mirror of
https://github.com/gatieme/LDD-LinuxDeviceDrivers.git
synced 2026-08-20 22:04:56 +08:00
description: update 20211204
This commit is contained in:
Cheng Jian
@@ -272,7 +272,7 @@ Linux 一开始是在一台i386上的机器开发的, i386 的硬件页表是2
|
||||
|
||||
| 时间 | 作者 | 特性 | 描述 | 是否合入主线 | 链接 |
|
||||
|:----:|:----:|:---:|:----:|:---------:|:----:|
|
||||
| 2021/07/15 | "Matthew Wilcox (Oracle)" <willy@infradead.org> | [Separate struct slab from struct page](https://patchwork.kernel.org/project/linux-mm/cover/20211004134650.4031813-1-willy@infradead.org) | struct page 结构定义中比较复杂的部分之一是 slab 分配器所使用的部分. 一般来说, 如果将 slab 的数据类型从 page 结构体中分离是有好处的, 而且它还有助于防止尾页滑落到任何地方. | v1 ☐ | [PatchWork 00/62](https://patchwork.kernel.org/project/linux-mm/cover/20211004134650.4031813-1-willy@infradead.org) |
|
||||
| 2021/12/01 | "Matthew Wilcox (Oracle)" <willy@infradead.org> | [Separate struct slab from struct page](https://patchwork.kernel.org/project/linux-mm/cover/20211004134650.4031813-1-willy@infradead.org) | struct page 结构定义中比较复杂的部分之一是 slab 分配器所使用的部分. 一般来说, 如果将 slab 的数据类型从 page 结构体中分离是有好处的, 而且它还有助于防止尾页滑落到任何地方. | v2 ☐ | [2021/07/15 PatchWork 00/62](https://patchwork.kernel.org/project/linux-mm/cover/20211004134650.4031813-1-willy@infradead.org)<br>*-*-*-*-*-*-*-* <br>[2021/12/01 PatchWork v2,00/33](https://patchwork.kernel.org/project/linux-mm/cover/20211201181510.18784-1-vbabka@suse.cz) |
|
||||
| 2021/10/12 | Johannes Weiner <hannes@cmpxchg.org> | [PageSlab: eliminate unnecessary compound_head() calls](https://patchwork.kernel.org/project/linux-mm/cover/20211012180148.1669685-1-hannes@cmpxchg.org) | 重构代码, 消除二义性, 使得代码更加简洁. PageSlab() 目前对所有调用站点施加一个 compound_head() 调用, 即使只有极少数情况会遇到尾页. 这组补丁气泡尾分辨率到少数需要它的网站, 并消除它在其他地方. 这个改动很独立, 它的灵感来自于 Willy 的补丁 Separate struct slab from struct page](https://patchwork.kernel.org/project/linux-mm/cover/20210715200030.899216-1-willy@infradead.org). 为了让逻辑更清晰, 代码更简洁. PageSlab() 的调用应该完全从对 compound_head() 的无限制调用中分离出来, 因为它们本身就有不必要的开销. | v1 ☐ | [PatchWork 00/11](https://patchwork.kernel.org/project/linux-mm/cover/20211012180148.1669685-1-hannes@cmpxchg.org), [LKML](https://lkml.org/lkml/2021/10/12/820) |
|
||||
| 2021/11/16 | Vlastimil Babka <vbabka@suse.cz> | [Separate struct slab from struct page](https://patchwork.kernel.org/project/linux-mm/cover/20211116001628.24216-1-vbabka@suse.cz) | NA | RFC ☐ | [PatchWork RFC,00/32](https://patchwork.kernel.org/project/linux-mm/cover/20211116001628.24216-1-vbabka@suse.cz) |
|
||||
|
||||
@@ -3244,6 +3244,7 @@ ARM 引入了一个[内存标签扩展](https://community.arm.com/developer/ip-p
|
||||
