TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Mining JIT traces for missing optimizations with Z3

125 点作者 matt_d11 个月前

4 条评论

adsharma11 个月前
Why not do this at a higher level on the python source itself?<p>I ask because z3 has been used for type inference (Typpete) and for solving equations written in Python.
评论 #41026314 未加载
评论 #41028922 未加载
puzzledobserver11 个月前
How does the PyPy JIT compare to the JIT in the upcoming CPython 3.13?
评论 #41028892 未加载
gus_massa11 个月前
From the article:<p>&gt; <i>(x &amp; c1) | (x &amp; c1) == x &amp; (c1 | c2)</i><p>Is this a typo? Shoud the second <i>c1</i> be <i>c2</i> instead?
评论 #41030228 未加载
aantix11 个月前
Compile enough traces with the accompanying optimizations and you have solid training data for an LLM that can suggest optimizations based on JIT traces.
评论 #41026750 未加载
评论 #41029470 未加载
评论 #41026317 未加载