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。

Core scheduling lands in Linux 5.14

375 点作者 phab超过 3 年前

10 条评论

earthscienceman超过 3 年前
This is such a great post, I&#x27;m surprised it doesn&#x27;t have more traction but maybe people aren&#x27;t into reading longform plain text.<p>&quot;spamming Internet users at scale requires a lot of parallel activity, after all. If those processes can be segregated so that all siblings of any given core run processes from the same customer, we can be spared the gruesome prospect of one spammer stealing another&#x27;s target list — or somebody else&#x27;s private keys.&quot;
评论 #29079444 未加载
评论 #29077959 未加载
评论 #29078043 未加载
评论 #29079836 未加载
ww520超过 3 年前
That&#x27;s very fast in adding the feature to kernel to address the loss of performance due to Sprectre.<p>Assigning OS threads of the same process to the hyperthreads in the same core is a good thing anyway. The threads probably share many data in the process and can benefit from the shared cache in the core.
po1nt超过 3 年前
Everything is cool, except for the naming. At first I thought this was about implementation of Intel&#x27;s new architecture.
tpxl超过 3 年前
Is there not a way to pin processes to cores in linux already, and&#x2F;or why it cannot be used to achieve a similar thing (pin user1 to cores 0,1 and user2 to cores 2,3)?
评论 #29078372 未加载
评论 #29080182 未加载
OldHand2018超过 3 年前
The first link in the linked article mentions Linus saying (in 2019) that performance needs to be better than simply running with SMT disabled.<p>There is no mention of performance though. How is it? Presumably it is better!! And if so, this sounds like a concept that the OpenBSD community would be interested in since they prefer SMT disabled for security reasons.
评论 #29083131 未加载
kevin_thibedeau超过 3 年前
<p><pre><code> int prctl(int option, unsigned long arg2, unsigned long arg3, unsigned long arg4, unsigned long arg5); </code></pre> Is there a reason why a more idiomatic void * to struct wasn&#x27;t used for the args?
评论 #29077459 未加载
评论 #29077974 未加载
评论 #29080458 未加载
评论 #29077485 未加载
jimmyed超过 3 年前
&gt; When one sibling is executing, the other must wait. SMT is useful because CPUs often go idle while waiting for events — usually the arrival of data from memory. While one CPU waits, the other can be executing<p>Is this accurate? I was under the impression SMT gains are not from running other threads when one is blocked (preemption is a old feature) but the processor having a multi stage pipeline so that the net number of instructions that are executed per cycle is more than 1 (closer to 2 in the above example)
评论 #29080716 未加载
评论 #29080142 未加载
评论 #29079800 未加载
jeff_vader超过 3 年前
Random question: does anyone use recent mainline kernels with Ubuntu 20.04? How? How&#x27;s the experience?<p>I&#x27;m currently on Ubuntu HWE line, but that only goes to 5.11.0. Ubuntu kernel devs have debs for [mainline][1], but I&#x27;m not finding any good feedback&#x2F;experience stories about these.<p>[1]: <a href="https:&#x2F;&#x2F;kernel.ubuntu.com&#x2F;~kernel-ppa&#x2F;mainline&#x2F;" rel="nofollow">https:&#x2F;&#x2F;kernel.ubuntu.com&#x2F;~kernel-ppa&#x2F;mainline&#x2F;</a>
exabrial超过 3 年前
&gt; While one might argue that cloud-computing providers are usually grumpy anyway, there is still value in anything that might improve their mood.<p>I&#x27;m dead haha. This is great.
656565656565超过 3 年前
I’m surprised this has made into the mainline kernel, it feels like it is a niche use case. Perhaps cloud provider pressure?