linux-btrfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Qu Wenruo <quwenruo@cn.fujitsu.com>
To: linux-btrfs@vger.kernel.org
Cc: Wang Xiaoguang <wangxg.fnst@cn.fujitsu.com>
Subject: [PATCH v8 03/27] btrfs: dedupe: Introduce function to add hash into in-memory tree
Date: Tue, 22 Mar 2016 09:35:28 +0800	[thread overview]
Message-ID: <1458610552-9845-4-git-send-email-quwenruo@cn.fujitsu.com> (raw)
In-Reply-To: <1458610552-9845-1-git-send-email-quwenruo@cn.fujitsu.com>

From: Wang Xiaoguang <wangxg.fnst@cn.fujitsu.com>

Introduce static function inmem_add() to add hash into in-memory tree.
And now we can implement the btrfs_dedupe_add() interface.

Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Wang Xiaoguang <wangxg.fnst@cn.fujitsu.com>
---
 fs/btrfs/dedupe.c | 162 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 162 insertions(+)

diff --git a/fs/btrfs/dedupe.c b/fs/btrfs/dedupe.c
index 9a0e03b..013600a 100644
--- a/fs/btrfs/dedupe.c
+++ b/fs/btrfs/dedupe.c
@@ -21,6 +21,25 @@
 #include "transaction.h"
 #include "delayed-ref.h"
 
