On 2018/1/25 15:05, Gaoxiang (OS) wrote:
> Previously, we attempt to flush the whole cp pack in a single bio,
> however, when suddenly powering off at this time, we could get into
> an extreme scenario that cp page1 and cp page2 are updated and
> latest, but payload or current summaries are still partially outdated.
> (see reliable write in the UFS specification)
>
> This patch submits the whole cp pack except cp page2 at first,
> and then writes the cp page2 with an extra independent
> bio with pre-io barrier.
>
> Signed-off-by: Gao Xiang <gaoxiang25@xxxxxxxxxx>
> ---
> Change log from v1:
> - Apply the review comments from Chao
> - time data from "finish block_ops" to " finish checkpoint" (tested on ARM64 with TOSHIBA 128GB UFS):
> Before patch: 0.002273 0.001973 0.002789 0.005159 0.002050
> After patch: 0.002502 0.001624 0.002487 0.003049 0.002696
Looks not very stable.
> fs/f2fs/checkpoint.c | 41 ++++++++++++++++++++++++++++++++++++-----
> 1 file changed, 36 insertions(+), 5 deletions(-)
>
> diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
> index 14d2fed..70deb09 100644
> --- a/fs/f2fs/checkpoint.c
> +++ b/fs/f2fs/checkpoint.c
> @@ -300,6 +300,36 @@ static int f2fs_write_meta_pages(struct address_space *mapping,
> return 0;
> }
>
> +static void commit_checkpoint_with_barrier(struct f2fs_sb_info *sbi,
Minor, {commit,sync}_checkpoint will be OK to me, since we will decide to
keep barrier in more deep places according to NOBARRIER mount option,
instead of this function.
> + void *src, block_t blk_addr)
> +{
> + struct writeback_control wbc = {
> + .for_reclaim = 0,
> + };
> + struct page *page = grab_meta_page(sbi, blk_addr);
> + int err;
> +
> + memcpy(page_address(page), src, PAGE_SIZE);
> + set_page_dirty(page);
> +
> + f2fs_wait_on_page_writeback(page, META, true);
> + BUG_ON(PageWriteback(page));
> + if (unlikely(!clear_page_dirty_for_io(page)))
> + BUG();
f2fs_bug_on(sbi, PageWriteback(page));
f2fs_bug_on(sbi, unlikely(!clear_page_dirty_for_io(page)));
> +
> + /* writeout cp page2 */
Please keep defined terms as the same in all places:
s/cp page2/cp pack 2 page
> + err = __f2fs_write_meta_page(page, &wbc, FS_CP_META_IO);
> + if (err) {
> + f2fs_put_page(page, 1);
> + f2fs_stop_checkpoint(sbi, false);
> + return;
How about f2fs_bug_on(sbi, err), since we should not fail in
__f2fs_write_meta_page, right?
> + }
> + f2fs_put_page(page, 0);
> +
> + /* submit checkpoint with barrier */
> + f2fs_submit_merged_write(sbi, META_FLUSH);
> +}
> +
> long sync_meta_pages(struct f2fs_sb_info *sbi, enum page_type type,
> long nr_to_write, enum iostat_type io_type)
> {
> @@ -1297,9 +1327,6 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
> start_blk += NR_CURSEG_NODE_TYPE;
> }
>
> - /* writeout checkpoint block */
> - update_meta_page(sbi, ckpt, start_blk);
> -
> /* wait for previous submitted node/meta pages writeback */
> wait_on_all_pages_writeback(sbi);
>
> @@ -1313,12 +1340,16 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
> sbi->last_valid_block_count = sbi->total_valid_block_count;
> percpu_counter_set(&sbi->alloc_valid_block_count, 0);
>
> - /* Here, we only have one bio having CP pack */
> - sync_meta_pages(sbi, META_FLUSH, LONG_MAX, FS_CP_META_IO);
> + /* Here, we have one bio having CP pack except cp page2 */
> + sync_meta_pages(sbi, META, LONG_MAX, FS_CP_META_IO);
>
> /* wait for previous submitted meta pages writeback */
> wait_on_all_pages_writeback(sbi);
/* wait for previous submitted meta pages writeback */
if (!test_opt(sbi, NOBARRIER)
wait_on_all_pages_writeback(sbi);
Thanks,
>
> + /* barrier and flush checkpoint cp page2 */
> + commit_checkpoint_with_barrier(sbi, ckpt, start_blk);
> + wait_on_all_pages_writeback(sbi);
> +
> release_ino_entry(sbi, false);
>
> if (unlikely(f2fs_cp_error(sbi)))
>