| 2020/11/23 | Andrey Konovalov <andreyknvl@google.com> | [kasan: boot parameters for hardware tag-based mode](https://lore.kernel.org/patchwork/cover/1344258) | NA | v4 ☑ [5.11-rc1](https://kernelnewbies.org/Linux_5.11#Memory_management) | [PatchWork mm,v4,00/19](https://patchwork.kernel.org/project/linux-mm/patch/748daf013e17d925b0fe00c1c3b5dce726dd2430.1606162397.git.andreyknvl@google.com) |
|
||||
| 2021/01/15 | Andrey Konovalov <andreyknvl@google.com> | [kasan: HW_TAGS tests support and fixes](https://lore.kernel.org/patchwork/cover/1366086) | NA | v4 ☑ [5.12-rc1](https://kernelnewbies.org/Linux_5.11#Memory_management) | [PatchWork mm,v4,00/15](https://patchwork.kernel.org/project/linux-mm/cover/cover.1610733117.git.andreyknvl@google.com) |
|
||||
| 2021/02/05 | Andrey Konovalov <andreyknvl@google.com> | [kasan: optimizations and fixes for HW_TAGS](https://lore.kernel.org/patchwork/cover/1376340) | NA | v3 ☑ [5.12-rc1](https://kernelnewbies.org/Linux_5.11#Memory_management) | [PatchWork mm,v3,mm,00/13](https://patchwork.kernel.org/project/linux-mm/cover/cover.1612546384.git.andreyknvl@google.com) |
|
||||
| 2021/11/30 | andrey.konovalov@linux.dev | [kasan, vmalloc, arm64: add vmalloc tagging support for SW/HW_TAGS](https://patchwork.kernel.org/project/linux-mm/cover/cover.1638308023.git.andreyknvl@google.com/) | NA | v1 ☐ | [PatchWork 00/31](https://patchwork.kernel.org/project/linux-mm/cover/cover.1638308023.git.andreyknvl@google.com) |
|
||||
|
||||
|
||||
### 13.3.5 KCSAN
|
||||
@@ -3253,7 +3254,7 @@ ARM 引入了一个[内存标签扩展](https://community.arm.com/developer/ip-p
|
||||
|
||||
| 时间 | 作者 | 特性 | 描述 | 是否合入主线 | 链接 |
|
||||
|:----:|:----:|:---:|:----:|:---------:|:----:|
|
||||
| 2020/10/05 | Marco Elver <elver@google.com> | [kcsan: Support detecting a subset of missing memory barriers](https://patchwork.kernel.org/project/linux-mm/cover/20211005105905.1994700-1-elver@google.com) | KCSAN 增加对 LKMM 定义的弱内存子集建模的支持, 它支持检测由于丢失内存障碍而导致的数据竞争子集.<br>当内存操作的结果应该由 barrier 来排序时, KCSAN 可以检测数据竞争, 在这种情况下, 冲突只发生在由于重新排序访问而丢失 barrier 的情况下.<br>KCSAN 检测内存障碍缺失的方法是基于对访问重新排序的建模, 设置了观察点检测对每个内存访问, 也选择在其功能范围内进行模拟重排序. 由于运行时不能"预取"访问, 我们只能对延迟访问效果进行建模, 一旦选择了某个访问进行重新排序, 就会在每次其他访问中检查它, 直到函数范围结束. 如果遇到适当的内存障碍，访问将不再考虑重新排序. | v1 ☐ | [2020/10/05 PatchWork 00/23](https://patchwork.kernel.org/project/linux-mm/cover/20211005105905.1994700-1-elver@google.com)<br>*-*-*-*-*-*-*-* <br>[2020/11/18 PatchWork v2,00/23](https://patchwork.kernel.org/project/linux-mm/cover/20211118081027.3175699-1-elver@google.com) |
|
||||