+struct inmem_hash {
+	struct rb_node hash_node;
+	struct rb_node bytenr_node;
+	struct list_head lru_list;
+
+	u64 bytenr;
+	u32 num_bytes;
+
+	u8 hash[];
+};
+
+static inline struct inmem_hash *inmem_alloc_hash(u16 type)
+{
+	if (WARN_ON(type >= ARRAY_SIZE(btrfs_dedupe_sizes)))
+		return NULL;
+	return kzalloc(sizeof(struct inmem_hash) + btrfs_dedupe_sizes[type],
+			GFP_NOFS);
+}
+
 int btrfs_dedupe_enable(struct btrfs_fs_info *fs_info, u16 type, u16 backend,
 			u64 blocksize, u64 limit_nr)
 {
@@ -95,3 +114,146 @@ out:
 		kfree(dedupe_info);
 	return ret;
 }
+
+static int inmem_insert_hash(struct rb_root *root,
+			     struct inmem_hash *hash, int hash_len)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct inmem_hash *entry = NULL;
+
+	while (*p) {
+		parent = *p;
+		entry = rb_entry(parent, struct inmem_hash, hash_node);
+		if (memcmp(hash->hash, entry->hash, hash_len) < 0)
+			p = &(*p)->rb_left;
+		else if (memcmp(hash->hash, entry->hash, hash_len) > 0)
+			p = &(*p)->rb_right;
+		else
+			return 1;
+	}
+	rb_link_node(&hash->hash_node, parent, p);
+	rb_insert_color(&hash->hash_node, root);
+	return 0;
+}
+
+static int inmem_insert_bytenr(struct rb_root *root,
+			       struct inmem_hash *hash)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct inmem_hash *entry = NULL;
+
+	while (*p) {
+		parent = *p;
+		entry = rb_entry(parent, struct inmem_hash, bytenr_node);
+		if (hash->bytenr < entry->bytenr)
+			p = &(*p)->rb_left;
+		else if (hash->bytenr > entry->bytenr)
+			p = &(*p)->rb_right;
+		else
+			return 1;
+	}
+	rb_link_node(&hash->bytenr_node, parent, p);
+	rb_insert_color(&hash->bytenr_node, root);
+	return 0;
+}
+
+static void __inmem_del(struct btrfs_dedupe_info *dedupe_info,
+			struct inmem_hash *hash)
+{
+	list_del(&hash->lru_list);
+	rb_erase(&hash->hash_node, &dedupe_info->hash_root);
+	rb_erase(&hash->bytenr_node, &dedupe_info->bytenr_root);
+
+	if (!WARN_ON(dedupe_info->current_nr == 0))
+		dedupe_info->current_nr--;
+
+	kfree(hash);
+}
+
+/*
+ * Insert a hash into in-memory dedupe tree
+ * Will remove exceeding last recent use hash.
+ *
+ * If the hash mathced with existing one, we won't insert it, to
+ * save memory
+ */
+static int inmem_add(struct btrfs_dedupe_info *dedupe_info,
+		     struct btrfs_dedupe_hash *hash)
+{
+	int ret = 0;
+	u16 type = dedupe_info->hash_type;
+	struct inmem_hash *ihash;
+
+	ihash = inmem_alloc_hash(type);
+
+	if (!ihash)
+		return -ENOMEM;
+
+	/* Copy the data out */
+	ihash->bytenr = hash->bytenr;
+	ihash->num_bytes = hash->num_bytes;
+	memcpy(ihash->hash, hash->hash, btrfs_dedupe_sizes[type]);
+
+	mutex_lock(&dedupe_info->lock);
+
+	ret = inmem_insert_bytenr(&dedupe_info->bytenr_root, ihash);
+	if (ret > 0) {
+		kfree(ihash);
+		ret = 0;
+		goto out;
+	}
+
+	ret = inmem_insert_hash(&dedupe_info->hash_root, ihash,
+				btrfs_dedupe_sizes[type]);
+	if (ret > 0) {
+		/*
+		 * We only keep one hash in tree to save memory, so if
+		 * hash conflicts, free the one to insert.
+		 */
+		rb_erase(&ihash->bytenr_node, &dedupe_info->bytenr_root);
+		kfree(ihash);
+		ret = 0;
+		goto out;
+	}
+
+	list_add(&ihash->lru_list, &dedupe_info->lru_list);
+	dedupe_info->current_nr++;
+
+	/* Remove the last dedupe hash if we exceed limit */
+	while (dedupe_info->current_nr > dedupe_info->limit_nr) {
+		struct inmem_hash *last;
+
+		last = list_entry(dedupe_info->lru_list.prev,
+				  struct inmem_hash, lru_list);
+		__inmem_del(dedupe_info, last);
+	}
+out:
+	mutex_unlock(&dedupe_info->lock);
+	return 0;
+}
+
+int btrfs_dedupe_add(struct btrfs_trans_handle *trans,
+		     struct btrfs_fs_info *fs_info,
+		     struct btrfs_dedupe_hash *hash)
+{
+	struct btrfs_dedupe_info *dedupe_info = fs_info->dedupe_info;
+
+	if (!fs_info->dedupe_enabled || !hash)
+		return 0;
+
+	if (WARN_ON(dedupe_info == NULL))
+		return -EINVAL;
+
+	if (WARN_ON(!btrfs_dedupe_hash_hit(hash)))
+		return -EINVAL;
+
+	/* ignore old hash */
+	if (dedupe_info->blocksize != hash->num_bytes)
+		return 0;
+
+	if (dedupe_info->backend == BTRFS_DEDUPE_BACKEND_INMEMORY)
+		return inmem_add(dedupe_info, hash);
+	return -EINVAL;
+}
-- 
2.7.3




