Make internal symbols hidden.

Only the malloc/free/... interface is exported.

This allows all of C++'s weak ODR symbols to be resolved at
compile-time, rather than staying weak and resolved at load-time.

Since internal calls don't need to go through the PLT anymore, we get a
nice speedup, at least on small objects.

CMakeLists.txt

@@ -54,16 +54,18 @@ if(USE_SBRK)
   add_definitions(-DUSE_SBRK)
 endif()
 
-if(NOT MSVC)
-  add_library(snmallocshim SHARED src/override/malloc.cc)
-  target_link_libraries(snmallocshim -pthread)
-  target_include_directories(snmallocshim PRIVATE src)
-endif()
+macro(add_shim name)
+  add_library(${name} SHARED src/override/malloc.cc)
+  target_link_libraries(${name} -pthread)
+  target_include_directories(${name} PRIVATE src)
+  target_compile_definitions(${name} PRIVATE "SNMALLOC_EXPORT=__attribute__((visibility(\"default\")))")
+  set_target_properties(${name} PROPERTIES CXX_VISIBILITY_PRESET hidden)
+endmacro()
 
 if(NOT MSVC)
-  add_library(snmallocshim-1mib SHARED src/override/malloc.cc)
-  target_link_libraries(snmallocshim-1mib -pthread)
-  target_include_directories(snmallocshim-1mib PRIVATE src)
+  add_shim(snmallocshim)
+
+  add_shim(snmallocshim-1mib)
   target_compile_definitions(snmallocshim-1mib PRIVATE IS_ADDRESS_SPACE_CONSTRAINED)
 endif()