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。

An Interview with Zen Chief Architect Mike Clark

151 点作者 Smaug123大约 2 个月前

4 条评论

adgjlsfhk1大约 2 个月前
This interview was great! I really wish there were more of these sorts of interviews published.
评论 #43469196 未加载
IshKebab大约 2 个月前
&gt; it also gives us better density overall by having smaller instructions. x86 can put more work into each instruction byte, so we can have denser binaries and increase performance that way.<p>Is that really true? I was under the impression that ARM core density was better than x86.<p>The first Google result I found agrees: <a href="https:&#x2F;&#x2F;www.bitsnbites.eu&#x2F;cisc-vs-risc-code-density&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.bitsnbites.eu&#x2F;cisc-vs-risc-code-density&#x2F;</a><p>&gt; It may take a little bit more microarchitectural work for us to account for stronger memory order on the x86 side, but of course the software has to account for the weaker memory ordering on the ARM side. So there are tradeoffs.<p>I don&#x27;t think this makes sense either since software has to have synchronisation anyway. I don&#x27;t think much software <i>relies</i> on x86 TSO (intentionally anyway; I&#x27;m sure many bugs are covered up by it).<p>Anyway fantastic interview! Very refreshing when they&#x27;re actually technical with interesting questions and answers.
评论 #43489112 未加载
评论 #43486649 未加载
评论 #43486608 未加载
whizzter大约 2 个月前
It&#x27;s an interesting interview but also horribly frustrating.<p>There&#x27;s talk about vectors, longer basic blocks and cache utilization and how they wish programmers used more of it, it misses the real world.<p>Regardless of how &quot;hardcore&quot; programmers feel about it, so much real world executed code is built in JS,etc. The JIT:ed code is super-branchy (to cater for deopt fallbacks) and won&#x27;t use vectors at all.<p>This is something Apple has gotten correct with their vertical integration as they seem to have put more focus on making real-world code go fast. Even most games will have huge swaths of non-vectorized code that would benefit from a more scalar focused way of optimizing.<p>Considering transistor counts in use today, instead of bigger vector units,etc it could be spent on bigger UOP buffers, BTB buffers, bigger caches to eat up branchy-but-linear code flows that are reality for less &quot;optimal&quot; languages).
评论 #43508949 未加载
评论 #43500318 未加载
评论 #43498627 未加载
评论 #43525632 未加载
levodelellis大约 2 个月前
Great interview, more please. Is there a way to submit question? I&#x27;d like to know 1) what affects branch predictors? From my understanding return statements do and cmov does not 2) Why isn&#x27;t there a conditional exception? to replace if (!cond) { __builtin_trap(); }
评论 #43490543 未加载