runtime_common: always use memcpy in STORE_PTR

This is one of the many causes of unaligned load warnings in
-fsanitize=alignment. memcpy should always be OK: on architectures where
unaligned stores are OK, this will turn into a single move.

This was discussed back in 2024 at
#2349.

Author: sjamesr

core/iwasm/common/wasm_runtime_common.h

@@ -6,6 +6,8 @@
 #ifndef _WASM_COMMON_H
 #define _WASM_COMMON_H
 
+#include <string.h>
+
 #include "bh_platform.h"
 #include "bh_common.h"
 #include "wasm_exec_env.h"
@@ -31,6 +33,12 @@ extern "C" {
 /* Internal use for setting default running mode */
 #define Mode_Default 0
 
+#define STORE_PTR(addr, ptr)                  \
+    do {                                      \
+        void *val = (ptr);                    \
+        memcpy((addr), &val, sizeof(void *)); \
+    } while (0)
+
 #if WASM_CPU_SUPPORTS_UNALIGNED_ADDR_ACCESS != 0
 
 #define PUT_I64_TO_ADDR(addr, value)       \
@@ -88,11 +96,6 @@ STORE_V128(void *addr, V128 value)
 #define LOAD_U16(addr) (*(uint16 *)(addr))
 #define LOAD_V128(addr) (*(V128 *)(addr))
 
-#define STORE_PTR(addr, ptr)          \
-    do {                              \
-        *(void **)addr = (void *)ptr; \
-    } while (0)
-
 #else /* WASM_CPU_SUPPORTS_UNALIGNED_ADDR_ACCESS != 0 */
 
 #define PUT_V128_TO_ADDR(addr, value)        \
@@ -465,12 +468,6 @@ LOAD_I16(void *addr)
 #define LOAD_U32(addr) ((uint32)LOAD_I32(addr))
 #define LOAD_U16(addr) ((uint16)LOAD_I16(addr))
 
-#if UINTPTR_MAX == UINT32_MAX
-#define STORE_PTR(addr, ptr) STORE_U32(addr, (uintptr_t)ptr)
-#elif UINTPTR_MAX == UINT64_MAX
-#define STORE_PTR(addr, ptr) STORE_I64(addr, (uintptr_t)ptr)
-#endif
-
 #endif /* WASM_CPU_SUPPORTS_UNALIGNED_ADDR_ACCESS != 0 */
 
 #if WASM_ENABLE_SHARED_MEMORY != 0