泰晓科技 -- 聚焦 Linux - 追本溯源,见微知著!

插上 Linux Lab 实验盘,“秒变” Linux 本本,可即刻开展 Linux 内核开发

泰晓资讯·6月 / 第二期 / 2021

unicornx 创作于 2021/06/12

“泰晓资讯”,广泛报道 “Linux/开源” 业界资讯。欢迎广大读者投递相关资讯来源和素材,本站将进一步收集整理后发布给大家。

  • Linux 内核最新消息


    Linux 5.13 继续火热开发中,Torvalds 在今天的发布公告中指出,虽然这个 -rc5 的大小 “fairly average”,但变更的热度尚未平静下来。 他现在希望随着我们向 5.13-rc6 迈进,改动量会逐渐小一些。

    rc5 中相当大部分的修复来自网络子系统(包括驱动程序和网络核心代码),其他不少补丁修改和架构上的修改有关复(针对 arm64 主要是设备树更新,以及其他对 x86、mips、powerpc 的修改)、其他驱动程序(譬如 GPU 驱动程序的修复也很多,以及与音频、HID、scsi、nvme 等等)。


    stable5.11.22 [EOL]2021-05-19

    关键词: Linux

  • 近期 Linux 内核的重要补丁(节选)


    Core kernel

    关键词: Linux

  • Google 开源了他们的 Fibers 用户空间调度框架

    For a number of years Google has developed Fibers (not to be confused with Google Fiber, their fiber Internet service) as a user-space scheduling framework. While it hasn’t been open-source, the few public papers and talks on Google Fibers has been quite interesting for great performance and a novel design. Finally though Google is working towards open-sourcing Fibers and hoping to get the necessary Linux kernel modifications upstreamed.

    多年来,谷歌一直在开发一款代号为 Fibers 的用户空间调度框架(注意不要与 “Google Fiber”,另一个 Google 的光纤互联网服务项目混淆)。目前,Google 正在努力将 Fibers 开源,并希望将相关的 Linux 内核修改提交给上游。

    近十年来,谷歌一直在开发和使用这个用户空间调度框架,他们希望最终能完全开源该框架。作为其中的一部分,他们正在开发一个 Linux 内核补丁 - 用户管理并发组 ( User-Managed Concurrency Groups, 简称 UMCG) 内核补丁,该补丁实现了该框架在内核端的服务和接口。

    谷歌实际上在 5 月底发送了 UMCG 的 RFC 补丁系列(https://lore.kernel.org/lkml/20210520183614.1227046-1-posk@google.com/),但碰巧直到最近在一些后续讨论中我们才开始注意到该补丁系列。Google 的工程师在补丁的描述中是这样介绍 UMCG 的: “User Managed Concurrency Groups is a fast context switching and in-process userspace scheduling framework. Two main use cases are security sandboxes and userspace scheduling. Security sandboxes: fast X-process context switching will open up a bunch of light-weight security tools, e.g. gVisor, or Tor Project’s Shadow simulator, to more use cases. In-process userspace scheduling is used extensively at Google to provide latency control and isolation guarantees for diverse workloads while maintaining high CPU utilization.” 该补丁提供了如下 API: “Enable UMCG core wait/wake/swap operations as well as UMCG group/server/worker API. The core API is useful for fast IPC and context switching, while the group/server/worker API, together with the core API, form the basis for an in-process M:N userspace scheduling framework implemented in lib/umcg.”

    让我们拭目以待,看看这个 UMCG / Google Fibers 开源工作在未来几个月内将会如何发展。

    新闻出处,”Google Working On Open-Sourcing Their Fibers User-Space Scheduling Framework”: https://www.phoronix.com/scan.php?page=news_item&px=Google-Fibers-Toward-Open

    关键词: Google,Fibers

  • Amazon 致力于为 Linux 内核开发新的主动内存回收机制

    For over a year Amazon engineers have been working on DAMON as a new means of monitoring data accesses under Linux. That patch series has yet to be mainlined but continues being worked on with the intention of getting it upstreamed when ready. More recently the engineers involved have been working on a DAMON-based page reclamation implementation for the Linux kernel in dealing proactively dealing with systems having high memory load.

    一年多以来,Amazon 工程师一直致力于将 DAMON 作为一种在 Linux 下监控数据访问的新方法。该补丁系列尚未进入主线,仍在继续开发过程中,并希望能够相对完备后再将其推向上游。最近,参与的工程师一直致力于为 Linux 内核实现基于 DAMON 的页面回收实现,以主动处理具有高内存负载的系统。

    本周,Amazon 发出了该补丁的第二个“征求意见”稿,该版本实现了基于 DAMON 的内存页面主动回收。也就是说新的方法是利用 DAMON 来监视数据访问以找出不经常被访问的内存页(cold pages),并在此基础上将这些 cold pages 回收掉,补丁系列的解释原话是这样说的,”It finds memory regions that didn’t accessed for specific time duration and page out. To avoid it consuming too much CPU for the paging out operations, the speed limit can be configured. Under the speed limit, it pages out memory regions that didn’t accessed longer first. Also, to remove the monitoring overhead under peaceful situation, and to fall back to the LRU-list based page granularity reclamation when it doesn’t make progress, the three watermarks based activation mechanism is used, with the free memory ratio as the watermark metric. For convenient configuration, it utilizes module parameters. Using these, sysadmins can enable/disable it and tune the coldness identification time threshold, the speed limit, and the three watermarks.”

