On Jun 22, 2020, at 7:14 AM, Wang Shilong <wangshilong1991@xxxxxxxxx> wrote:
>
> From: Wang Shilong <wshilong@xxxxxxx>
>
> Currently WAS_TRIMMED flag is not persistent, whenever filesystem was
> remounted, fstrim need walk all block groups again, the problem with
> this is FSTRIM could be slow on very large LUN SSD based filesystem.
>
> To avoid this kind of problem, we introduce a block group flag
> EXT4_BG_WAS_TRIMMED, the side effect of this is we need introduce
> extra one block group dirty write after trimming block group.
>
> And When clearing TRIMMED flag, block group will be journalled
> anyway, so it won't introduce any overhead.
>
> Cc: Shuichi Ihara <sihara@xxxxxxx>
> Cc: Andreas Dilger <adilger@xxxxxxxxx>
> Cc: Wang Shilong <wangshilong1991@xxxxxxxxx>
> Signed-off-by: Wang Shilong <wshilong@xxxxxxx>
Reviewed-by: Andreas Dilger <adilger@xxxxxxxxx>
> ---
> v2->v3:
> don't renumber EXT4_GROUP_INFO_* bits.
> v1->v2:
> call ext4_journal_get_write_access() before modify buffer.
> ---
> fs/ext4/ext4.h | 14 +++++------
> fs/ext4/ext4_jbd2.h | 3 ++-
> fs/ext4/mballoc.c | 59 +++++++++++++++++++++++++++++++++------------
> 3 files changed, 53 insertions(+), 23 deletions(-)
>
> diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
> index 42f5060f3cdf..252754da2f1b 100644
> --- a/fs/ext4/ext4.h
> +++ b/fs/ext4/ext4.h
> @@ -368,6 +368,7 @@ struct flex_groups {
> #define EXT4_BG_INODE_UNINIT 0x0001 /* Inode table/bitmap not in use */
> #define EXT4_BG_BLOCK_UNINIT 0x0002 /* Block bitmap not in use */
> #define EXT4_BG_INODE_ZEROED 0x0004 /* On-disk itable initialized to zero */
> +#define EXT4_BG_WAS_TRIMMED 0x0008 /* block group was trimmed */
>
> /*
> * Macro-instructions used to manage group descriptors
> @@ -3138,7 +3139,6 @@ struct ext4_group_info {
> };
>
> #define EXT4_GROUP_INFO_NEED_INIT_BIT 0
> -#define EXT4_GROUP_INFO_WAS_TRIMMED_BIT 1
> #define EXT4_GROUP_INFO_BBITMAP_CORRUPT_BIT 2
> #define EXT4_GROUP_INFO_IBITMAP_CORRUPT_BIT 3
> #define EXT4_GROUP_INFO_BBITMAP_CORRUPT \
> @@ -3153,12 +3153,12 @@ struct ext4_group_info {
> #define EXT4_MB_GRP_IBITMAP_CORRUPT(grp) \
> (test_bit(EXT4_GROUP_INFO_IBITMAP_CORRUPT_BIT, &((grp)->bb_state)))
>
> -#define EXT4_MB_GRP_WAS_TRIMMED(grp) \
> - (test_bit(EXT4_GROUP_INFO_WAS_TRIMMED_BIT, &((grp)->bb_state)))
> -#define EXT4_MB_GRP_SET_TRIMMED(grp) \
> - (set_bit(EXT4_GROUP_INFO_WAS_TRIMMED_BIT, &((grp)->bb_state)))
> -#define EXT4_MB_GRP_CLEAR_TRIMMED(grp) \
> - (clear_bit(EXT4_GROUP_INFO_WAS_TRIMMED_BIT, &((grp)->bb_state)))
> +#define EXT4_MB_GDP_WAS_TRIMMED(gdp) \
> + (gdp->bg_flags & cpu_to_le16(EXT4_BG_WAS_TRIMMED))
> +#define EXT4_MB_GDP_SET_TRIMMED(gdp) \
> + (gdp->bg_flags |= cpu_to_le16(EXT4_BG_WAS_TRIMMED))
> +#define EXT4_MB_GDP_CLEAR_TRIMMED(gdp) \
> + (gdp->bg_flags &= ~cpu_to_le16(EXT4_BG_WAS_TRIMMED))
>
> #define EXT4_MAX_CONTENTION 8
> #define EXT4_CONTENTION_THRESHOLD 2
> diff --git a/fs/ext4/ext4_jbd2.h b/fs/ext4/ext4_jbd2.h
> index 00dc668e052b..a37e438f4b4d 100644
> --- a/fs/ext4/ext4_jbd2.h
> +++ b/fs/ext4/ext4_jbd2.h
> @@ -123,7 +123,8 @@
> #define EXT4_HT_MOVE_EXTENTS 9
> #define EXT4_HT_XATTR 10
> #define EXT4_HT_EXT_CONVERT 11
> -#define EXT4_HT_MAX 12
> +#define EXT4_HT_FS_TRIM 12
> +#define EXT4_HT_MAX 13
>
> /**
> * struct ext4_journal_cb_entry - Base structure for callback information.
> diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
> index c0a331e2feb0..235a316584d0 100644
> --- a/fs/ext4/mballoc.c
> +++ b/fs/ext4/mballoc.c
> @@ -2923,15 +2923,6 @@ static void ext4_free_data_in_buddy(struct super_block *sb,
> rb_erase(&entry->efd_node, &(db->bb_free_root));
> mb_free_blocks(NULL, &e4b, entry->efd_start_cluster, entry->efd_count);
>
> - /*
> - * Clear the trimmed flag for the group so that the next
> - * ext4_trim_fs can trim it.
> - * If the volume is mounted with -o discard, online discard
> - * is supported and the free blocks will be trimmed online.
> - */
> - if (!test_opt(sb, DISCARD))
> - EXT4_MB_GRP_CLEAR_TRIMMED(db);
> -
> if (!db->bb_free_root.rb_node) {
> /* No more items in the per group rb tree
> * balance refcounts from ext4_mb_free_metadata()
> @@ -5084,8 +5075,7 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode,
> " group:%d block:%d count:%lu failed"
> " with %d", block_group, bit, count,
> err);
> - } else
> - EXT4_MB_GRP_CLEAR_TRIMMED(e4b.bd_info);
> + }
>
> ext4_lock_group(sb, block_group);
> mb_clear_bits(bitmap_bh->b_data, bit, count_clusters);
> @@ -5095,6 +5085,14 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode,
> ret = ext4_free_group_clusters(sb, gdp) + count_clusters;
> ext4_free_group_clusters_set(sb, gdp, ret);
> ext4_block_bitmap_csum_set(sb, block_group, gdp, bitmap_bh);
> + /*
> + * Clear the trimmed flag for the group so that the next
> + * ext4_trim_fs can trim it.
> + * If the volume is mounted with -o discard, online discard
> + * is supported and the free blocks will be trimmed online.
> + */
> + if (!test_opt(sb, DISCARD))
> + EXT4_MB_GDP_CLEAR_TRIMMED(gdp);
> ext4_group_desc_csum_set(sb, block_group, gdp);
> ext4_unlock_group(sb, block_group);
>
> @@ -5348,8 +5346,15 @@ ext4_trim_all_free(struct super_block *sb, ext4_group_t group,
> ext4_grpblk_t next, count = 0, free_count = 0;
> struct ext4_buddy e4b;
> int ret = 0;
> + struct ext4_group_desc *gdp;
> + struct buffer_head *gdp_bh;
>
> trace_ext4_trim_all_free(sb, group, start, max);
> + gdp = ext4_get_group_desc(sb, group, &gdp_bh);
> + if (!gdp) {
> + ret = -EIO;
> + return ret;
> + }
>
> ret = ext4_mb_load_buddy(sb, group, &e4b);
> if (ret) {
> @@ -5360,7 +5365,7 @@ ext4_trim_all_free(struct super_block *sb, ext4_group_t group,
> bitmap = e4b.bd_bitmap;
>
> ext4_lock_group(sb, group);
> - if (EXT4_MB_GRP_WAS_TRIMMED(e4b.bd_info) &&
> + if (EXT4_MB_GDP_WAS_TRIMMED(gdp) &&
> minblocks >= atomic_read(&EXT4_SB(sb)->s_last_trim_minblks))
> goto out;
>
> @@ -5399,14 +5404,38 @@ ext4_trim_all_free(struct super_block *sb, ext4_group_t group,
> break;
> }
>
> - if (!ret) {
> + if (!ret)
> ret = count;
> - EXT4_MB_GRP_SET_TRIMMED(e4b.bd_info);
> - }
> out:
> ext4_unlock_group(sb, group);
> ext4_mb_unload_buddy(&e4b);
> + if (ret > 0) {
> + int err;
> + handle_t *handle;
>
> + handle = ext4_journal_start_sb(sb, EXT4_HT_FS_TRIM, 1);
> + if (IS_ERR(handle)) {
> + ret = PTR_ERR(handle);
> + goto out_return;
> + }
> + err = ext4_journal_get_write_access(handle, gdp_bh);
> + if (err) {
> + ret = err;
> + goto out_journal;
> + }
> + ext4_lock_group(sb, group);
> + EXT4_MB_GDP_SET_TRIMMED(gdp);
> + ext4_group_desc_csum_set(sb, group, gdp);
> + ext4_unlock_group(sb, group);
> + err = ext4_handle_dirty_metadata(handle, NULL, gdp_bh);
> + if (err)
> + ret = err;
> +out_journal:
> + err = ext4_journal_stop(handle);
> + if (err)
> + ret = err;
> + }
> +out_return:
> ext4_debug("trimmed %d blocks in the group %d\n",
> count, group);
>
> --
> 2.25.4
>
Cheers, Andreas
Attachment:
signature.asc
Description: Message signed with OpenPGP
