From: Naohiro Aota <naohiro.aota@wdc.com>
To: linux-btrfs@vger.kernel.org, David Sterba <dsterba@suse.com>
Cc: Chris Mason <clm@fb.com>, Josef Bacik <josef@toxicpanda.com>,
Nikolay Borisov <nborisov@suse.com>,
Damien Le Moal <damien.lemoal@wdc.com>,
Johannes Thumshirn <Johannes.Thumshirn@wdc.com>,
Hannes Reinecke <hare@suse.com>,
Anand Jain <anand.jain@oracle.com>,
linux-fsdevel@vger.kernel.org,
Naohiro Aota <naohiro.aota@wdc.com>
Subject: [PATCH v2 21/21] btrfs: factor out prepare_allocation()
Date: Wed, 12 Feb 2020 16:20:48 +0900 [thread overview]
Message-ID: <20200212072048.629856-22-naohiro.aota@wdc.com> (raw)
In-Reply-To: <20200212072048.629856-1-naohiro.aota@wdc.com>
This function finally factor out prepare_allocation() form
find_free_extent(). This function is called before the allocation loop and
a specific allocator function like prepare_allocation_clustered() should
initialize their private information and can set proper hint_byte to
indicate where to start the allocation with.
Signed-off-by: Naohiro Aota <naohiro.aota@wdc.com>
---
fs/btrfs/extent-tree.c | 110 +++++++++++++++++++++++++----------------
1 file changed, 68 insertions(+), 42 deletions(-)
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 3ab0d2f5d718..6da7f6e03767 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -3868,6 +3868,71 @@ static int find_free_extent_update_loop(struct btrfs_fs_info *fs_info,
return -ENOSPC;
}
+static int prepare_allocation_clustered(struct btrfs_fs_info *fs_info,
+ struct find_free_extent_ctl *ffe_ctl,
+ struct btrfs_space_info *space_info,
+ struct btrfs_key *ins)
+{
+ /*
+ * If our free space is heavily fragmented we may not be able to make
+ * big contiguous allocations, so instead of doing the expensive search
+ * for free space, simply return ENOSPC with our max_extent_size so we
+ * can go ahead and search for a more manageable chunk.
+ *
+ * If our max_extent_size is large enough for our allocation simply
+ * disable clustering since we will likely not be able to find enough
+ * space to create a cluster and induce latency trying.
+ */
+ if (unlikely(space_info->max_extent_size)) {
+ spin_lock(&space_info->lock);
+ if (space_info->max_extent_size &&
+ ffe_ctl->num_bytes > space_info->max_extent_size) {
+ ins->offset = space_info->max_extent_size;
+ spin_unlock(&space_info->lock);
+ return -ENOSPC;
+ } else if (space_info->max_extent_size) {
+ ffe_ctl->use_cluster = false;
+ }
+ spin_unlock(&space_info->lock);
+ }
+
+ ffe_ctl->last_ptr = fetch_cluster_info(fs_info, space_info,
+ &ffe_ctl->empty_cluster);
+ if (ffe_ctl->last_ptr) {
+ struct btrfs_free_cluster *last_ptr = ffe_ctl->last_ptr;
+
+ spin_lock(&last_ptr->lock);
+ if (last_ptr->block_group)
+ ffe_ctl->hint_byte = last_ptr->window_start;
+ if (last_ptr->fragmented) {
+ /*
+ * We still set window_start so we can keep track of the
+ * last place we found an allocation to try and save
+ * some time.
+ */
+ ffe_ctl->hint_byte = last_ptr->window_start;
+ ffe_ctl->use_cluster = false;
+ }
+ spin_unlock(&last_ptr->lock);
+ }
+
+ return 0;
+}
+
+static int prepare_allocation(struct btrfs_fs_info *fs_info,
+ struct find_free_extent_ctl *ffe_ctl,
+ struct btrfs_space_info *space_info,
+ struct btrfs_key *ins)
+{
+ switch (ffe_ctl->policy) {
+ case BTRFS_EXTENT_ALLOC_CLUSTERED:
+ return prepare_allocation_clustered(fs_info, ffe_ctl,
+ space_info, ins);
+ default:
+ BUG();
+ }
+}
+
/*
* walks the btree of allocated extents and find a hole of a given size.
