From: Qu Wenruo <quwenruo.btrfs@gmx.com>
To: Mark Harmstone <maharmstone@fb.com>, linux-btrfs@vger.kernel.org
Subject: Re: [RFC PATCH 06/10] btrfs: redirect I/O for remapped block groups
Date: Fri, 23 May 2025 19:39:00 +0930 [thread overview]
Message-ID: <d5aeaff6-3e04-4525-861d-36dfa358eb45@gmx.com> (raw)
In-Reply-To: <20250515163641.3449017-7-maharmstone@fb.com>
在 2025/5/16 02:06, Mark Harmstone 写道:
> Change btrfs_map_block() so that if the block group has the REMAPPED
> flag set, we call btrfs_translate_remap() to obtain a new address.
I'm wondering if we can do it a little simpler:
- Delete the chunk item for a fully relocated/remapped chunk
So that future read/write into that logical range will not find a chunk.
- If chunk map lookup failed, search remap tree instead
By this we do not need the REMAPPED flag at all.
Thanks,
Qu
>
> btrfs_translate_remap() searches the remap tree for a range
> corresponding to the logical address passed to btrfs_map_block(). If it
> is within an identity remap, this part of the block group hasn't yet
> been relocated, and so we use the existing address.
>
> If it is within an actual remap, we subtract the start of the remap
> range and add the address of its destination, contained in the item's
> payload.
>
> Signed-off-by: Mark Harmstone <maharmstone@fb.com>
> ---
> fs/btrfs/ctree.c | 11 ++++---
> fs/btrfs/ctree.h | 3 ++
> fs/btrfs/relocation.c | 75 +++++++++++++++++++++++++++++++++++++++++++
> fs/btrfs/relocation.h | 2 ++
> fs/btrfs/volumes.c | 19 +++++++++++
> 5 files changed, 105 insertions(+), 5 deletions(-)
>
> diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
> index a2e7979372cc..7808f7bc2303 100644
> --- a/fs/btrfs/ctree.c
> +++ b/fs/btrfs/ctree.c
> @@ -2331,7 +2331,8 @@ int btrfs_search_old_slot(struct btrfs_root *root, const struct btrfs_key *key,
> * This may release the path, and so you may lose any locks held at the
> * time you call it.
> */
> -static int btrfs_prev_leaf(struct btrfs_root *root, struct btrfs_path *path)
> +int btrfs_prev_leaf(struct btrfs_trans_handle *trans, struct btrfs_root *root,
> + struct btrfs_path *path, int ins_len, int cow)
> {
> struct btrfs_key key;
> struct btrfs_key orig_key;
> @@ -2355,7 +2356,7 @@ static int btrfs_prev_leaf(struct btrfs_root *root, struct btrfs_path *path)
> }
>
> btrfs_release_path(path);
> - ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
> + ret = btrfs_search_slot(trans, root, &key, path, ins_len, cow);
> if (ret <= 0)
> return ret;
>
> @@ -2454,7 +2455,7 @@ int btrfs_search_slot_for_read(struct btrfs_root *root,
> }
> } else {
> if (p->slots[0] == 0) {
> - ret = btrfs_prev_leaf(root, p);
> + ret = btrfs_prev_leaf(NULL, root, p, 0, 0);
> if (ret < 0)
> return ret;
> if (!ret) {
> @@ -5003,7 +5004,7 @@ int btrfs_previous_item(struct btrfs_root *root,
>
> while (1) {
> if (path->slots[0] == 0) {
> - ret = btrfs_prev_leaf(root, path);
> + ret = btrfs_prev_leaf(NULL, root, path, 0, 0);
> if (ret != 0)
> return ret;
> } else {
> @@ -5044,7 +5045,7 @@ int btrfs_previous_extent_item(struct btrfs_root *root,
>
> while (1) {
> if (path->slots[0] == 0) {
> - ret = btrfs_prev_leaf(root, path);
> + ret = btrfs_prev_leaf(NULL, root, path, 0, 0);
> if (ret != 0)
> return ret;
> } else {
> diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
> index 075a06db43a1..90a0d38a31c9 100644
> --- a/fs/btrfs/ctree.h
> +++ b/fs/btrfs/ctree.h
> @@ -721,6 +721,9 @@ static inline int btrfs_next_leaf(struct btrfs_root *root, struct btrfs_path *pa
> return btrfs_next_old_leaf(root, path, 0);
> }
>
> +int btrfs_prev_leaf(struct btrfs_trans_handle *trans, struct btrfs_root *root,
> + struct btrfs_path *path, int ins_len, int cow);
> +
> static inline int btrfs_next_item(struct btrfs_root *root, struct btrfs_path *p)
> {
> return btrfs_next_old_item(root, p, 0);
> diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
> index 02086191630d..e5571c897906 100644
> --- a/fs/btrfs/relocation.c
> +++ b/fs/btrfs/relocation.c
> @@ -3897,6 +3897,81 @@ static const char *stage_to_string(enum reloc_stage stage)
> return "unknown";
> }
>
> +int btrfs_translate_remap(struct btrfs_fs_info *fs_info, u64 *logical,
> + u64 *length)
> +{
> + int ret;
> + struct btrfs_key key, found_key;
> + struct extent_buffer *leaf;
> + struct btrfs_remap *remap;
> + BTRFS_PATH_AUTO_FREE(path);
> +
> + path = btrfs_alloc_path();
> + if (!path)
> + return -ENOMEM;
> +
> + key.objectid = *logical;
> + key.type = BTRFS_IDENTITY_REMAP_KEY;
> + key.offset = 0;
> +
> + ret = btrfs_search_slot(NULL, fs_info->remap_root, &key, path,
> + 0, 0);
> + if (ret < 0)
> + return ret;
> +
> + leaf = path->nodes[0];
> +
> + if (path->slots[0] >= btrfs_header_nritems(leaf)) {
> + ret = btrfs_next_leaf(fs_info->remap_root, path);
> + if (ret < 0)
> + return ret;
> +
> + leaf = path->nodes[0];
> + }
> +
> + btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
> +
> + if (found_key.objectid > *logical) {
> + if (path->slots[0] == 0) {
> + ret = btrfs_prev_leaf(NULL, fs_info->remap_root, path,
> + 0, 0);
> + if (ret) {
> + if (ret == 1)
> + ret = -ENOENT;
> + return ret;
> + }
> +
> + leaf = path->nodes[0];
> + } else {
> + path->slots[0]--;
> + }
> +
> + btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
> + }
> +
> + if (found_key.type != BTRFS_REMAP_KEY &&
> + found_key.type != BTRFS_IDENTITY_REMAP_KEY) {
> + return -ENOENT;
> + }
> +
> + if (found_key.objectid > *logical ||
> + found_key.objectid + found_key.offset <= *logical) {
> + return -ENOENT;
> + }
> +
> + if (*logical + *length > found_key.objectid + found_key.offset)
> + *length = found_key.objectid + found_key.offset - *logical;
> +
> + if (found_key.type == BTRFS_IDENTITY_REMAP_KEY)
> + return 0;
> +
> + remap = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_remap);
> +
> + *logical = *logical - found_key.objectid + btrfs_remap_address(leaf, remap);
> +
> + return 0;
> +}
> +
> /*
> * function to relocate all extents in a block group.
