For various reasons including an impending disk space quota, I need to permanently discontinue the Boost regression tests I'm running for gcc-3.4.3 on SunOS 5.8/SPARC.  This is, unfortunately, one of the release target platforms.  These tests can continue for a couple of more days, but I will need to stop them completely by the end of next week (June 30).
<br><br>On the plus side of the ledger, I have a machine running SunOS 5.10/x86 + gcc-4.1.1 on which I can (and have) start running the regression tests to pick up some of the slack.&nbsp; The results are all-green right now except for two MPL tests:
<br><br>* apply.cpp produces an ICE on gcc 4.1.x and has been reported to the gcc bugzilla (<a href="http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28088">http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28088</a>)<br>* multiset.cpp
 is failing on several platforms, so I think this is not a specific SunOS problem (<b><a href="http://tinyurl.com/r4p6j">http://tinyurl.com/r4p6j</a>)</b><br><br>So, would there be any objections to me marking the gcc-4.1.1_sunos_i86pc
 toolset as &quot;required&quot; and removing gcc-3.4.3_sunos from explicit-failures-markup.xml?<br><br>Is there anyone who can pick up the SunOS/SPARC + gcc testing?&nbsp; Perhaps Doug (OSL4-V2)?<br clear="all"><br>-- <br>Caleb Epstein