2. EL 7
  3. kernel-doc
  4. write_cache_pages(9)

Scroll to navigation

WRITE_CACHE_PAGES(9) Memory Management in Linux WRITE_CACHE_PAGES(9)

NAME

write_cache_pages - walk the list of dirty pages of the given address space and write all of them.

SYNOPSIS

int write_cache_pages(struct address_space * mapping, struct writeback_control * wbc, writepage_t writepage, void * data);

ARGUMENTS

mapping

address space structure to write

wbc

subtract the number of written pages from *wbc->nr_to_write

writepage

function called for each page

data

data passed to writepage function

DESCRIPTION

If a page is already under I/O, write_cache_pages skips it, even if it's dirty. This is desirable behaviour for memory-cleaning writeback, but it is INCORRECT for data-integrity system calls such as fsync. fsync and msync need to guarantee that all the data which was dirty at the time the call was made get new I/O started against them. If wbc->sync_mode is WB_SYNC_ALL then we were called for data integrity and we must wait for existing IO to complete.

To avoid livelocks (when other process dirties new pages), we first tag pages which should be written back with TOWRITE tag and only then start writing them. For data-integrity sync we have to be careful so that we do not miss some pages (e.g., because some other process has cleared TOWRITE tag we set). The rule we follow is that TOWRITE tag can be cleared only by the process clearing the DIRTY tag (and submitting the page for IO).

To avoid deadlocks between range_cyclic writeback and callers that hold pages in PageWriteback to aggregate IO until write_cache_pages returns, we do not loop back to the start of the file. Doing so causes a page lock/page writeback access order inversion - we should only ever lock multiple pages in ascending page->index order, and looping back to the start of the file violates that rule and causes deadlocks.

COPYRIGHT

June 2024 Kernel Hackers Manual 3.10