Hi,
f2fs_dirty_inode just set fi->flag as FI_DIRTY_INODE not to
call update_inode_page. Instead, we do it when f2fs_write_indoe is called.
Do you have any reason to do like this?
How about move update_inode_page from write_inode to dirty_inode?
And we can update inode page when mark_inode_dirty or
Hi Jaeguek,
Thansk for review.
Changes from V2
o rename iradix to iroot
o declare gc_inode_list in gc.h
Changes from V1
o introduce gc_inode_list containing ilist and iradix
o use local variable
o call radix_tree_delete before iput
o retry to add inode into radix_tree
o call list_del expl
On Mon, Dec 01, 2014 at 04:05:20PM +0900, Changman Lee wrote:
> It makes sense to check inode's state than check if
> inode page is dirty or not.
Nice catch.
However, we should leave the original condition, since write_inode can be called
in prior to this fsync call.
And, this is not a proper fix,
Hi,
On Fri, Nov 28, 2014 at 04:12:50PM +0900, Changman Lee wrote:
> Hi Jaegeuk,
>
> I've modified as you mentioned. Review again, please.
>
> Changes from V1
> o introduce gc_inode_list containing ilist and iradix
> o use local variable
> o call radix_tree_delete before iput
> o retry to add
Hi guys,
I was wondering why f2fs_write_inode doesn't submit any I/O according to
wbc->sync_mode.
If you have any idea, answer to my questions, please.
And at f2fs_sync_file,
if (need_cp) {
Q: We've already called sync_fs. Is there any scenario like below ?
I refered to 354a3399dc6f7e55