| 2021/11/30 | Andrey Konovalov <andreyknvl@google.com> | [kcsan: Support detecting a subset of missing memory barriers](https://patchwork.kernel.org/project/linux-mm/cover/20211005105905.1994700-1-elver@google.com) | KCSAN 增加对 LKMM 定义的弱内存子集建模的支持, 它支持检测由于丢失内存障碍而导致的数据竞争子集.<br>当内存操作的结果应该由 barrier 来排序时, KCSAN 可以检测数据竞争, 在这种情况下, 冲突只发生在由于重新排序访问而丢失 barrier 的情况下.<br>KCSAN 检测内存障碍缺失的方法是基于对访问重新排序的建模, 设置了观察点检测对每个内存访问, 也选择在其功能范围内进行模拟重排序. 由于运行时不能"预取"访问, 我们只能对延迟访问效果进行建模, 一旦选择了某个访问进行重新排序, 就会在每次其他访问中检查它, 直到函数范围结束. 如果遇到适当的内存障碍，访问将不再考虑重新排序. | v3 ☐ | [2021/10/05 PatchWork 00/23](https://patchwork.kernel.org/project/linux-mm/cover/20211005105905.1994700-1-elver@google.com)<br>*-*-*-*-*-*-*-* <br>[2021/11/18 PatchWork v2,00/23](https://patchwork.kernel.org/project/linux-mm/cover/20211118081027.3175699-1-elver@google.com)<br>*-*-*-*-*-*-*-* <br>[2021/11/30 PatchWork v3,00/25](https://patchwork.kernel.org/project/linux-mm/cover/20211130114433.2580590-1-elver@google.com) |
|
||||
|
||||
|
||||
|
||||
@@ -3533,7 +3534,12 @@ DAMON 利用两个核心机制 : **基于区域的采样**和**自适应区域
|
||||
| 2017/09/03 | Ard Biesheuvel <ard.biesheuvel@linaro.org> | [implement KASLR for ARM](https://lwn.net/Articles/732891) | ARM 支持 KASLR. | v1 ☐ | [PatchWork v2,0/6](https://git.kernel.org/pub/scm/linux/kernel/git/ardb/linux.git/log/?h=arm-kaslr-latest) |
|
||||
|
||||
|
||||
* 随机函数偏移
|
||||
* 随机函数偏移(FGKASLR)
|
||||
|
||||
[Intel Open-Source Developer Has Been Working On "FGKASLR" For Better Kernel Security](https://www.phoronix.com/scan.php?page=news_item&px=Intel-Linux-FGKASLR-Proposal)
|
||||
|
||||
[Linux 5.16 Has Early Preparations For Supporting FGKASLR
|
||||
](https://www.phoronix.com/scan.php?page=news_item&px=Linux-5.16-Preps-For-FGKASLR)
|
||||
|
||||
| 时间 | 作者 | 特性 | 描述 | 是否合入主线 | 链接 |
|
||||
|:----:|:----:|:---:|:----:|:---------:|:----:|
|
||||
|
||||
@@ -74,6 +74,7 @@
|
||||
|:-----:|:-----:|:-------:|:--------:|
|
||||
| [5.14](https://lwn.net/Articles/867540) | [2021/07/12 rc1](https://lwn.net/Articles/861695)<br>*-*-*-*-*-*-*-* <br>[2021/07/12 5.14 conclusion](https://lwn.net/Articles/861695) | [8月 / 第一期 / 2021](http://tinylab.org/tinylab-weekly-8-1st-2021)<br>*-*-*-*-*-*-*-* <br>[8月 / 第二期 / 2021](http://tinylab.org/tinylab-weekly-8-2nd-2021/)<br>*-*-*-*-*-*-*-* <br>[2021 年 9 月 第 一 期](http://tinylab.org/tinylab-weekly-9-1st-2021) |
|
||||
| [5.15](https://lwn.net/Articles/874283) | [2021/09/02 5.15 Merge window, part 1](https://lwn.net/Articles/867821)<br>*-*-*-*-*-*-*-* <br>[2021/09/13 The rest of the 5.15 merge window](https://lwn.net/Articles/868221) | [2021/07/12 rc1](https://lwn.net/Articles/861695) | [9月 / 第二期 / 2021](http://tinylab.org/tinylab-weekly-9-2nd-2021) | [Linux 5.15 Delivers Many Features](https://www.phoronix.com/scan.php?page=article&item=linux-515-features&num=1) |
|
||||
| [5.16](https://lwn.net/Articles/874283) | [2021/09/02 5.15 Merge window, part 1](https://lwn.net/Articles/867821)<br>*-*-*-*-*-*-*-* <br>[2021/09/13 The rest of the 5.15 merge window](https://lwn.net/Articles/868221) | [2021/07/12 rc1](https://lwn.net/Articles/861695) | [9月 / 第二期 / 2021](http://tinylab.org/tinylab-weekly-9-2nd-2021) | [Linux 5.16-rc1 ](https://www.phoronix.com/scan.php?page=article&item=linux-516-features&num=1) |
