<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Parameterized Construction of Program Representations for Sparse Dataflow Analyses · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/1403.5952/</link><description>Collected HN comments and submissions linking this paper.</description><language>en</language><lastBuildDate>Mon, 02 Nov 2026 11:16:22 GMT</lastBuildDate><atom:link href="https://hackerpaper.kaipfister.de/p/1403.5952/feed.xml" rel="self" type="application/rss+xml" /><item><title>Parameterized Construction of Program Representations for Sparse Dataflow Analyses — mentioned by DannyBee</title><link>https://news.ycombinator.com/item?id=43356941</link><guid isPermaLink="false">urn:hackerpapers:mention:1403.5952:43356941</guid><pubDate>Thu, 13 Mar 2025 20:28:23 GMT</pubDate><description>&lt;p&gt;I remember implementing this decades ago. Brings me back to see it again :) Your implementation looks nice. You are right that often lots of lattice tricks and such are played in real implementations. If you are worried about speed and memory usage while dealing with context sensitivity, and want to make the lattice taller or wider, incorporate BDD&amp;#x27;s or a similar mechanism. Somewhat pointless for single-functions, but as context-sensitivity grows, you either need BDD&amp;#x27;s or summaries (or some form of auto-sharing) if you want to represent nice large lattices. Lattice value bdd&amp;#x27;s are a thing…&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/1403.5952/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.PL</category></item></channel></rss>
