a few finalizer testcases, old thread implementation uses now VMThread, no mor contex...

[cacao.git] / src / vm / builtin.c

diff --git a/src/vm/builtin.c b/src/vm/builtin.c
index 6f99656c38b292b4c066fa086a096c729383bf79..ee0e36f2e1568204ab7fc71011e8cd8fa5eac809 100644 (file)
--- a/src/vm/builtin.c
+++ b/src/vm/builtin.c
@@ -34,7 +34,7 @@
    calls instead of machine instructions, using the C calling
    convention.
 
-   $Id: builtin.c 1113 2004-06-02 10:31:09Z twisti $
+   $Id: builtin.c 1173 2004-06-16 14:56:18Z jowenn $
 
 */
 
@@ -1814,7 +1814,10 @@ stacktraceelement *builtin_stacktrace_copy(stacktraceelement **el,stacktraceelem
        size_t s;
        s=(end-begin);
        /*printf ("begin: %p, end: %p, diff: %ld, size :%ld\n",begin,end,s,s*sizeof(stacktraceelement));*/
+       *el=heap_allocate(sizeof(stacktraceelement)*(s+1), true, 0);
+#if 0
        *el=MNEW(stacktraceelement,s+1); /*GC*/
+#endif
        memcpy(*el,begin,(end-begin)*sizeof(stacktraceelement));
        (*el)[s].method=0;
 #if defined(__GNUC__)