Merge branch 'incoming'

8 files changed, 200 insertions, 43 deletions

diff --git a/AUTHORS b/AUTHORS
index a1e995c5..264f1e4c 100644
--- a/AUTHORS
+++ b/AUTHORS
@@ -41,4 +41,5 @@ a license to everyone to use it as detailed in LICENSE.)
 * Joel Martin <github@martintribe.org>
 * Manuel Wellmann <manuel.wellmann@autodesk.com> (copyright owned by Autodesk, Inc.)
 * Xuejie Xiao <xxuejie@gmail.com>
+* Dominic Wong <dom@slowbunyip.org>
 
diff --git a/emcc b/emcc
index dc8bd630..48631d64 100755
--- a/emcc
+++ b/emcc
@@ -374,10 +374,12 @@ or LLVM assembly files in human-readable form.
 emcc is affected by several environment variables. For details, view
 the source of emcc (search for 'os.environ').
 
+emcc: supported targets: llvm bitcode, javascript, NOT elf
+(autoconf likes to see elf above to enable shared object support)
 ''' % (this, this, this)
   exit(0)
 elif len(sys.argv) == 2 and sys.argv[1] == '-v': # -v with no inputs
-  print 'emcc (Emscripten GCC-like replacement) 2.0'
+  print 'emcc (Emscripten GCC-like replacement + linker emulating GNU ld ) 2.0'
   exit(subprocess.call([shared.CLANG, '-v']))
 
 def is_minus_s_for_emcc(newargs,i):
@@ -540,6 +542,15 @@ for i in range(len(sys.argv)-1):
     sys.argv = sys.argv[:i] + sys.argv[i+2:]
     break
 
+specified_target = target
+target = specified_target if specified_target is not None else 'a.out.js' # specified_target is the user-specified one, target is what we will generate
+target_basename = unsuffixed_basename(target)
+
+if '.' in target:
+  final_suffix = target.split('.')[-1]
+else:
+  final_suffix = ''
+
 if header: # header or such
   if len(sys.argv) >= 3: # if there is a source and a target, then copy, otherwise do nothing
     sys.argv = filter(lambda arg: not arg.startswith('-I'), sys.argv)
@@ -751,7 +762,7 @@ try:
 
     if i > 0:
       prev = newargs[i-1]
-      if prev in ['-MT', '-install_name']: continue # ignore this gcc-style argument
+      if prev in ['-MT', '-install_name', '-I', '-L']: continue # ignore this gcc-style argument
 
     if not arg.startswith('-') and (arg.endswith(SOURCE_SUFFIXES + BITCODE_SUFFIXES + DYNAMICLIB_SUFFIXES + ASSEMBLY_SUFFIXES) or shared.Building.is_ar(arg)): # we already removed -o <target>, so all these should be inputs
       newargs[i] = ''
@@ -789,6 +800,17 @@ try:
 
   newargs = [ arg for arg in newargs if arg is not '' ]
 
+  # -c means do not link in gcc, and for us, the parallel is to not go all the way to JS, but stop at bitcode
+  has_dash_c = '-c' in newargs
+  if has_dash_c:
+    assert has_source_inputs, 'Must have source code inputs to use -c'
+    target = target_basename + '.o'
+    final_suffix = 'o'
+
+  # do not link in libs when just generating object code (not an 'executable', i.e. JS, or a library)
+  if ('.' + final_suffix) in BITCODE_SUFFIXES:
+    libs = []
+
   # Find library files
   for lib in libs:
     if DEBUG: print >> sys.stderr, 'emcc: looking for library "%s"' % lib
@@ -816,22 +838,6 @@ try:
 
   newargs += CC_ADDITIONAL_ARGS
 
-  specified_target = target
-  target = specified_target if specified_target is not None else 'a.out.js' # specified_target is the user-specified one, target is what we will generate
-
-  target_basename = unsuffixed_basename(target)
-
-  # -c means do not link in gcc, and for us, the parallel is to not go all the way to JS, but stop at bitcode
-  has_dash_c = '-c' in newargs
-  if has_dash_c:
-    assert has_source_inputs, 'Must have source code inputs to use -c'
-    target = target_basename + '.o'
-
-  if '.' in target:
-    final_suffix = target.split('.')[-1]
-  else:
-    final_suffix = ''
-
   assert not (Compression.on and final_suffix != 'html'), 'Compression only works when generating HTML'
 
