Understanding the Linux Kernel

16.6. Page Swap-In

Swap-in must take place when a process attempts to address a page within its address space that has been swapped out to disk. The "Page fault" exception handler triggers a swap-in operation when the following conditions occur (see Section 7.4.2, Chapter 7):

• The page including the address that caused the exception is a valid one, that is, it belongs to a memory region of the current process.

• The page is not present in memory, that is, the Present flag in the Page Table entry is cleared.

• The Page Table entry associated with the page is not null, which means it contains a swapped-out page identifier.

As described in Section 7.4.3 in Chapter 7, the handle_pte_fault( ) function, invoked by the do_page_fault( ) exception handler, checks whether the Page Table entry is non-null. If so, it invokes do_swap_page( ), which acts on the following parameters:

tsk

Process descriptor address of the process that caused the "Page fault" exception

address

Linear address that caused the exception

vma

Memory region descriptor address of the region that includes address

page_table

Address of the Page Table entry that maps address

entry

Identifier of the swapped-out page

write_access

Flag denoting whether the attempted access was a read or a write

Linux allows each memory region to include a customized function for performing swap-in. A region that needs such a customized function stores a pointer to it in the swapin field of its descriptor. Until recently, IPC shared memory regions ...