Sent a log off list <div>With logging on it does not really want to hang. </div><div>Instead (with ~5% probability) it loops forever. </div><div>I think this is the same bug -- the process misses its own death time...<br><br>

</div><div>--kcc <br><div class="gmail_quote">On Thu, Jan 28, 2010 at 10:40 AM, Julian Seward <span dir="ltr">&lt;<a href="mailto:jseward@acm.org">jseward@acm.org</a>&gt;</span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">

<div><div></div><div class="h5">On Wednesday 27 January 2010, Julian Seward wrote:<br>
&gt; On Wednesday 27 January 2010, Konstantin Serebryany wrote:<br>
&gt; &gt; I&#39;ve minimized the problem to a small test (below).<br>
&gt; &gt; It spawns many threads and doesn&#39;t join them before exiting.<br>
&gt; &gt; It will hang (or loop forever) one out of 40-100 runs:<br>
&gt; &gt; % g++ -g -lpthread hang.cc<br>
&gt; &gt; % for((i=10;i&lt;=99;i++)); do date; time ~/valgrind/trunk/inst/bin/valgrind<br>
&gt; &gt; --tool=none --trace-syscalls=yes --trace-signals=yes  -q ./a.out 2&gt;<br>
&gt; &gt; $i.log ; done<br>
&gt;<br>
&gt; Ok; managed to reproduce it.  2 threads were still stuck in some syscall<br>
&gt; (don&#39;t know which yet).  Investigating.<br>
<br>
</div></div>I can reproduce it, but only in the case where there is no logging,<br>
which isn&#39;t useful.  If you have a logfile where it hangs for<br>
--trace-syscalls=yes --trace-signals=yes, can you compress it and<br>
send it to me?  afaics the log is about 40MB long, but it should<br>
bzip2 nicely.<br>
<font color="#888888"><br>
J<br>
</font></blockquote></div><br></div>