|
||||
|
||||
|
||||
# 5 业界会议
|
||||
|
||||
@@ -63,6 +63,7 @@ blogexcerpt: 虚拟化 & KVM 子系统
|
||||
|:----:|:----:|:---:|:----:|:---------:|:----:|
|
||||
| 2021/06/01 | Shaokun Zhang <zhangshaokun@hisilicon.com> | [fs: Optimized file struct to improve performance](https://patchwork.kernel.org/project/linux-fsdevel/patch%20/1622513557-46189-1-git-send-email-zhangshaokun@hisilicon.com) | 通过调整 struct file 中各字段的布局来提升性能. 在系统调用过程中, 经常使用 struct file 结构体中 `f_count`和 `f_mod` 两个[字段](https://patchwork.kernel.org/project/linux-fsdevel/patch/1592987548-8653-1-git-send-email-zhangshaokun@hisilicon.com), 如果我们将它们放在一起, 将能有效地共享同一 cache line, 这对性能非常有用. intel 0-day CI 工程发现该补丁可以提升 UnixBench System Call Overhead 子项的性能, 参见 [aec499039e: unixbench.score 19.2% improvement](https://lkml.org/lkml/2021/4/20/28). | v1 ☐ | [2021/04/09 Patchwork RESEND](https://patchwork.kernel.org/project/linux-fsdevel/patch/1617940057-52843-1-git-send-email-zhangshaokun@hisilicon.com)<br>*-*-*-*-*-*-*-* <br>[2021/06/01 Patchwork RESEND](https://patchwork.kernel.org/project/linux-fsdevel/patch%20/1622513557-46189-1-git-send-email-zhangshaokun@hisilicon.com) |
|
||||
| 2021/08/03 | Peter Oskolkov <posk@google.com> | [5-10% increase in IO latencies with nohz balance patch](https://lkml.org/lkml/2021/11/29/1108) | [7fd7a9e0caba ("sched/fair: Trigger nohz.next_balance updates when a CPU goes NOHZ-idle")](https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=7fd7a9e0caba) 这个补丁导致了 FIO 性能测试劣化. | v1 ☐ | [PatchWork v4,00/10](https://lore.kernel.org/patchwork/cover/1471548) |
|
||||
| 2021/08/03 | Peter Oskolkov <posk@google.com> | [New Linux Scheduler Patches Can Improve AMD Zen Performance For Some Workloads](https://www.phoronix.com/scan.php?page=news_item&px=Linux-Sched-NUMA-Imbalance-Zen) | NA | v1 ☐ | [PatchWork v4,00/10](https://lore.kernel.org/patchwork/cover/1471548) |
|
||||
|
||||
|
||||
# 2 网络
|
||||
|
||||
@@ -1029,7 +1029,15 @@ avg_idle 可以反应目前 RQ 进入 idle 的时间长短.
|
||||
## 5.5 cluster_scheduler
|
||||
-------
|
||||
|
||||
ARM64 机器(如 kunpeng920)和 x86 机器(如 Jacobsville)具有一定的硬件拓扑级别 cluster, 其中一些 CPU 核(通常为 4 核)共享 L3 tag 或二级缓存. 这意味着在 cluster 之间分散这些任务将带来更多的内存带宽并减少缓存争用, 但是打包任务可能有助于减少缓存同步的延迟.
|
||||
|
||||
|
||||
在最后一级缓存中, 可以有多个 cluster, 每个 cluster 下的多个 CPU 有一些共享资源.
|
||||
|
||||
例如 ARM64 机器(如 kunpeng920)和 x86 机器(如 Jacobsville)具有一定的硬件拓扑级别 cluster, 其中一些 CPU 核(通常为 4 核)共享 L3 tag 或二级缓存. cluster 之间分散这些任务将带来更多的内存带宽并减少缓存争用, 但是打包任务可能有助于减少缓存同步的延迟.
|
||||
|
||||
|
||||
在 5.16 中引入了 Cluster Scheduling Domain, 让系统调度域感知 cluster 这一层次. 通过 Cluster Scheduling 可以减少对 cluster 资源 (例如 L2 缓存) 的争用, 从而获得更好的性能.