    这个基于 DAMON 的 Linux 页面回收的性能听起来不错,”DAMON_RECLAIM on v5.12 Linux kernel with ZRAM swap device and 10GB/s speed limit achieves 32% memory saving with only 1.91% runtime overhead. For this, DAMON_RECLAIM consumes only 5.72% of single CPU time. Among the CPU consumption, only about 1.448% of single CPU time is expected to be used for the access pattern monitoring.”

    补丁的更多信息可以参考内核邮件列表 https://lore.kernel.org/lkml/20210608115254.11930-1-sj38.park@gmail.com/

    新闻出处,“Amazon Working On New Proactive Memory Reclamation For The Linux Kernel” https://www.phoronix.com/scan.php?page=news_item&px=DAMON-Page-Reclamation-RFC

    关键词: Amazon, Linux

  • -O3 编译器优化级别仍然被认为对 Linux 内核来说太不安全

    Due to not too old versions of the GNU Compiler Collection (GCC) possibly generating bad code with the “-O3” compiler optimization level and sometimes there not being performance benefits, Linus Torvalds remains against using this optimization flag when compiling the Linux kernel.

    由于新近的 GCC 版本采用 “-O3” 选项编译时仍然可能会生成有问题的代码,并且有时也并没有体现出运行性能上优势,所以 Linus Torvalds 仍然反对在编译 Linux 内核时使用此优化选项。

    本周作为 Linux 5.13 内核网络修复程序的一部分提交了 WireGuard 修复程序,其中删除了使用“-O3”C 标志编译 WireGuard 模块。 自项目启动以来,WireGuard 首席开发人员 Jason Donenfeld 一直在使用 “-O3” 优化级别构建 WireGuard,并在过去研究了生成的代码。 虽然他没有亲眼看到使用 “-O3” 产生错误代码的情况,但由于 Linus Torvalds 不建议使用,所以他放弃了这个选项。

    鉴于许多供应商转向新编译器版本的节奏缓慢,以及内核维护者需要更长的时间才会放弃采用旧编译器版本,Linux 内核可能需要数年时间才能决定默认情况下使用 “-O3” 是否是安全的。

    新闻出处,“-O3 Compiler Optimization Level Still Deemed Too Unsafe For The Linux Kernel” https://www.phoronix.com/scan.php?page=news_item&px=No-O3-For-Linux-Kernel

    关键词: GCC, Linux

  • Intel 有意收购 RISC-V 公司 SiFive

    Back in March during the announcement of Intel Foundry Services it was mentioned that SiFive and Intel were working together to allow RISC-V chips to be fabbed within Intel’s facilities. Additionally, Intel Capital previously invested in SiFive during prior funding rounds. Now it turns out Intel is reportedly positioning to potentially acquire SiFive.

    早在 3 月份,英特尔代工服务的公告中就提到 SiFive 和英特尔正在合作,允许在英特尔制造 RISC-V 芯片。此外,Intel Capital 在之前的几轮融资中也投资了 SiFive。现在,据报道,英特尔正准备收购 SiFive。

    根据彭博社今天发布的一份报告,英特尔已提出以超过 20 亿美元的价格收购 SiFive。该报告未经证实,据说仍处于谈判初期,其他公司可能对收购 SiFive 感兴趣。

    SiFive 的 RISC-V IP 是目前最引人注目的 IP,他们一直致力于通过发布 HiFive Unmatched 等开发板让更多的开发人员投入 RISCV 社区。

    SiFive 由加州大学伯克利分校的研究人员于 2015 年创立。目前,让我们看看 SiFive 是否最终能被英特尔收购吧。

    新闻出处,“Intel Reportedly Interested In Acquiring RISC-V Firm SiFive” https://www.phoronix.com/scan.php?page=news_item&px=Intel-Interested-In-SiFive

    关键词: Intel, SiFive,RISC-V

  • Git 2.32 发布并进行了大量改进

    Version 2.32 of the Git distributed revision control system is now available with a variety of improvements from performance optimizations to usability enhancements and a whole lot of bug fixes.

    Git 分布式修订控制系统的 2.32 版现已推出,其中包含从性能优化到可用性增强以及大量错误修复的各种改进。

    感兴趣的同学可以通过发布公告 (https://lore.kernel.org/lkml/xmqqa6o3xj2e.fsf@gitster.g/T/#u)找到有关新 Git 2.32 版本中所有改进和修复的更多详细信息。

    新闻出处,“Git 2.32 Released With Numerous Improvements” https://www.phoronix.com/scan.php?page=news_item&px=Git-2.32-Released

    关键词: Git


资讯内容部分来自 “LWN.net“。如果您对某些 LWN 文章感兴趣(譬如希望获得全文翻译的)请扫描二维码加微信联系我们:

tinylab wechat

Read Album:

Read Related:

Read Latest: