Here in module cleanup you flush the particular work queue ( which blocks until the handler has completed processing of the work ) and then destroy the work queue .

在模块清理过程中，会清理一些特别的 工作 队列（它们将保持阻塞状态直到处理程序完成对任务的处理），然后销毁工作队列。

The work queue is generally implemented as some sort of linked list with an associated monitor object .

该 工作 队列通常被实现成具有相关监视器对象的某种链表。

You can use a global kernel-global work queue with four functions that address this work queue .

可以使用全局的内核全局 工作 队列，利用4个函数来为工作队列定位。

For this sample solution you only need one ECM In-basket for each of the Work Queues that you created and used when you implemented the process flow .

对于这个样例解决方案，您在实现流程流时创建并使用的每个 Work Queue只需一个ECMIn-basket。

A new request comes in and stays in the new service class work queue .

新的请求传入并滞留在新的服务分类 工作 队列中。

In this case we may not want to queue every incoming request to our work queue because the tasks queued for execution may consume too many system resources and cause resource starvation .

在这种情形下，我们可能不想将每个到来的请求都排队到我们的 工作 队列，因为排在队列中等待执行的任务可能会消耗太多的系统资源并引起资源缺乏。

The following class is a model of a common concurrent utility : a work queue .

下面的类是一种常见的并行实用程序模型： 任务 队列。

The work to be communicated through the work queue is defined by the work_struct structure .

通过 工作 队列与之通信的任务可以由结构体workstruct来定义。

Fork / join addresses the issue of queue contention by having one work queue per thread .

通过让一个线程拥有一个 工作 队列，Fork/join解决了队列争用的问题。

You can find the implementation of the work queue API in . / kernel / workqueue . c with API definitions in . / include / linux / workqueue . h.

在./kernel/workqueue.c中能够找到 工作 队列API的实现方法，API在./include/linux/workqueue.h中定义。

The work queue depth is the number of replications waiting to be processed by the cluster replicator .

工作 队列深度是指等待被集群复制器处理的复制数量。

Source adapter agent-Source adapters are configured to monitor an incoming work queue and pass incoming events to the server .

源适配器代理&源适配器配置为对传入 工作 队列进行监视，并将传入的事件传递给服务器。

As the work_struct indicates the handler function to use you can use the work queue to queue work for a variety of handlers .

由于workstruct中指出了要采用的处理程序函数，因此可以利用 工作 队列来为不同的处理程序进行任务排队。

The work queue API is slightly more complicated that tasklets primarily because a number of options are supported .

工作 队列API比微线程稍复杂，主要是因为它支持很多选项。

If the source adapter agent is a bottleneck events will build up in the work queue that the agent is monitoring .

如果源适配器代理为瓶颈，则事件会在该代理监视的 工作 队列中堆积。

All work on a given work queue can be completed using a call to flush_workqueue .

指定 工作 队列中的所有任务能够通过调用flushworkqueue来完成。

When multiple replicators are used the cluster replication queue and work queue depth times are reduced .

当使用多个集群复制器时，集群复制队列和 工作 队列深度时间都减少了。

The events / X kernel threads ( one per CPU ) extract work from the work queue and activates one of the bottom-half handlers ( as indicated by the handler function in the struct work_struct ) .

events/X内核线程（每CPU一个）从 工作 队列中抽取任务并激活一个bottom-half处理程序（由处理程序函数在结构体workstruct中指定）。

That 's the core of the work queue API .

这就是 工作 队列API的核心。

Thus if the Gearman daemon fails it can recreate the work queue on restart .

这样一来，如果Gearman守护程序故障，它就可以在重启后重新创建这个 工作 队列。

The following example illustrates a few of the core work queue API functions .

下面的例子说明了几个核心的 工作 队列API函数。

Any shared objects ( such as the work queue and work item statuses ) are managed ( and effectively synchronized ) by the Terracotta server .

任何共享对象（比如 工作 队列和工作项状态）都由Terracotta服务器管理（并有效地同步）。

Suppose we wanted to add Business Process management ( BPM ) to the solution such that when the message came in it is stored in some sort of work queue for an editor to approve .

假设我们想在解决方案中添加业务流程管理（BPM），比如在消息传入时，以某种 工作 队列的分类存储该消息，以便编辑器进行核准。

The fork-join framework reduces contention for the work queue by using a technique known as work stealing .

fork-join框架通过一种称作工作窃取（workstealing）的技术减少了 工作 队列的争用情况。

Let 's now continue with a simple example of the work queue API .

下面我们看一个 工作 队列API的简单例子。

On an N-processor system for a work queue that will hold entirely compute-bound tasks you will generally achieve maximum CPU utilization with a thread pool of N or N + 1 threads .

若在一个具有N个处理器的系统上只有一个 工作 队列，其中全部是计算性质的任务，在线程池具有N或N+1个线程时一般会获得最大的CPU利用率。

The Activities act like a work queue where work is assigned and completed .

这些活动像一个 工作 队列，工作在这里被分配和完成。

To flush the kernel-global work queue call flush_scheduled_work .

为了清理内核全局 工作 队列，可调用flushscheduledwork。

The work queue mechanism is inherited from the WLM queuing manager service .

工作 队列机制是从WLM队列管理器服务继承而来的。

What we usually want is a work queue combined with a fixed group of worker threads which uses wait () and notify () to signal waiting threads that new work has arrived .

我们通常想要的是同一组固定的工作线程相结合的 工作 队列，它使用wait（）和notify（）来通知等待线程新的工作已经到达了。