* The key ins is changed to record the hole:
@@ -3937,48 +4002,9 @@ static noinline int find_free_extent(struct btrfs_fs_info *fs_info,
return -ENOSPC;
}
- /*
- * If our free space is heavily fragmented we may not be able to make
- * big contiguous allocations, so instead of doing the expensive search
- * for free space, simply return ENOSPC with our max_extent_size so we
- * can go ahead and search for a more manageable chunk.
- *
- * If our max_extent_size is large enough for our allocation simply
- * disable clustering since we will likely not be able to find enough
- * space to create a cluster and induce latency trying.
- */
- if (unlikely(space_info->max_extent_size)) {
- spin_lock(&space_info->lock);
- if (space_info->max_extent_size &&
- num_bytes > space_info->max_extent_size) {
- ins->offset = space_info->max_extent_size;
- spin_unlock(&space_info->lock);
- return -ENOSPC;
- } else if (space_info->max_extent_size) {
- ffe_ctl.use_cluster = false;
- }
- spin_unlock(&space_info->lock);
- }
-
- ffe_ctl.last_ptr = fetch_cluster_info(fs_info, space_info,
- &ffe_ctl.empty_cluster);
- if (ffe_ctl.last_ptr) {
- struct btrfs_free_cluster *last_ptr = ffe_ctl.last_ptr;
-
- spin_lock(&last_ptr->lock);
- if (last_ptr->block_group)
- ffe_ctl.hint_byte = last_ptr->window_start;
- if (last_ptr->fragmented) {
- /*
- * We still set window_start so we can keep track of the
- * last place we found an allocation to try and save
- * some time.
- */
- ffe_ctl.hint_byte = last_ptr->window_start;
- ffe_ctl.use_cluster = false;
- }
- spin_unlock(&last_ptr->lock);
- }
+ ret = prepare_allocation(fs_info, &ffe_ctl, space_info, ins);
+ if (ret < 0)
+ return ret;
ffe_ctl.search_start = max(ffe_ctl.search_start,
first_logical_byte(fs_info, 0));
--
2.25.0
next prev parent reply other threads:[~2020-02-12 7:21 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-12 7:20 [PATCH v2 00/21] btrfs: refactor and generalize chunk/dev_extent/extent allocation Naohiro Aota
2020-02-12 7:20 ` [PATCH v2 01/21] btrfs: change type of full_search to bool Naohiro Aota
2020-02-12 7:20 ` [PATCH v2 02/21] btrfs: do not BUG_ON with invalid profile Naohiro Aota
2020-02-12 7:58 ` Johannes Thumshirn
2020-02-12 14:14 ` Josef Bacik
2020-02-12 7:20 ` [PATCH v2 03/21] btrfs: introduce chunk allocation policy Naohiro Aota
2020-02-12 7:20 ` [PATCH v2 04/21] btrfs: refactor find_free_dev_extent_start() Naohiro Aota
2020-02-12 14:16 ` Josef Bacik
2020-02-13 11:32 ` Johannes Thumshirn
2020-02-12 7:20 ` [PATCH v2 05/21] btrfs: introduce alloc_chunk_ctl Naohiro Aota
2020-02-12 8:00 ` Johannes Thumshirn
2020-02-13 16:17 ` Josef Bacik
2020-02-12 7:20 ` [PATCH v2 06/21] btrfs: factor out init_alloc_chunk_ctl Naohiro Aota
2020-02-13 16:20 ` Josef Bacik
2020-02-12 7:20 ` [PATCH v2 07/21] btrfs: factor out gather_device_info() Naohiro Aota
