New Defects reported by Coverity Scan for LibreOffice

[Stephan Bergmann]

On 08/25/2016 08:56 AM, scan-admin at coverity.com wrote:
> *** CID 1371768:  Control flow issues  (DEADCODE)
> /jvmfwk/plugins/sunmajor/pluginlib/sunjavaplugin.cxx: 760 in jfw_plugin_startJavaVirtualMachine(const JavaInfo *, const JavaVMOption *, int, JavaVM_ **, JNIEnv_ **)()
> 754             if (opt.startsWith("-Xss")) {
> 755                 hasStackSize = true;
> 756             }
> 757             options.push_back(Option(opt, arOptions[i].extraInfo));
> 758         }
> 759         if (addForceInterpreted) {
>>>>     CID 1371768:  Control flow issues  (DEADCODE)
>>>>     Execution cannot reach the expression "options.push_back(jfw_plugin_startJavaVirtualMachine(JavaInfo const *, JavaVMOption const *, int, JavaVM_ **, JNIEnv_ **)::Option (instance 42280)(rtl::OString const("-Xint", rtl::libreoffice_internal::Dummy({})), NULL))" inside this statement: "options.push_back(jfw_plugi...".
> 760             options.push_back(Option("-Xint", nullptr));
> 761         }
> 762         if (!hasStackSize) {
> 763     #if defined LINUX && (defined X86 || defined X86_64)
> 764             // At least OpenJDK 1.8.0's os::workaround_expand_exec_shield_cs_limit
> 765             // (hotspot/src/os_cpu/linux_x86/vm/os_linux_x86.cpp) can mmap an rwx

Was there a way to annotate false positives like this (caused by a 
FALSE_POSITIVE macro being defined as either 0 or 1, depending on 
configuration) in the source code?