how to determine page(cache) boundaries when writing to a file

aKumara · 08-13-2019, 05:06 AM

In linux when writing to a file, kernel maintains multiple in memory pages (4KB in size). Data is first written to the pages and background process bdflush sends these data to disk drive.

Is there a way to determine page boundaries when writing sequentially to a file ?

Can I assume it is always 1-4096 : page 1 and 4097-8192 : page 2 ?

or can it vary ?

say if I start writing from 10 (i.e. first 10 bytes already written to the file previously and I set file position to 10 before start writing) will the page boundary still be

1-4096 : page 1

OR

10-5096 : page 1 ?

Reason for asking,

I can use sync_file_range http://man7.org/linux/man-pages/man2...e_range.2.html to flush data from kernel pages to disk drive in a orderly manner. If I can determine page boundaries I can call sync_file_range only when a page boundary is reached, so that unnecessary sync_file_range calls are avoided.

berndbausch · 08-13-2019, 05:54 AM

Quote:

Originally Posted by aKumara

In linux when writing to a file, kernel maintains multiple in memory pages (4KB in size). Data is first written to the pages and background process bdflush sends these data to disk drive.

Is there a way to determine page boundaries when writing sequentially to a file ?

The page size is usually 4K, but to be certain, run

Code:

$ getconf PAGESIZE
4096

As you assume, pages go from 0 to 4095, 4096 to 8191, 8192 to 12xxx etc. Boundaries are multiples of the page size.

EDIT: Boundaries are multiple of page sizes, but I am not entirely certain whether there is a guarantee that a file always starts at the beginning of a page. By default, this seems to be the case in an ext4 filesystem, but perhaps there are options that allow a file to start somewhere in the middle of a block, if it is small.

A quick test:

Code:

$ echo x >testfile
$ ls -ls testfile
4 -rw-rw-r-- 1 ubuntu ubuntu 2 Aug 13 19:57 testfile

The 4 at the beginning of the last line is the actual space occupied by this file. It's 4K. I conclude that the file starts at a block boundary, therefore at a page boundary in memory.