QNX 6.6软件开发平台的自适应分区用户指南

需积分: 30 122 浏览量更新于2024-07-17 收藏 694KB PDF 举报

本指南是关于QNX® Software Development Platform 6.6版本的《QNX Adaptive Partitioning User's Guide》。QNX是一款由BlackBerry Limited子公司QNX Software Systems Limited开发的操作系统，专为嵌入式和实时系统设计，尤其是汽车电子和航空电子应用。Adaptive Partitioning是一种关键特性，它允许在多任务环境下动态地管理和分配系统资源，以优化性能并确保关键任务得到优先处理。章节一介绍了什么是Adaptive Partitioning。在这个部分，读者会了解到分区（partition）的概念，即系统如何将硬件资源划分为独立的逻辑区域，每个分区可以运行独立的应用程序或服务。Adaptive Partitioning则是这种划分的增强版本，它可以根据系统负载自动调整各个分区的CPU时间片，确保在低负载时节省资源，而在高负载时能够快速响应。此外，该章节还讨论了系统和用户需求，以及如何通过QNX的线程调度器来实现这些功能。第二章详述了如何使用线程调度器。这部分内容涵盖了如何监控CPU使用情况，包括在不同负载状态下的处理方式：在系统处于“Underload”（空闲或轻载）时，剩余的CPU时间会被分配给非关键任务，以提高整体效率；而在“Freetime”（空闲时段）和“FullLoad”（满载状态）下，调度器会确保关键任务得到优先处理，避免性能瓶颈。总结这一章的内容，用户可以学习到如何根据应用需求调整分区策略，以达到最佳的资源利用率和系统稳定性。这份文档提供了一个全面的指南，帮助开发人员理解和利用QNX Adaptive Partitioning来构建高效、灵活的嵌入式系统，确保在各种工作负载条件下都能保持高性能和响应能力。无论是初学者还是经验丰富的开发者，都能从中获益匪浅，从而提升他们在QNX平台上进行软件开发的能力。

Keeping track of CPU time

The adaptive partitioning thread scheduler throttles CPU usage by measuring the

average CPU usage of each partition. The average is computed over an averaging

window (typically 100 milliseconds), a value that is configurable.

However, the thread scheduler doesn't wait 100 milliseconds to compute the average.

In fact, it calculates it very often. As soon as 1 millisecond passes, the usage for this

1 millisecond is added to the usage of the previous 99 milliseconds to compute the

total CPU usage over the averaging window (i.e. 100 milliseconds).

Time

Averaging window

Current time

Time

Averaging window

Current time

...

Figure 1: The averaging window moves forward as time advances.

The window size defines the averaging time over which the thread scheduler attempts

to balance the partitions to their guaranteed CPU limits. You can set the averaging

window size to any value from 8 milliseconds to 400 milliseconds.

Different choices of the window size affect both the accuracy of load balancing and,

in extreme cases, the maximum delays seen by ready-to-run threads. For more

information, see the Considerations for the Thread Scheduler (p. 39) chapter.

Because the averaging window slides, it can be difficult for you to keep statistics over

a longer period, so the scheduler keeps track of two other windows:

• Window 2 — typically 10 times the window size.

• Window 3 — typically 100 times the window size.

To view the statistics for these additional windows, use the show -v or show -vv

option with the aps command.

The thread scheduler accounts for time spent to the actual fraction of clock ticks used,

and accounts for the time spent in interrupt threads and in the kernel on behalf of

user threads. We refer to this as microbilling.

Microbilling may be approximated on ARM targets if the board can't provide

a micro clock.

Using the Thread Scheduler

This is a consequence of the principle that the thread scheduler chooses between

partitions strictly by priority, as long as no partition is being limited to its budget. This

strategy ensures the most real time behavior.

But, there may be circumstances when you don't want partition Pb to receive all of

the free time just because it has the highest-priority thread. That may occur when,

say, when you choose to use Pb to encapsulate an untrusted or third-party application,

particularly when you are unable control its code.

In that case, you may want to configure the thread scheduler to run a more restrictive

algorithm that divides free time by the budgets of the busy partitions (rather than

assigning all of it to the highest-priority thread). To do so, set the

SCHED_APS_FREETIME_BY_RATIO flag (see “Scheduling policies” in the entry for

SchedCtl() in the QNX Neutrino C Library Reference), or use the aps modify -S

freetime_by_ratio command (see the Utilities Reference).

In our example, freetime-by-ratio mode might cause the aps show command to

display:

+---- CPU Time ----+-- Critical Time --

Partition name id | Budget | Used | Budget | Used

--------------------+------------------+-------------------

System 0 | 70% | 0.04% | 200ms | 0.000ms

Pa 1 | 20% | 65.96% | 0ms | 0.000ms

Pb 2 | 10% | 33.96% | 0ms | 0.000ms

--------------------+------------------+-------------------

Total | 100% | 99.96% |

which indicates that in freetime-by-ratio mode, the thread scheduler divides free time

between partitions Pa and Pb in roughly a 2:1 ratio, which is the ratio of their budgets.

Full Load

Full load occurs when all partitions demand their full budget. A simple way to

demonstrate this is to run while(1) loops in all of the sample partitions. In this

case, the aps show command might display:

+---- CPU Time ----+-- Critical Time --

Partition name id | Budget | Used | Budget | Used

--------------------+------------------+-------------------

System 0 | 70% | 69.80% | 200ms | 0.000ms

Pa 1 | 20% | 19.99% | 0ms | 0.000ms

Pb 2 | 10% | 9.81% | 0ms | 0.000ms

--------------------+------------------+-------------------

Total | 100% | 99.61% |

In this example, the requirement to meet the partitions' guaranteed budgets takes

precedence over priority.

In general, when partitions are at or over their budget, the thread scheduler divides

time between them by the ratios of their budgets, and balances usage to a few

percentage points of the partitions' budgets. (For more information on budget accuracy,

see “ Choosing the window size (p. 43) ” in the Considerations for Scheduling chapter

of this guide.)

How is CPU time divided between partitions?

剩余97页未读，继续阅读

embeddedman

粉丝: 18
资源: 108

QNX 6.6软件开发平台的自适应分区用户指南

QNX虚拟化技术文档

tracelog:基于LogLogger标准库的日志记录系统

Adaptive Partitioning User's Guide

cmake -DCMAKE_TOOLCHAIN_FILE=../qnx_aarch64_toolchain.cmake ..

Main.Func_CanDataProcess.ReadCanData(UniversalData.CanBus1, QNX_canvar.R_205_0, false , this, "setYuanKongJingKong");

Android编译显示QNX_HOST environment variable not set.

The QNX_HOST/QNX_TARGET environment variables must be set

qnx_hyp_rx_dispatch

最新资源