildLTOPreLinkDefaultPipeline(OL));
-#endif
+            if (comp_ctx->comp_data->func_count >= 10)
+                /* Adds the pre-link optimizations if the func count
+                   is large enough */
+                MPM.addPass(PB.buildLTOPreLinkDefaultPipeline(OL));
+            else
+                MPM.addPass(PB.buildLTODefaultPipeline(OL, NULL));
         }
         else {
             MPM.addPass(PB.buildPerModuleDefaultPipeline(OL));
diff --git a/core/iwasm/fast-jit/jit_frontend.c b/core/iwasm/fast-jit/jit_frontend.c
index 50cc3717f6..995bfb80a1 100644
--- a/core/iwasm/fast-jit/jit_frontend.c
+++ b/core/iwasm/fast-jit/jit_frontend.c
@@ -1286,11 +1286,8 @@ jit_compile_func(JitCompContext *cc)
 
 #if WASM_ENABLE_TAIL_CALL != 0
             case WASM_OP_RETURN_CALL:
-                if (!cc->enable_tail_call) {
-                    jit_set_last_error(cc, "unsupported opcode");
-                    return false;
-                }
                 read_leb_uint32(frame_ip, frame_ip_end, func_idx);
+
                 if (!jit_compile_op_call(cc, func_idx, true))
                     return false;
                 if (!jit_compile_op_return(cc, &frame_ip))
@@ -1301,11 +1298,6 @@ jit_compile_func(JitCompContext *cc)
             {
                 uint32 tbl_idx;
 
-                if (!cc->enable_tail_call) {
-                    jit_set_last_error(cc, "unsupported opcode");
-                    return false;
-                }
-
                 read_leb_uint32(frame_ip, frame_ip_end, type_idx);
 #if WASM_ENABLE_REF_TYPES != 0
                 read_leb_uint32(frame_ip, frame_ip_end, tbl_idx);
diff --git a/core/iwasm/include/wasm_c_api.h b/core/iwasm/include/wasm_c_api.h
index 8900978957..85fcd1868a 100644
--- a/core/iwasm/include/wasm_c_api.h
+++ b/core/iwasm/include/wasm_c_api.h
@@ -186,6 +186,9 @@ typedef union MemAllocOption {
         void *malloc_func;
         void *realloc_func;
         void *free_func;
+        /* allocator user data, only used when
+          WASM_MEM_ALLOC_WITH_USER_DATA is defined */
+        void *user_data;
     } allocator;
 } MemAllocOption;
 #endif
diff --git a/core/iwasm/include/wasm_export.h b/core/iwasm/include/wasm_export.h
index cc173af92d..2d5e85155d 100644
--- a/core/iwasm/include/wasm_export.h
+++ b/core/iwasm/include/wasm_export.h
@@ -117,6 +117,9 @@ typedef union MemAllocOption {
         void *malloc_func;
         void *realloc_func;
         void *free_func;
+        /* allocator user data, only used when
+            WASM_MEM_ALLOC_WITH_USER_DATA is defined */
+        void *user_data;
     } allocator;
 } MemAllocOption;
 #endif
diff --git a/core/iwasm/interpreter/wasm_runtime.c b/core/iwasm/interpreter/wasm_runtime.c
index f702a0e843..8e8f96b5c3 100644
--- a/core/iwasm/interpreter/wasm_runtime.c
+++ b/core/iwasm/interpreter/wasm_runtime.c
@@ -2781,7 +2781,11 @@ wasm_interp_dump_call_stack(struct WASMExecEnv *exec_env, bool print, char *buf,
 void
 jit_set_exception_with_id(WASMModuleInstance *module_inst, uint32 id)
 {
-    wasm_set_exception_with_id(module_inst, id);
+    if (id != EXCE_ALREADY_THROWN)
+        wasm_set_exception_with_id(module_inst, id);
+#ifdef OS_ENABLE_HW_BOUND_CHECK
+    wasm_runtime_access_exce_check_guard_page();
+#endif
 }
 
