Boost C++ Libraries: Ticket #1491: value_initialized leaves data uninitialized, when using GCC https://svn.boost.org/trac10/ticket/1491 <p> The data of an instance of boost::value_initialized&lt;T&gt; may not be initialized properly, when using GCC. Therefore some tests recently added to <a class="ext-link" href="http://svn.boost.org/trac/boost/browser/trunk/libs/utility/value_init_test.cpp"><span class="icon">​</span>trunk/libs/utility/value_init_test.cpp</a> are expected to fail on GCC test platforms. This is caused by some GCC compiler issues: </p> <p> GCC Bugzilla Bug 30111 <a class="ext-link" href="http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30111"><span class="icon">​</span>Value-initialization of POD base class doesn't initialize members</a>, reported by Jonathan Wakely </p> <p> GCC Bugzilla Bug 33916 <a class="ext-link" href="http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33916"><span class="icon">​</span>Default constructor fails to initialize array members</a>, reported by Michael Elizabeth Chastain </p> <p> Fernando Cacciola and I have already discussed this issue. I hope to have a workaround within a few weeks... Hopefully this workaround will be the same for GCC, Borland and MSVC. See also Ticket <a class="closed ticket" href="https://svn.boost.org/trac10/ticket/1459" title="#1459: Bugs: value_initialized leaves data uninitialized, when using Borland (closed: fixed)">#1459</a> and <a class="closed ticket" href="https://svn.boost.org/trac10/ticket/1217" title="#1217: Bugs: value_initialized leaves data uninitialized, when using MSVC (closed: fixed)">#1217</a>. </p> en-us Boost C++ Libraries /htdocs/site/boost.png https://svn.boost.org/trac10/ticket/1491 Trac 1.4.3 niels_dekker Mon, 10 Dec 2007 10:46:49 GMT <link>https://svn.boost.org/trac10/ticket/1491#comment:1 </link> <guid isPermaLink="false">https://svn.boost.org/trac10/ticket/1491#comment:1</guid> <description> <p> A workaround is provided at the trunk, with changeset <a class="changeset" href="https://svn.boost.org/trac10/changeset/41942" title="value_init.hpp now no longer distinguished between workaround and ...">[41942]</a>. See also the Boost developer mailing list, Re: [utility] value_init workaround fixed [Re: value_init_test failure]. </p> </description> <category>Ticket</category> </item> <item> <dc:creator>niels_dekker</dc:creator> <pubDate>Sun, 06 Jan 2008 09:24:09 GMT</pubDate> <title>status changed; resolution set https://svn.boost.org/trac10/ticket/1491#comment:2 https://svn.boost.org/trac10/ticket/1491#comment:2 <ul> <li><strong>status</strong> <span class="trac-field-old">new</span> → <span class="trac-field-new">closed</span> </li> <li><strong>resolution</strong> → <span class="trac-field-new">fixed</span> </li> </ul> <p> The workaround is merged from the trunk to the release branch, changeset <a class="changeset" href="https://svn.boost.org/trac10/changeset/42501" title="Merged value_init.hpp from trunk [42277] to release branch, thereby ...">[42501]</a>, so the issue should be resolved with the next release of Boost, version 1.35.0. </p> Ticket niels_dekker Sun, 06 Jan 2008 09:30:49 GMT milestone changed https://svn.boost.org/trac10/ticket/1491#comment:3 https://svn.boost.org/trac10/ticket/1491#comment:3 <ul> <li><strong>milestone</strong> <span class="trac-field-old">To Be Determined</span> → <span class="trac-field-new">Boost 1.35.0</span> </li> </ul> Ticket