   # If we are using embind and generating JS, now is the time to link in bind.cpp
@@ -904,7 +910,7 @@ try:
         assert len(original_input_files) == 1 or not has_dash_c, 'fatal error: cannot specify -o with -c with multiple files' + str(sys.argv) + ':' + str(original_input_files)
         # We have a specified target (-o <target>), which is not JavaScript or HTML, and
         # we have multiple files: Link them
-        if DEBUG: print >> sys.stderr, 'emcc: link: ' + str(temp_files)
+        if DEBUG: print >> sys.stderr, 'emcc: link: ' + str(temp_files), specified_target
         shared.Building.link(temp_files, specified_target, remove_duplicates=remove_duplicates)
     exit(0)
 
@@ -985,20 +991,32 @@ try:
     for name, create, fix, library_symbols in [('libcxx',    create_libcxx,    fix_libcxx,    libcxx_symbols),
                                                ('libcxxabi', create_libcxxabi, fix_libcxxabi, libcxxabi_symbols),
                                                ('dlmalloc',  create_dlmalloc,  fix_dlmalloc,  dlmalloc_symbols)]:
-      need = []
-      has = []
+      need = set()
+      has = set()
       for temp_file in temp_files:
         symbols = shared.Building.llvm_nm(temp_file)
         for library_symbol in library_symbols:
           if library_symbol in symbols.undefs:
-            need.append(library_symbol)
+            need.add(library_symbol)
           if library_symbol in symbols.defs:
-            has.append(library_symbol)
-      if DEBUG: print >> sys.stderr, 'emcc: considering including %s: we need |%s| and have |%s|' % (name, str(need), str(has))
-      if force or (need and not has):
+            has.add(library_symbol)
+      for haz in has: # remove symbols that are supplied by another of the inputs
+        if haz in need:
+          need.remove(haz)
+      if DEBUG: print >> sys.stderr, 'emcc: considering including %s: we need %s and have %s' % (name, str(need), str(has))
+      if force or len(need) > 0:
         # We need to build and link the library in
         if DEBUG: print >> sys.stderr, 'emcc: including %s' % name
-        extra_files_to_link.append(shared.Cache.get(name, create))
+        libfile = shared.Cache.get(name, create)
+        if len(has) > 0:
+          # remove the symbols we do not need
+          fixed = in_temp(uniquename(libfile)) + '.bc'
+          shutil.copyfile(libfile, fixed)
+          for haz in has:
+            if DEBUG: print >> sys.stderr, 'emcc: including: removing symbol "%s" that we have' % haz
+            shared.Building.remove_symbol(fixed, haz)
+          libfile = fixed
+        extra_files_to_link.append(libfile)
         force = True
         if fix:
           fix()
diff --git a/src/parseTools.js b/src/parseTools.js
index 7387bf31..4a76a9a2 100644
--- a/src/parseTools.js
+++ b/src/parseTools.js
@@ -1750,7 +1750,7 @@ function processMathop(item) {
           }
         }
       }
-      case 'uitofp': case 'sitofp': return low1 + ' + ' + high1 + '*4294967296';
+      case 'uitofp': case 'sitofp': return RuntimeGenerator.makeBigInt(low1, high1, op[0] == 'u');
       case 'fptoui': case 'fptosi': return finish(splitI64(idents[0]));
       case 'icmp': {
         switch (variant) {
diff --git a/system/include/emscripten/bind.h b/system/include/emscripten/bind.h
index 4a82eacb..8ce95de7 100644
--- a/system/include/emscripten/bind.h
+++ b/system/include/emscripten/bind.h
@@ -218,7 +218,7 @@ namespace emscripten {
         template<typename ClassType, typename ReturnType, typename... Args>
         struct MethodInvoker {
             typedef ReturnType (ClassType::*MemberPointer)(Args...);
-            typename internal::BindingType<ReturnType>::WireType invoke(
+            static typename internal::BindingType<ReturnType>::WireType invoke(
                 ClassType* ptr,
                 const MemberPointer& method,
                 typename internal::BindingType<Args>::WireType... args
diff --git a/tests/embind/embind_test.cpp b/tests/embind/embind_test.cpp
index e7b4d985..dc052d1a 100644
--- a/tests/embind/embind_test.cpp
+++ b/tests/embind/embind_test.cpp
@@ -96,6 +96,10 @@ public:
         this->v = v;

