Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 873a64c7 authored by Artem Bityutskiy's avatar Artem Bityutskiy
Browse files

UBIFS: amend commentaries 



This patch amends and nicifies commentaries in file.c, as well as
fixes some spelling problems.

Signed-off-by: default avatarArtem Bityutskiy <Artem.Bityutskiy@nokia.com>
parent 0dcd18e4

fs/ubifs/file.c

+30 −32
Original line number Diff line number Diff line
@@ -21,34 +21,32 @@ 
 */



/*

 * This file implements VFS file and inode operations of regular files, device

 * This file implements VFS file and inode operations for regular files, device

 * nodes and symlinks as well as address space operations.

 *

 * UBIFS uses 2 page flags: PG_private and PG_checked. PG_private is set if the

 * page is dirty and is used for budgeting purposes - dirty pages should not be

 * budgeted. The PG_checked flag is set if full budgeting is required for the

 * page e.g., when it corresponds to a file hole or it is just beyond the file

 * size. The budgeting is done in 'ubifs_write_begin()', because it is OK to

 * fail in this function, and the budget is released in 'ubifs_write_end()'. So

 * the PG_private and PG_checked flags carry the information about how the page

 * was budgeted, to make it possible to release the budget properly.

 * UBIFS uses 2 page flags: @PG_private and @PG_checked. @PG_private is set if

 * the page is dirty and is used for optimization purposes - dirty pages are

 * not budgeted so the flag shows that 'ubifs_write_end()' should not release

 * the budget for this page. The @PG_checked flag is set if full budgeting is

 * required for the page e.g., when it corresponds to a file hole or it is

 * beyond the file size. The budgeting is done in 'ubifs_write_begin()', because

 * it is OK to fail in this function, and the budget is released in

 * 'ubifs_write_end()'. So the @PG_private and @PG_checked flags carry

 * information about how the page was budgeted, to make it possible to release

 * the budget properly.

 *

 * A thing to keep in mind: inode's 'i_mutex' is locked in most VFS operations

 * we implement. However, this is not true for '->writepage()', which might be

 * called with 'i_mutex' unlocked. For example, when pdflush is performing

 * write-back, it calls 'writepage()' with unlocked 'i_mutex', although the

 * inode has 'I_LOCK' flag in this case. At "normal" work-paths 'i_mutex' is

 * locked in '->writepage', e.g. in "sys_write -> alloc_pages -> direct reclaim

 * path'. So, in '->writepage()' we are only guaranteed that the page is

 * locked.

 * A thing to keep in mind: inode @i_mutex is locked in most VFS operations we

 * implement. However, this is not true for 'ubifs_writepage()', which may be

 * called with @i_mutex unlocked. For example, when pdflush is doing background

 * write-back, it calls 'ubifs_writepage()' with unlocked @i_mutex. At "normal"

 * work-paths the @i_mutex is locked in 'ubifs_writepage()', e.g. in the

 * "sys_write -> alloc_pages -> direct reclaim path". So, in 'ubifs_writepage()'

 * we are only guaranteed that the page is locked.

 *

 * Similarly, 'i_mutex' does not have to be locked in readpage(), e.g.,

 * readahead path does not have it locked ("sys_read -> generic_file_aio_read

 * -> ondemand_readahead -> readpage"). In case of readahead, 'I_LOCK' flag is

 * not set as well. However, UBIFS disables readahead.

 *

 * This, for example means that there might be 2 concurrent '->writepage()'

 * calls for the same inode, but different inode dirty pages.

 * Similarly, @i_mutex is not always locked in 'ubifs_readpage()', e.g., the

 * read-ahead path does not lock it ("sys_read -> generic_file_aio_read ->

 * ondemand_readahead -> readpage"). In case of readahead, @I_LOCK flag is not

 * set as well. However, UBIFS disables readahead.

 */



#include "ubifs.h"
@@ -449,9 +447,9 @@ static int ubifs_write_begin(struct file *file, struct address_space *mapping, 
			/*

			 * We change whole page so no need to load it. But we

			 * have to set the @PG_checked flag to make the further

			 * code the page is new. This might be not true, but it

			 * is better to budget more that to read the page from

			 * the media.

			 * code know that the page is new. This might be not

			 * true, but it is better to budget more than to read

			 * the page from the media.

			 */

			SetPageChecked(page);

			skipped_read = 1;
@@ -497,8 +495,8 @@ static int ubifs_write_begin(struct file *file, struct address_space *mapping, 
	}



	/*

	 * Whee, we aquired budgeting quickly - without involving

	 * garbage-collection, committing or forceing write-back. We return

	 * Whee, we acquired budgeting quickly - without involving

	 * garbage-collection, committing or forcing write-back. We return

	 * with @ui->ui_mutex locked if we are appending pages, and unlocked

	 * otherwise. This is an optimization (slightly hacky though).

	 */
@@ -562,7 +560,7 @@ static int ubifs_write_end(struct file *file, struct address_space *mapping, 


		/*

		 * Return 0 to force VFS to repeat the whole operation, or the

		 * error code if 'do_readpage()' failes.

		 * error code if 'do_readpage()' fails.

		 */

		copied = do_readpage(page);

		goto out;
@@ -1175,11 +1173,11 @@ static int do_truncation(struct ubifs_info *c, struct inode *inode, 
	ui->ui_size = inode->i_size;

	/* Truncation changes inode [mc]time */

	inode->i_mtime = inode->i_ctime = ubifs_current_time(inode);

	/* The other attributes may be changed at the same time as well */

	/* Other attributes may be changed at the same time as well */

	do_attr_changes(inode, attr);



	err = ubifs_jnl_truncate(c, inode, old_size, new_size);

	mutex_unlock(&ui->ui_mutex);



out_budg:

	if (budgeted)

		ubifs_release_budget(c, &req);