Linux Kernel Development (10)

The Page Cache and Page Writeback

In this manner, the page cache contains chunks of recently accessed files. During a page I/O operation, such as read(),2 the kernel checks whether the data resides in the
page cache. If the data is in the page cache, the kernel can quickly return the requested page from memory rather than read the data off the comparatively slow disk. In the rest of this chapter, we explore the data structures and kernel facilities that maintain Linux’s page cache.

The address_space Object

The Linux page cache aims to cache any pagebased object, which includes many forms of files and memory mappings.

Radix Tree
Because the kernel must check for the existence of a page in the page cache before initiating any page I/O, such a check must be quick. Otherwise, the overhead of searching and checking the page cache could nullify any benefits the cache might provide.

 

The Buffer Cache

the page cache also reduces disk access during block I/O operations by both caching disk blocks and buffering block I/O operations until later.This caching is often referred to as the buffer cache, although as implemented it is not a separate cache but is part of the page cache.

In earlier kernels, there were two separate disk caches: the page cache and the buffer cache.The former cached pages; the latter cached buffers.

The Flusher Threads

Dirty page writeback occurs in three situations:

These three jobs have rather different goals. In fact, two separate kernel threads performed the work in older kernels (see the following section). In 2.6, however, a gang5 of kernel threads, the flusher threads, performs all three jobs.