     }

 

+    int returnIntPlusFive( int x ) {

+        return x + 5;

+    }

+

     static int some_class_method(int i) {

         return i;

     }

@@ -286,6 +290,7 @@ EMSCRIPTEN_BINDINGS(([]() {
         .constructor<val>()

         .method("getVal", &ValHolder::getVal)

         .method("setVal", &ValHolder::setVal)

+        .method("returnIntPlusFive", &ValHolder::returnIntPlusFive)

         .classmethod("some_class_method", &ValHolder::some_class_method)

         ;

     function("emval_test_return_ValHolder", &emval_test_return_ValHolder);

diff --git a/tests/embind/embind_test.js b/tests/embind/embind_test.js
index e01f0236..8c61553b 100644
--- a/tests/embind/embind_test.js
+++ b/tests/embind/embind_test.js
@@ -137,6 +137,8 @@ module({
             c.setVal('1234');
             assert.equal('1234', c.getVal());
 
+            assert.equal(1239, c.returnIntPlusFive(1234));
+
             c.delete();
             assert.equal(0, cm.count_emval_handles());
         },
diff --git a/tests/runner.py b/tests/runner.py
index cef14e95..1480afe7 100755
--- a/tests/runner.py
+++ b/tests/runner.py
@@ -1083,6 +1083,32 @@ m_divisor is 1091269979
       '''
       self.do_run(src, '<=0')
 
+    def test_i64_qdouble(self):
+      if Settings.USE_TYPED_ARRAYS != 2: return self.skip('full i64 stuff only in ta2')
+
+      src = r'''
+        #include <stdio.h>
+        typedef long long qint64; /* 64 bit signed */
+        typedef double qreal;
+
+
+        int main(int argc, char **argv)
+        {
+          qreal c = 111;
+          qint64 d = -111 + (argc - 1);
+          c += d;
+          if (c < -1 || c > 1)
+          {
+                  printf("Failed!\n");
+          }
+          else
+          {
+                  printf("Succeeded!\n");
+          }
+        };
+      '''
+      self.do_run(src, 'Succeeded!')
+
     def test_i32_mul_precise(self):
       if self.emcc_args == None: return self.skip('needs ta2')
 
@@ -5779,6 +5805,39 @@ int main(int argc, char **argv) {
         ]:
           self.do_run(src.replace('{{{ NEW }}}', new).replace('{{{ DELETE }}}', delete), '*1,0*')
 
+    def test_dlmalloc_partial(self):
+      if self.emcc_args is None: return self.skip('only emcc will link in dlmalloc')
+      # present part of the symbols of dlmalloc, not all
+      src = open(path_from_root('tests', 'new.cpp')).read().replace('{{{ NEW }}}', 'new int').replace('{{{ DELETE }}}', 'delete') + '''    
+void *
+operator new(size_t size)
+{
+  printf("new %d!\\n", size);
+  return malloc(size);
+}
+'''
+      self.do_run(src, 'new 4!\n*1,0*')
+
+    def test_dlmalloc_partial_2(self):
+      if self.emcc_args is None or 'SAFE_HEAP' in str(self.emcc_args): return self.skip('only emcc will link in dlmalloc, and we do unsafe stuff')
+      # present part of the symbols of dlmalloc, not all. malloc is harder to link than new which is weak.
+      src = r'''
+        #include <stdio.h>
+        #include <stdlib.h>
+        void *malloc(size_t size)
+        {
+          return (void*)123;
+        }
+        int main() {
+          void *x = malloc(10);
+          printf("got %p\n", x);
+          free(x);
+          printf("freed the faker\n");
+          return 1;
+        }
+'''
+      self.do_run(src, 'got 0x7b\nfreed')
+
     def test_libcxx(self):
       self.do_run(open(path_from_root('tests', 'hashtest.cpp')).read(),
                    'june -> 30\nPrevious (in alphabetical order) is july\nNext (in alphabetical order) is march')
@@ -7850,6 +7909,70 @@ f.close()
       self.assertContained('hello from lib', run_js(os.path.join(self.get_dir(), 'a.out.js')))
       assert not os.path.exists('a.out') and not os.path.exists('a.exe'), 'Must not leave unneeded linker stubs'
 
