<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic IPP Needs Love in Intel® Integrated Performance Primitives</title>
    <link>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Needs-Love/m-p/1747866#M29222</link>
    <description>The IPP libraries need an update. We need a good generic memcpy() and LZ4 has fallen behind in compression performance, the compressor is no faster than ISA-L deflate nowadays. The doc for LZ4 is not very accurate, what is the acceleration, so we need to call TableInit once or before each call to Encode, etc. A big refresh is in order. Latest version is 2022.3 it cannot be up-to-date after four years.</description>
    <pubDate>Thu, 14 May 2026 05:22:27 GMT</pubDate>
    <dc:creator>LwMQ</dc:creator>
    <dc:date>2026-05-14T05:22:27Z</dc:date>
    <item>
      <title>IPP Needs Love</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Needs-Love/m-p/1747866#M29222</link>
      <description>The IPP libraries need an update. We need a good generic memcpy() and LZ4 has fallen behind in compression performance, the compressor is no faster than ISA-L deflate nowadays. The doc for LZ4 is not very accurate, what is the acceleration, so we need to call TableInit once or before each call to Encode, etc. A big refresh is in order. Latest version is 2022.3 it cannot be up-to-date after four years.</description>
      <pubDate>Thu, 14 May 2026 05:22:27 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Needs-Love/m-p/1747866#M29222</guid>
      <dc:creator>LwMQ</dc:creator>
      <dc:date>2026-05-14T05:22:27Z</dc:date>
    </item>
  </channel>
</rss>

