mirror of
https://github.com/gatieme/LDD-LinuxDeviceDrivers.git
synced 2026-08-18 17:17:39 +08:00
descrption: scheduler
This commit is contained in:
@@ -549,8 +549,8 @@ https://lore.kernel.org/lkml/157476581065.5793.4518979877345136813.stgit@buzz/
|
||||
|
||||
| 时间 | 特性 | 描述 | 是否合入主线 | 链接 |
|
||||
|:----:|:----:|:---:|:------:|:---:|
|
||||
| 2013/07/04 | | [sched: smart wake-affine](https://lore.kernel.org/patchwork/cover/390846) | 引入wakee 翻转次数, 通过巧妙的启发式算法, 识别系统中 1:N/N:M 等唤醒模型, 作为是否进行 wake_affine 的依据 | V3 3.12-rc1 | [PatchWork](https://lore.kernel.org/patchwork/cover/390846)<br>*-*-*-*-*-*-*-* <br>[commit 1](https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=62470419e993f8d9d93db0effd3af4296ecb79a5), [commit2](https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=7d9ffa8961482232d964173cccba6e14d2d543b2) |
|
||||
| | | [Reduce scheduler migrations due to wake_affine](https://lore.kernel.org/patchwork/cover/864391) | 优化 wake_affine 减少迁移次数 | [PatchWork](https://lore.kernel.org/patchwork/cover/864391) |
|
||||
| 2013/07/04 | Michael wang | [sched: smart wake-affine](https://lore.kernel.org/patchwork/cover/390846) | 引入wakee 翻转次数, 通过巧妙的启发式算法, 识别系统中 1:N/N:M 等唤醒模型, 作为是否进行 wake_affine 的依据 | v3 ☑ 3.12-rc1 | [PatchWork](https://lore.kernel.org/patchwork/cover/390846)<br>*-*-*-*-*-*-*-* <br>[commit 1](https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=62470419e993f8d9d93db0effd3af4296ecb79a5), [commit2](https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=7d9ffa8961482232d964173cccba6e14d2d543b2) |
|
||||
| 2017/12/18 | Mel Gorman | [Reduce scheduler migrations due to wake_affine](https://lore.kernel.org/patchwork/cover/864391) | 优化 wake_affine 减少迁移次数 | | [PatchWork](https://lore.kernel.org/patchwork/cover/864391) |
|
||||
| 2020/05/24 | Mel Gorman | [Optimise try_to_wake_up() when wakee is descheduling](https://lore.kernel.org/patchwork/cover/1246560) | 唤醒时如果 wakee 进程正在睡眠或者调度(释放 CPU), 优化在 on_cpu 的自旋等待时间 | v1 ☑ 5.8-rc1 | [PatchWork](https://lore.kernel.org/patchwork/cover/1246560) |
|
||||
|
||||
## 1.4.2 提升 CPU 的查找效率
|
||||
@@ -674,6 +674,16 @@ NUMA 机器一个重要特性就是不同 node 之间的内存访问速度有差
|
||||
| 2020/03/11 | [sched: Streamline select_task_rq() & select_task_rq_fair()](https://lore.kernel.org/patchwork/patch/1208449) | 选核流程上的重构和优化, 当然除此之外还做了其他操作, 比如清理了 sd->flags 信息, 甚至 sysfs 接口都变成只读了 | | |
|
||||
| 2020/03/11 | [sched: Instrument sched domain flags](https://lore.kernel.org/patchwork/cover/1208463) | 基于上一组补丁, 重构了 SD_FLAGS 的定义 | | |
|
||||
|
||||
## 1.5.4
|
||||
-------
|
||||
|
||||
| 时间 | 作者 | 特性 | 描述 | 是否合入主线 | 链接 |
|
||||
|:----:|:----:|:---:|:---:|:----------:|:----:|
|
||||
| 2017/12/18 | Mel Gorman | [Reduce scheduler migrations due to wake_affine](https://lore.kernel.org/patchwork/cover/864391) | | | [PatchWork](https://lore.kernel.org/patchwork/cover/864391)
|
||||
| 2018/01/30 | Mel Gorman | [Reduce migrations and unnecessary spreading of load to multiple CPUs](https://lore.kernel.org/patchwork/cover/878789) | | v1 ☑ 4.16-rc1 | [PatchWork](https://lore.kernel.org/patchwork/cover/878789) |
|
||||
| 2018/02/12 | Mel Gorman | [Reduce migrations due to load imbalance and process exits](https://lore.kernel.org/patchwork/cover/886577) | | v1 ☑ 4.17-rc1 | [PatchWork](https://lore.kernel.org/patchwork/cover/886577) |
|
||||
| 2018/02/12 | Mel Gorman | [Stop wake_affine fighting with automatic NUMA balancing](https://lore.kernel.org/patchwork/cover/886622) | |v1 | [PatchWork](https://lore.kernel.org/patchwork/cover/886622) |
|
||||
| 2018/02/13 | Mel Gorman | [Reduce migrations and conflicts with automatic NUMA balancing v2](https://lore.kernel.org/patchwork/cover/886940) | | v1 | [PatchWork](https://lore.kernel.org/patchwork/cover/886940) |
|
||||
|
||||
|
||||
# 1.6 pick_next_task
|
||||
|
||||
Reference in New Issue
Block a user