<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Recent posts to news</title><link>https://sourceforge.net/p/lmbench/news/</link><description>Recent posts to news</description><atom:link href="https://sourceforge.net/p/lmbench/news/feed.rss" rel="self"/><language>en</language><lastBuildDate>Wed, 11 Apr 2007 07:35:23 -0000</lastBuildDate><atom:link href="https://sourceforge.net/p/lmbench/news/feed.rss" rel="self" type="application/rss+xml"/><item><title>lmbench-3.0-a8 released</title><link>https://sourceforge.net/p/lmbench/news/2007/04/lmbench-30-a8-released/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;LMbench version 3.0 alpha 8 was just released.  It contains the latest bug fixes and patches, and should be considered release candidate 1 (RC1) for LMbench version 3.&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Carl Staelin</dc:creator><pubDate>Wed, 11 Apr 2007 07:35:23 -0000</pubDate><guid isPermaLink="false">https://sourceforge.neteca7a343550fd9b9e229084eed68e6c731a52bc4</guid></item><item><title>lmbench-3.0-a4 released</title><link>https://sourceforge.net/p/lmbench/news/2004/08/lmbench-30-a4-released/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;This is primarily a bug-fix release.  Some buggy areas that were cleaned up include:&lt;br /&gt;
- 32bit vs. 64bit architecture issues&lt;br /&gt;
- automatic memory sizing and size detection&lt;br /&gt;
- per timing interval initialization/cleanup&lt;br /&gt;
- lat_mem_rd strided and random memory access patterns&lt;br /&gt;
- lat_mem_rd memory size now grows exponentially instead of linearly&lt;br /&gt;
- lat_ops / par_ops are now more accurate on more architectures&lt;br /&gt;
- stream used the wrong kernel in one instance&lt;br /&gt;
- automatic configuration of random number generators&lt;br /&gt;
- potential infinite loop in timing harness&lt;/p&gt;
&lt;p&gt;The most significant non-bugfix change is in lat_mem_rd.  It now has two modes: sequential strided access, and random access.  Each access mode is common in real applications, and each is important.  The old lat_mem_rd access pattern (reverse strided access) was a hybrid of the two patterns that was predictable by some hardware and not others and did not really reflect any common memory access pattern.  We feel the new lat_mem_rd is more accurate and better matches common system usage.&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Carl Staelin</dc:creator><pubDate>Tue, 10 Aug 2004 11:26:50 -0000</pubDate><guid isPermaLink="false">https://sourceforge.net4eaab6fe36e9881ced4d5b174cd5454eda433691</guid></item><item><title>lmbench-3.0-a1</title><link>https://sourceforge.net/p/lmbench/news/2002/10/lmbench-30-a1/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;This is the first alpha release of the developer branch, lmbench version 3.0.  It includes a large number of new features, such as the ability to measure performance under scalable load, and new benchmarks such as tlb (# of TLB entries), line (cache line size), stream (lmbench version of John McCalpin's STREAM benchmark), lat_usleep, lat_ops (add/mul/div latency for int/uint64/float/double), par_mem (memory subsystem instruction-level parallelism), par_ops (instruction-level parallelism for many basic operations), and several others.&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Carl Staelin</dc:creator><pubDate>Mon, 28 Oct 2002 13:07:46 -0000</pubDate><guid isPermaLink="false">https://sourceforge.net19fb3b662d3aca9b5a0c5e3c41b83a1648836760</guid></item><item><title>lmbench-2.0.3</title><link>https://sourceforge.net/p/lmbench/news/2002/10/lmbench-203/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;lmbench version 2.0.3 is a patch release to the stable branch, version 2.0.  This supercedes lmbench-2.0, lmbench-2.0-patch1, and lmbench-2.0-patch2.&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Carl Staelin</dc:creator><pubDate>Mon, 28 Oct 2002 13:04:09 -0000</pubDate><guid isPermaLink="false">https://sourceforge.net49ab0208f1d3a5cb2fd37827316b37ae9f7ce28a</guid></item><item><title>lmbench-2.0 patch1</title><link>https://sourceforge.net/p/lmbench/news/2001/07/lmbench-20-patch1/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;I just released the first patch for lmbench-2.0,&lt;br /&gt;
which fixes a bug in the summary result report&lt;br /&gt;
generation.&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Carl Staelin</dc:creator><pubDate>Mon, 23 Jul 2001 10:58:31 -0000</pubDate><guid isPermaLink="false">https://sourceforge.net4a5fa71d8a0f2d901fc945f93a03026116e844b5</guid></item><item><title>lmbench-2.0 release</title><link>https://sourceforge.net/p/lmbench/news/2001/07/lmbench-20-release/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;lmbench-2.0 has finally been released.  I ran out bugs to fix, so I decided to call version 2.0 done.&lt;/p&gt;
&lt;p&gt;Work has been progressing on lmbench-3.0, and the first alpha releases should start hitting the net in the not-too-distant future...  lmbench3.0 includes new benchmarks, such as par_mem which measures the available parallelism in the memory heirarchy, and new capabilities for measuring system scalability by running benchmark processes in parallel in a controlled fashion.&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Carl Staelin</dc:creator><pubDate>Mon, 09 Jul 2001 13:11:06 -0000</pubDate><guid isPermaLink="false">https://sourceforge.net1ed27a9cbed903c89a02fc3225693f3514d62f91</guid></item><item><title>lmbench-2beta1</title><link>https://sourceforge.net/p/lmbench/news/2001/01/lmbench-2beta1/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;lmbench-2beta1 has just been released.  It contains several bugfixes and updated documentation.  If you are running an alpha release, it is important to update to 2beta1.&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Carl Staelin</dc:creator><pubDate>Wed, 10 Jan 2001 07:21:28 -0000</pubDate><guid isPermaLink="false">https://sourceforge.net26b12c73cb22516b34048966ca2a6f43b36793c6</guid></item><item><title>lmbench-2alpha12</title><link>https://sourceforge.net/p/lmbench/news/2000/11/lmbench-2alpha12/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;lmbench-2alpha12 has been released.  This should be the last alpha release for version 2, and is the first patch release in nearly two years.&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Carl Staelin</dc:creator><pubDate>Mon, 13 Nov 2000 09:21:11 -0000</pubDate><guid isPermaLink="false">https://sourceforge.net323a79aa015fa6eea96b1d26445ab95da108c136</guid></item></channel></rss>