|
||||
|
||||
|
||||
[Cluster Scheduler Support Queued Ahead Of Linux 5.16](https://www.phoronix.com/scan.php?page=news_item&px=Linux-5.16-Cluster-Scheduler)
|
||||
|
||||
@@ -1048,6 +1056,7 @@ ARM64 机器(如 kunpeng920)和 x86 机器(如 Jacobsville)具有一定的硬件
|
||||
|:----:|:----:|:----:|:---:|:----------:|:---:|
|
||||
| 2021/09/20 | Barry Song <song.bao.hua@hisilicon.com> | [scheduler: expose the topology of clusters and add cluster scheduler](https://lore.kernel.org/patchwork/cover/1415806) | 增加了 cluster 层次的 CPU select. 多个架构都是有 CLUSTER 域的概念的, 比如 Kunpeng 920 一个 NODE(DIE) 24个 CPU 分为 8 个 CLUSTER, 整个 DIE 共享 L3 tag, 但是一个 CLUSTER 使用一个 L3 TAG. 这种情况下对于有数据共享的进程, 在一个 cluster 上运行, 通讯的时延更低. | RFC v6 ☐ | [2020/12/01 PatchWork RFC,v2,0/2](https://patchwork.kernel.org/project/linux-arm-kernel/cover/20201201025944.18260-1-song.bao.hua@hisilicon.com)<br>*-*-*-*-*-*-*-* <br>[2021/03/01 PatchWork RFC,v4,0/4](https://patchwork.kernel.org/project/linux-arm-kernel/cover/20210301225940.16728-1-song.bao.hua@hisilicon.com)<br>*-*-*-*-*-*-*-* <br>[2021/03/19 PatchWork RFC,v5,0/4](https://patchwork.kernel.org/project/linux-arm-kernel/cover/20210319041618.14316-1-song.bao.hua@hisilicon.com)<br>*-*-*-*-*-*-*-* <br>[2021/09/20 PatchWork v6,0/4](https://patchwork.kernel.org/project/linux-arm-kernel/cover/20210420001844.9116-1-song.bao.hua@hisilicon.com) |
|
||||
| 2021/06/15 | Peter Zijlstra | [Represent cluster topology and enable load balance between clusters](https://patchwork.kernel.org/project/linux-arm-kernel/cover/20210820013008.12881-1-21cnbao@gmail.com) | 第一个系列(series): 让拓扑域感知 cluster 的存在, 在 sysfs 接口中提供 cluster 的信息(包括 id 和 cpumask 等), 并添加 CONFIG_SCHED_CLUSTER, 可以在 cluster 之间实现负载平衡, 从而使大量工作负载受益. 测试表明, 在 Jacobsville 上增加 25.1% 的 SPECrate mcf, 在 kunpeng920 上增加 13.574% 的 mcf. 但是社区测试在 alder lake 上造成了一定的性能回归, [Linux 5.16's New Cluster Scheduling Is Causing Regression, Further Hurting Alder Lake](https://www.phoronix.com/scan.php?page=article&item=linux-516-regress&num=1), [Windows 11 Better Than Linux Right Now For Intel Alder Lake Performance](https://www.phoronix.com/scan.php?page=article&item=alderlake-windows-linux&num=1) | RFC ☑ [5.16-rc1](https://lore.kernel.org/lkml/163572864855.3357115.17938524897008353101.tglx@xen13/) | [2021/09/20 PatchWork 0/3](https://patchwork.kernel.org/project/linux-arm-kernel/cover/20210820013008.12881-1-21cnbao@gmail.com), [2021/09/20 PatchWork RESEND,0/3](https://patchwork.kernel.org/project/linux-arm-kernel/cover/20210924085104.44806-1-21cnbao@gmail.com), [LKML](https://lkml.org/lkml/2021/9/24/178), [LWN](https://lwn.net/Articles/866914) |
|
||||
| 2021/12/03 | Tim Chen <tim.c.chen@linux.intel.com> | [Make Cluster Scheduling Configurable](https://lkml.org/lkml/2021/12/3/891 ) | Cluster Scheduling 并不适用于所有场景, 因此这组补丁支持了在运行时和引导时可以动态配置 Cluster Scheduling. 可以通过启动参数 `sched_cluster={1|0}` 来在启动时开启和关闭, 也可以通过 `/proc/sys/kernel/sched_cluster` 接口在运行时动态开启和关闭.<br>当系统负载适中时, 值得做额外的负载平衡来平衡 cluster 之间的负载, 以减少 cluster 内资源的争用. 但是如果系统负载较大, 各个资源已经得到充分利用, cluster 之间的负载平衡不太可能有助于减少 cluster 的资源争用, 因为 cluster 内已经完全繁忙.<br>同时由于不感知性能异构的 CPU 类型, 造成了 Intel Alder Lake CPU 上性能退化, 参见 [Linux 5.16's New Cluster Scheduling Is Causing Regression, Further Hurting Alder Lake](https://www.phoronix.com/scan.php?page=article&item=linux-516-regress&num=3). 因此在 x86 hybrid 类型的 CPU 上禁用 Cluster Scheduling.<br>在一个有 24 个 Atom 内核的 Jacobsville 系统上 (每个 cluster 有 4 个 Atom CPU 核共享一个 L2), 在 24 个 CPU 的系统上运行 mcf 基准测试, 从非常低的负载 1 个基准测试副本到 24 个基准测试副本. 我们看到, 在中等负载时吞吐量得到了提高, 但当系统满负载时, Cluster Scheduling 几乎没有什么提升. | | v1 ☐ | [LORE 0/5](https://lkml.kernel.org/lkml/cover.1638563225.git.tim.c.chen@linux.intel.com) |
|
||||
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user