+    def test_multiply_defined_libsymbols(self):
+      lib = "int mult() { return 1; }"
+      lib_name = os.path.join(self.get_dir(), 'libA.c')
+      open(lib_name, 'w').write(lib)
+      a2 = "void x() {}"
+      a2_name = os.path.join(self.get_dir(), 'a2.c')
+      open(a2_name, 'w').write(a2)
+      b2 = "void y() {}"
+      b2_name = os.path.join(self.get_dir(), 'b2.c')
+      open(b2_name, 'w').write(b2)
+      main = r'''
+        #include <stdio.h>
+        int mult();
+        int main() {
+          printf("result: %d\n", mult());
+          return 0;
+        }
+      '''
+      main_name = os.path.join(self.get_dir(), 'main.c')
+      open(main_name, 'w').write(main)
+
+      Building.emcc(lib_name, output_filename='libA.so')
+
+      Building.emcc(a2_name, ['-L.', '-lA'])
+      Building.emcc(b2_name, ['-L.', '-lA'])
+
+      Building.emcc(main_name, ['-L.', '-lA', a2_name+'.o', b2_name+'.o'], output_filename='a.out.js')
+
+      self.assertContained('result: 1', run_js(os.path.join(self.get_dir(), 'a.out.js')))
+
+    def test_multiply_defined_libsymbols_2(self):
+      a = "int x() { return 55; }"
+      a_name = os.path.join(self.get_dir(), 'a.c')
+      open(a_name, 'w').write(a)
+      b = "int y() { return 2; }"
+      b_name = os.path.join(self.get_dir(), 'b.c')
+      open(b_name, 'w').write(b)
+      c = "int z() { return 5; }"
+      c_name = os.path.join(self.get_dir(), 'c.c')
+      open(c_name, 'w').write(c)
+      main = r'''
+        #include <stdio.h>
+        int x();
+        int y();
+        int z();
+        int main() {
+          printf("result: %d\n", x() + y() + z());
+          return 0;
+        }
+      '''
+      main_name = os.path.join(self.get_dir(), 'main.c')
+      open(main_name, 'w').write(main)
+
+      Building.emcc(a_name) # a.c.o
+      Building.emcc(b_name) # b.c.o
+      Building.emcc(c_name) # c.c.o
+      lib_name = os.path.join(self.get_dir(), 'libLIB.a')
+      Building.emar('cr', lib_name, [a_name + '.o', b_name + '.o']) # libLIB.a with a and b
+
+      # a is in the lib AND in an .o, so should be ignored in the lib. We do still need b from the lib though
+      Building.emcc(main_name, ['-L.', '-lLIB', a_name+'.o', c_name + '.o'], output_filename='a.out.js')
+
+      self.assertContained('result: 62', run_js(os.path.join(self.get_dir(), 'a.out.js')))
+
     def test_abspaths(self):
       # Includes with absolute paths are generally dangerous, things like -I/usr/.. will get to system local headers, not our portable ones.
 
diff --git a/tools/shared.py b/tools/shared.py
index e292dbf3..86a5ba77 100644
--- a/tools/shared.py
+++ b/tools/shared.py
@@ -707,6 +707,11 @@ set(CMAKE_FIND_ROOT_PATH_MODE_PACKAGE ONLY)''' % { 'winfix': '' if not WINDOWS e
     return generated_libs
 
   @staticmethod
+  def remove_symbol(filename, symbol):
+    Popen([LLVM_EXTRACT, filename, '-delete', '-glob=' + symbol, '-o', filename], stderr=PIPE).communicate()
+    Popen([LLVM_EXTRACT, filename, '-delete', '-func=' + symbol, '-o', filename], stderr=PIPE).communicate()
+
+  @staticmethod
   def link(files, target, remove_duplicates=False):
     actual_files = []
     unresolved_symbols = set(['main']) # tracking unresolveds is necessary for .a linking, see below. (and main is always a necessary symbol)
@@ -731,6 +736,7 @@ set(CMAKE_FIND_ROOT_PATH_MODE_PACKAGE ONLY)''' % { 'winfix': '' if not WINDOWS e
             os.makedirs(temp_dir)
           os.chdir(temp_dir)
           contents = filter(lambda x: len(x) > 0, Popen([LLVM_AR, 't', f], stdout=PIPE).communicate()[0].split('\n'))
+          #print >> sys.stderr, '  considering archive', f, ':', contents
           if len(contents) == 0:
             print >> sys.stderr, 'Warning: Archive %s appears to be empty (recommendation: link an .so instead of .a)' % f
           else:
@@ -741,20 +747,23 @@ set(CMAKE_FIND_ROOT_PATH_MODE_PACKAGE ONLY)''' % { 'winfix': '' if not WINDOWS e
             Popen([LLVM_AR, 'x', f], stdout=PIPE).communicate() # if absolute paths, files will appear there. otherwise, in this directory
             contents = map(lambda content: os.path.join(temp_dir, content), contents)
             contents = filter(os.path.exists, map(os.path.abspath, contents))
-            needed = False # We add or do not add the entire archive. We let llvm dead code eliminate parts we do not need, instead of
-                           # doing intra-dependencies between archive contents
-            for content in contents:
-              new_symbols = Building.llvm_nm(content)
-              # Link in the .o if it provides symbols, *or* this is a singleton archive (which is apparently an exception in gcc ld)
-              if new_symbols.defs.intersection(unresolved_symbols) or len(files) == 1:
-                needed = True
-            if needed:
+            added_contents = set()
+            added = True
+            while added: # recursively traverse until we have everything we need
+              added = False
               for content in contents:
-                if Building.is_bitcode(content):
-                  new_symbols = Building.llvm_nm(content)
-                  resolved_symbols = resolved_symbols.union(new_symbols.defs)
-                  unresolved_symbols = unresolved_symbols.union(new_symbols.undefs.difference(resolved_symbols)).difference(new_symbols.defs)
-                  actual_files.append(content)
+                if content in added_contents: continue 
+                new_symbols = Building.llvm_nm(content)
+                # Link in the .o if it provides symbols, *or* this is a singleton archive (which is apparently an exception in gcc ld)
+                #print >> sys.stderr, 'need', content, '?', unresolved_symbols, 'and we can supply', new_symbols.defs
+                if new_symbols.defs.intersection(unresolved_symbols) or len(files) == 1:
+                  if Building.is_bitcode(content):
+                    #print >> sys.stderr, '  adding object', content
+                    resolved_symbols = resolved_symbols.union(new_symbols.defs)
+                    unresolved_symbols = unresolved_symbols.union(new_symbols.undefs.difference(resolved_symbols)).difference(new_symbols.defs)
+                    actual_files.append(content)
+                    added_contents.add(content)
+                    added = True
         finally:
           os.chdir(cwd)
     try_delete(target)
@@ -771,8 +780,7 @@ set(CMAKE_FIND_ROOT_PATH_MODE_PACKAGE ONLY)''' % { 'winfix': '' if not WINDOWS e
           print >> sys.stderr, 'emcc: warning: removing duplicates in', actual
           for dupe in dupes:
             print >> sys.stderr, 'emcc: warning: removing duplicate', dupe
-            Popen([LLVM_EXTRACT, actual, '-delete', '-glob=' + dupe, '-o', actual], stderr=PIPE).communicate()
-            Popen([LLVM_EXTRACT, actual, '-delete', '-func=' + dupe, '-o', actual], stderr=PIPE).communicate()
+            Building.remove_symbol(actual, dupe)
           Popen([LLVM_EXTRACT, actual, '-delete', '-glob=.str', '-o', actual], stderr=PIPE).communicate() # garbage that appears here
         seen_symbols = seen_symbols.union(symbols.defs)