  parent reply	other threads:[~2016-03-22  1:38 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-03-22  1:35 [PATCH v8 00/27][For 4.7] Btrfs: Add inband (write time) de-duplication framework Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 01/27] btrfs: dedupe: Introduce dedupe framework and its header Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 02/27] btrfs: dedupe: Introduce function to initialize dedupe info Qu Wenruo
2016-03-22  1:35 ` Qu Wenruo [this message]
2016-03-22  1:35 ` [PATCH v8 04/27] btrfs: dedupe: Introduce function to remove hash from in-memory tree Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 05/27] btrfs: delayed-ref: Add support for increasing data ref under spinlock Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 06/27] btrfs: dedupe: Introduce function to search for an existing hash Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 07/27] btrfs: dedupe: Implement btrfs_dedupe_calc_hash interface Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 08/27] btrfs: ordered-extent: Add support for dedupe Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 09/27] btrfs: dedupe: Inband in-memory only de-duplication implement Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 10/27] btrfs: dedupe: Add basic tree structure for on-disk dedupe method Qu Wenruo
2016-03-24 20:58   ` Chris Mason
2016-03-25  1:59     ` Qu Wenruo
2016-03-25 15:11       ` Chris Mason
2016-03-26 13:11         ` Qu Wenruo
2016-03-28 14:09           ` Chris Mason
2016-03-29  1:47             ` Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 11/27] btrfs: dedupe: Introduce interfaces to resume and cleanup dedupe info Qu Wenruo
2016-03-29 17:31   ` Alex Lyakas
2016-03-30  0:26     ` Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 12/27] btrfs: dedupe: Add support for on-disk hash search Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 13/27] btrfs: dedupe: Add support to delete hash for on-disk backend Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 14/27] btrfs: dedupe: Add support for adding " Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 15/27] btrfs: dedupe: Add ioctl for inband dedupelication Qu Wenruo
2016-03-22  2:29   ` kbuild test robot
2016-03-22  2:48   ` kbuild test robot
2016-03-22  1:35 ` [PATCH v8 16/27] btrfs: dedupe: add an inode nodedupe flag Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 17/27] btrfs: dedupe: add a property handler for online dedupe Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 18/27] btrfs: dedupe: add per-file online dedupe control Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 19/27] btrfs: try more times to alloc metadata reserve space Qu Wenruo
2016-04-22 18:06   ` Josef Bacik
2016-04-25  0:54     ` Qu Wenruo
2016-04-25 14:05       ` Josef Bacik
2016-04-26  0:50         ` Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 20/27] btrfs: dedupe: Fix a bug when running inband dedupe with balance Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 21/27] btrfs: Fix a memory leak in inband dedupe hash Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 22/27] btrfs: dedupe: Fix metadata balance error when dedupe is enabled Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 23/27] btrfs: dedupe: Avoid submit IO for hash hit extent Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 24/27] btrfs: dedupe: Preparation for compress-dedupe co-work Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 25/27] btrfs: dedupe: Add support for compression and dedpue Qu Wenruo
2016-03-24 20:35   ` Chris Mason
2016-03-25  1:44     ` Qu Wenruo
2016-03-25 15:12       ` Chris Mason
2016-03-22  1:35 ` [PATCH v8 26/27] btrfs: relocation: Enhance error handling to avoid BUG_ON Qu Wenruo
2016-03-22  1:35 ` [PATCH v8 27/27] btrfs: dedupe: Fix a space cache delalloc bytes underflow bug Qu Wenruo
2016-03-22 13:38 ` [PATCH v8 00/27][For 4.7] Btrfs: Add inband (write time) de-duplication framework David Sterba
2016-03-23  2:25   ` Qu Wenruo
2016-03-24 13:42     ` David Sterba
2016-03-25  1:38       ` Qu Wenruo
2016-04-04 16:55         ` David Sterba
2016-04-05  3:08           ` Qu Wenruo
2016-04-20  2:02             ` Qu Wenruo
2016-04-20 19:14               ` Chris Mason
2016-04-06  3:47           ` Nicholas D Steeves
2016-04-06  5:22             ` Qu Wenruo
2016-04-22 22:14               ` Nicholas D Steeves
2016-04-25  1:25                 ` Qu Wenruo
2016-03-29 17:22 ` Alex Lyakas
2016-03-30  0:34   ` Qu Wenruo
2016-03-30 10:36     ` Alex Lyakas
2016-04-03  8:22     ` Alex Lyakas
2016-04-05  3:51       ` Qu Wenruo

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=1458610552-9845-4-git-send-email-quwenruo@cn.fujitsu.com \
    --to=quwenruo@cn.fujitsu.com \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=wangxg.fnst@cn.fujitsu.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).