<br><br>
<div class="gmail_quote">On Thu, Jun 19, 2008 at 9:21 AM, Andrew Williams <<a href="mailto:andy@tensixtyone.com">andy@tensixtyone.com</a>> wrote:<br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">2008/6/19 Udo van den Heuvel <<a href="mailto:udovdh@xs4all.nl">udovdh@xs4all.nl</a>>:<br>
<div class="Ih2E3d">>> As many others have said, there is a difference between<br>>> virtual (VSZ) and resident (RSS). This is not as simple as<br>>> "swapped out" and "swapped in".<br>
><br>> What does top report?<br>> And even if there's not 400 megs in RAM, why does it get to big?<br>> I still think it is not OK when a process starts at less than 10% grows<br>> to 40% over time.<br>
<br></div>The usage at the top of top reports total memory usage, including<br>buffers and cache etc. In a idea world that should be close to 100% as<br>possible with alot of memory in use as cache. My system is currently<br>
running at 787mb "In Use" with 560mb being used as cache.</blockquote>
<div> </div>
<div>Just his post above alone gives me enough to know that this discussion is a fruitless exercise</div>
<div> </div>
<div>1. He has yet to post his top numbers or any other benchmark of his "issue" despite others here doing so</div>
<div>2. He *believes* there is a problem because a process grows in memory from 10% to 40% over time with no evidence other than that to base his claim on</div>
<div> </div>
<div>I guess I should consider all the processes on my PC to be leaking memory because they all pretty much do that. Just because he can't imagine why a process would do that doesn't make it a bug.</div>
<div> </div>
<div>Kevin</div></div>