2020-02-12 8:04 ` Johannes Thumshirn
2020-02-12 7:20 ` [PATCH v2 08/21] btrfs: factor out decide_stripe_size() Naohiro Aota
2020-02-12 7:20 ` [PATCH v2 09/21] btrfs: factor out create_chunk() Naohiro Aota
2020-02-12 8:07 ` Johannes Thumshirn
2020-02-13 16:24 ` Josef Bacik
2020-02-20 10:17 ` Naohiro Aota
2020-02-12 7:20 ` [PATCH v2 10/21] btrfs: parameterize dev_extent_min Naohiro Aota
2020-02-13 16:27 ` Josef Bacik
2020-02-12 7:20 ` [PATCH v2 11/21] btrfs: introduce extent allocation policy Naohiro Aota
2020-02-13 19:55 ` Josef Bacik
2020-02-12 7:20 ` [PATCH v2 12/21] btrfs: move hint_byte into find_free_extent_ctl Naohiro Aota
2020-02-12 14:11 ` Johannes Thumshirn
2020-02-13 19:56 ` Josef Bacik
2020-02-12 7:20 ` [PATCH v2 13/21] btrfs: move vairalbes for clustered allocation " Naohiro Aota
2020-02-12 14:14 ` Johannes Thumshirn
2020-02-13 19:57 ` Josef Bacik
2020-02-20 14:27 ` Christoph Hellwig
2020-02-12 7:20 ` [PATCH v2 14/21] btrfs: factor out do_allocation() Naohiro Aota
2020-02-12 14:45 ` Johannes Thumshirn
2020-02-13 19:57 ` Josef Bacik
2020-02-12 7:20 ` [PATCH v2 15/21] btrfs: drop unnecessary arguments from clustered allocation functions Naohiro Aota
2020-02-12 14:47 ` Johannes Thumshirn
2020-02-13 19:58 ` Josef Bacik
2020-02-12 7:20 ` [PATCH v2 16/21] btrfs: factor out release_block_group() Naohiro Aota
2020-02-12 14:48 ` Johannes Thumshirn
2020-02-13 19:58 ` Josef Bacik
2020-02-12 7:20 ` [PATCH v2 17/21] btrfs: factor out found_extent() Naohiro Aota
2020-02-12 14:50 ` Johannes Thumshirn
2020-02-13 19:58 ` Josef Bacik
2020-02-12 7:20 ` [PATCH v2 18/21] btrfs: drop unnecessary arguments from find_free_extent_update_loop() Naohiro Aota
2020-02-12 14:51 ` Johannes Thumshirn
2020-02-13 19:58 ` Josef Bacik
2020-02-12 7:20 ` [PATCH v2 19/21] btrfs: factor out chunk_allocation_failed() Naohiro Aota
2020-02-12 14:56 ` Johannes Thumshirn
2020-02-13 19:52 ` Josef Bacik
2020-02-20 9:57 ` Naohiro Aota
2020-02-12 7:20 ` [PATCH v2 20/21] btrfs: skip LOOP_NO_EMPTY_SIZE if not clustered allocation Naohiro Aota
2020-02-13 19:55 ` Josef Bacik
2020-02-20 9:56 ` Naohiro Aota
2020-02-20 15:40 ` Josef Bacik
2020-02-12 7:20 ` Naohiro Aota [this message]
2020-02-13 19:59 ` [PATCH v2 21/21] btrfs: factor out prepare_allocation() Josef Bacik
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=20200212072048.629856-22-naohiro.aota@wdc.com \
--to=naohiro.aota@wdc.com \
--cc=Johannes.Thumshirn@wdc.com \
--cc=anand.jain@oracle.com \
--cc=clm@fb.com \
--cc=damien.lemoal@wdc.com \
--cc=dsterba@suse.com \
--cc=hare@suse.com \
--cc=josef@toxicpanda.com \
--cc=linux-btrfs@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=nborisov@suse.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;
as well as URLs for NNTP newsgroup(s).