<div>
Thank you!
</div>
<div><div><br></div><div>igor</div><div><br></div></div>
<p style="color: #A0A0A8;">On Tuesday, December 27, 2011 at 3:21 PM, Vladimir Kozlov wrote:</p>
<blockquote type="cite" style="border-left-style:solid;border-width:1px;margin-left:0px;padding-left:10px;">
<span><div><div><div>Looks good.</div><div><br></div><div>Thanks,</div><div>Vladimir</div><div><br></div><div>Igor Veresov wrote:</div><blockquote type="cite"><div><div> Right, thanks! I've updated the webrev.</div><div><br></div><div>igor</div><div><br></div><div>On Tuesday, December 27, 2011 at 10:40 AM, Vladimir Kozlov wrote:</div><div><br></div><blockquote type="cite"><div><div>You missed changes (additional parameter) in other OS files: bsd, solaris.</div><div><br></div><div>Vladimir</div><div><br></div><div>Igor Veresov wrote:</div><blockquote type="cite"><div><div>The issue is with os::free_memory() changes made in 7060842 when the</div><div>“madvise” call was replaced with “mmap”. This is the reason of</div><div>exhaustive memory consumption. Seems when we need to remap some memory</div><div>we need to do it using the same flags as we mapped it initially,</div><div>particularly with “MAP_HUGETLB”. Otherwise system maps small pages</div><div>instead of already reserved huge pages.</div><div><br></div><div>Contributed by: Aleksey Ignatenko <<a href="mailto:aleksey.v.ignatenko@intel.com">aleksey.v.ignatenko@intel.com</a> </div><div><<a href="mailto:aleksey.v.ignatenko@intel.com">mailto:aleksey.v.ignatenko@intel.com</a>>></div><div><br></div><div>Webrev: <a href="http://cr.openjdk.java.net/~iveresov/7124829/webrev.00/">http://cr.openjdk.java.net/~iveresov/7124829/webrev.00/</a></div><div><br></div><div>igor</div></div></blockquote></div></blockquote></div></blockquote></div></div></span>
</blockquote>
<div>
<br>
</div>