Re:Re: [PATCH v5 09/16] f2fs: optimize small block size large folio read

[Nanzhe Zhao]

>
>Oops, sorry, I didn't notice there is unnecessary f2fs_truncate_meta_caches()
>in f2fs_wait_on_block_writeback_range(), so still we'd better just use
>f2fs_wait_on_block_writeback here?
>
>    void f2fs_wait_on_block_writeback_range(struct inode *inode, block_t blkaddr,
>                                    block_t len)
>    {
>        struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
>        block_t i;
>
>        if (!f2fs_meta_inode_gc_required(inode))
>            return;
>
>        for (i = 0; i < len; i++)
>            f2fs_wait_on_block_writeback(inode, blkaddr + i);
>
>        f2fs_truncate_meta_caches(sbi, blkaddr, len);  <------------
>    }
>
>Thanks,
>
I have checked the commit history of f2fs_wait_on_block_writeback_range....
It seems f2fs_truncate_meta_caches is used to invalidate old dirty meta data to
prevent data corruption for DIO case. But I think now the helper's name is quite
misleading here.................. We may need a cleanup patch for this future.