 bool
@@ -2789,8 +2793,15 @@ jit_check_app_addr_and_convert(WASMModuleInstance *module_inst, bool is_str,
                                uint32 app_buf_addr, uint32 app_buf_size,
                                void **p_native_addr)
 {
-    return wasm_check_app_addr_and_convert(module_inst, is_str, app_buf_addr,
-                                           app_buf_size, p_native_addr);
+    bool ret = wasm_check_app_addr_and_convert(
+        module_inst, is_str, app_buf_addr, app_buf_size, p_native_addr);
+
+#ifdef OS_ENABLE_HW_BOUND_CHECK
+    if (!ret)
+        wasm_runtime_access_exce_check_guard_page();
+#endif
+
+    return ret;
 }
 #endif /* end of WASM_ENABLE_FAST_JIT != 0 || WASM_ENABLE_JIT != 0 \
           || WASM_ENABLE_WAMR_COMPILER != 0 */
@@ -2811,12 +2822,20 @@ bool
 llvm_jit_call_indirect(WASMExecEnv *exec_env, uint32 tbl_idx, uint32 elem_idx,
                        uint32 argc, uint32 *argv)
 {
+    bool ret;
+
 #if WASM_ENABLE_JIT != 0
     if (Wasm_Module_AoT == exec_env->module_inst->module_type) {
         return aot_call_indirect(exec_env, tbl_idx, elem_idx, argc, argv);
     }
 #endif
-    return call_indirect(exec_env, tbl_idx, elem_idx, argc, argv, false, 0);
+
+    ret = call_indirect(exec_env, tbl_idx, elem_idx, argc, argv, false, 0);
+#ifdef OS_ENABLE_HW_BOUND_CHECK
+    if (!ret)
+        wasm_runtime_access_exce_check_guard_page();
+#endif
+    return ret;
 }
 
 bool
@@ -2833,6 +2852,7 @@ llvm_jit_invoke_native(WASMExecEnv *exec_env, uint32 func_idx, uint32 argc,
     const char *signature;
     void *attachment;
     char buf[96];
+    bool ret = false;
 
 #if WASM_ENABLE_JIT != 0
     if (Wasm_Module_AoT == exec_env->module_inst->module_type) {
@@ -2855,27 +2875,34 @@ llvm_jit_invoke_native(WASMExecEnv *exec_env, uint32 func_idx, uint32 argc,
                  "failed to call unlinked import function (%s, %s)",
                  import_func->module_name, import_func->field_name);
         wasm_set_exception(module_inst, buf);
-        return false;
+        goto fail;
     }
 
     attachment = import_func->attachment;
     if (import_func->call_conv_wasm_c_api) {
-        return wasm_runtime_invoke_c_api_native(
+        ret = wasm_runtime_invoke_c_api_native(
             (WASMModuleInstanceCommon *)module_inst, func_ptr, func_type, argc,
             argv, import_func->wasm_c_api_with_env, attachment);
     }
     else if (!import_func->call_conv_raw) {
         signature = import_func->signature;
-        return wasm_runtime_invoke_native(exec_env, func_ptr, func_type,
-                                          signature, attachment, argv, argc,
-                                          argv);
+        ret =
+            wasm_runtime_invoke_native(exec_env, func_ptr, func_type, signature,
+                                       attachment, argv, argc, argv);
     }
     else {
         signature = import_func->signature;
-        return wasm_runtime_invoke_native_raw(exec_env, func_ptr, func_type,
-                                              signature, attachment, argv, argc,
-                                              argv);
+        ret = wasm_runtime_invoke_native_raw(exec_env, func_ptr, func_type,
+                                             signature, attachment, argv, argc,
+                                             argv);
     }
+
+fail:
+#ifdef OS_ENABLE_HW_BOUND_CHECK
+    if (!ret)
+        wasm_runtime_access_exce_check_guard_page();
+#endif
+    return ret;
 }
 
 #if WASM_ENABLE_BULK_MEMORY != 0
diff --git a/doc/source_debugging.md b/doc/source_debugging.md
index a07cebc43b..c18c61267e 100644
--- a/doc/source_debugging.md
+++ b/doc/source_debugging.md
@@ -28,6 +28,7 @@ mkdir build && cd build
 cmake .. -DWAMR_BUILD_DEBUG_INTERP=1
 make
 ```
+> Note: On MacOS M1 environment, pass the additional `-DWAMR_DISABLE_HW_BOUND_CHECK=1` cmake configuration.
 
 3. Execute iwasm with debug engine enabled
 ``` bash
@@ -43,6 +44,7 @@ mkdir build-lldb && cd build-lldb
 cmake -DCMAKE_BUILD_TYPE:STRING="Release" -DLLVM_ENABLE_PROJECTS="clang;lldb" -DLLVM_TARGETS_TO_BUILD:STRING="X86;WebAssembly" -DLLVM_ENABLE_LIBXML2:BOOL=ON ../llvm
 make -j $(nproc)
 ```
+> Note: If using `CommandLineTools` on MacOS, make sure only one SDK is present in `/Library/Developer/CommandLineTools/SDKs`.
 
 5. Launch customized lldb and connect to iwasm
 ``` bash