> */
> diff --git a/fs/btrfs/relocation.h b/fs/btrfs/relocation.h
> index 788c86d8633a..f07dbd9a89c6 100644
> --- a/fs/btrfs/relocation.h
> +++ b/fs/btrfs/relocation.h
> @@ -30,5 +30,7 @@ int btrfs_should_cancel_balance(const struct btrfs_fs_info *fs_info);
> struct btrfs_root *find_reloc_root(struct btrfs_fs_info *fs_info, u64 bytenr);
> bool btrfs_should_ignore_reloc_root(const struct btrfs_root *root);
> u64 btrfs_get_reloc_bg_bytenr(const struct btrfs_fs_info *fs_info);
> +int btrfs_translate_remap(struct btrfs_fs_info *fs_info, u64 *logical,
> + u64 *length);
>
> #endif
> diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
> index 77194bb46b40..4777926213c0 100644
> --- a/fs/btrfs/volumes.c
> +++ b/fs/btrfs/volumes.c
> @@ -6620,6 +6620,25 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
> if (IS_ERR(map))
> return PTR_ERR(map);
>
> + if (map->type & BTRFS_BLOCK_GROUP_REMAPPED) {
> + u64 new_logical = logical;
> +
> + ret = btrfs_translate_remap(fs_info, &new_logical, length);
> + if (ret)
> + return ret;
> +
> + if (new_logical != logical) {
> + btrfs_free_chunk_map(map);
> +
> + map = btrfs_get_chunk_map(fs_info, new_logical,
> + *length);
> + if (IS_ERR(map))
> + return PTR_ERR(map);
> +
> + logical = new_logical;
> + }
> + }
> +
> num_copies = btrfs_chunk_map_num_copies(map);
> if (io_geom.mirror_num > num_copies)
> return -EINVAL;
next prev parent reply other threads:[~2025-05-23 10:09 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-15 16:36 [RFC PATCH 00/10] Remap tree Mark Harmstone
2025-05-15 16:36 ` [RFC PATCH 01/10] btrfs: add definitions and constants for remap-tree Mark Harmstone
2025-05-21 12:43 ` Johannes Thumshirn
2025-05-23 13:06 ` Mark Harmstone
2025-05-15 16:36 ` [RFC PATCH 02/10] btrfs: add REMAP chunk type Mark Harmstone
2025-05-15 16:36 ` [RFC PATCH 03/10] btrfs: allow remapped chunks to have zero stripes Mark Harmstone
2025-05-15 16:36 ` [RFC PATCH 04/10] btrfs: add extended version of struct block_group_item Mark Harmstone
2025-05-23 9:53 ` Qu Wenruo
2025-05-23 12:00 ` Mark Harmstone
2025-05-15 16:36 ` [RFC PATCH 05/10] btrfs: allow mounting filesystems with remap-tree incompat flag Mark Harmstone
2025-05-15 16:36 ` [RFC PATCH 06/10] btrfs: redirect I/O for remapped block groups Mark Harmstone
2025-05-23 10:09 ` Qu Wenruo [this message]
2025-05-23 11:53 ` Mark Harmstone
2025-05-15 16:36 ` [RFC PATCH 07/10] btrfs: handle deletions from remapped block group Mark Harmstone
2025-05-15 16:36 ` [RFC PATCH 08/10] btrfs: handle setting up relocation of block group with remap-tree Mark Harmstone
2025-05-15 16:36 ` [RFC PATCH 09/10] btrfs: move existing remaps before relocating block group Mark Harmstone
[not found] ` <202505161726.w1lqCZxG-lkp@intel.com>
2025-05-16 11:43 ` Mark Harmstone
2025-05-15 16:36 ` [RFC PATCH 10/10] btrfs: replace identity maps with actual remaps when doing relocations Mark Harmstone
2025-05-21 0:04 ` Boris Burkov
2025-05-23 14:54 ` Mark Harmstone
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=d5aeaff6-3e04-4525-861d-36dfa358eb45@gmx.com \
--to=quwenruo.btrfs@gmx.com \
--cc=linux-btrfs@vger.kernel.org \
--cc=maharmstone@fb.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox