linux-ext4.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Kemeng Shi <shikemeng@huaweicloud.com>
To: ritesh.list@gmail.com, tytso@mit.edu, adilger.kernel@dilger.ca
Cc: ojaswin@linux.ibm.com, linux-ext4@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: [PATCH v8 04/12] ext4: extend ext4_mb_mark_context to support allocation under journal
Date: Fri, 29 Sep 2023 00:03:59 +0800	[thread overview]
Message-ID: <20230928160407.142069-5-shikemeng@huaweicloud.com> (raw)
In-Reply-To: <20230928160407.142069-1-shikemeng@huaweicloud.com>

Previously, ext4_mb_mark_context is only called under fast commit
replay path, so there is no valid handle when we update block bitmap
and group descriptor. This patch try to extend ext4_mb_mark_context
to be used by code under journal. There are several improvement:
1. Add "handle_t *handle" to struct ext4_mark_context to journal block
bitmap and group descriptor update inside ext4_mb_mark_context (the
added journal code is based on ext4_mb_mark_diskspace_used where
ext4_mb_mark_context is going to be used.)
2. Adds a flag argument to ext4_mb_mark_context() which controls
a. EXT4_MB_BITMAP_MARKED_CHECK - whether block bitmap checking is needed.
b. EXT4_MB_SYNC_UPDATE - whether dirty buffers (bitmap and group
descriptor) needs sync.

Signed-off-by: Kemeng Shi <shikemeng@huaweicloud.com>
Reviewed-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
---
 fs/ext4/mballoc.c | 64 ++++++++++++++++++++++++++++++++++++-----------
 1 file changed, 49 insertions(+), 15 deletions(-)

diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index f50f6fa5e65c..b3e418c0f3d5 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -3953,26 +3953,47 @@ void ext4_exit_mballoc(void)
 	ext4_groupinfo_destroy_slabs();
 }
 
+#define EXT4_MB_BITMAP_MARKED_CHECK 0x0001
+#define EXT4_MB_SYNC_UPDATE 0x0002
 static int
-ext4_mb_mark_context(struct super_block *sb, bool state, ext4_group_t group,
-		     ext4_grpblk_t blkoff, ext4_grpblk_t len)
+ext4_mb_mark_context(handle_t *handle, struct super_block *sb, bool state,
+		     ext4_group_t group, ext4_grpblk_t blkoff,
+		     ext4_grpblk_t len, int flags, ext4_grpblk_t *ret_changed)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	struct buffer_head *bitmap_bh = NULL;
 	struct ext4_group_desc *gdp;
 	struct buffer_head *gdp_bh;
 	int err;
-	unsigned int i, already, changed;
+	unsigned int i, already, changed = len;
 
+	if (ret_changed)
+		*ret_changed = 0;
 	bitmap_bh = ext4_read_block_bitmap(sb, group);
 	if (IS_ERR(bitmap_bh))
 		return PTR_ERR(bitmap_bh);
 
+	if (handle) {
+		BUFFER_TRACE(bitmap_bh, "getting write access");
+		err = ext4_journal_get_write_access(handle, sb, bitmap_bh,
+						    EXT4_JTR_NONE);
+		if (err)
+			goto out_err;
+	}
+
 	err = -EIO;
 	gdp = ext4_get_group_desc(sb, group, &gdp_bh);
 	if (!gdp)
 		goto out_err;
 
+	if (handle) {
+		BUFFER_TRACE(gdp_bh, "get_write_access");
+		err = ext4_journal_get_write_access(handle, sb, gdp_bh,
+						    EXT4_JTR_NONE);
+		if (err)
+			goto out_err;
+	}
+
 	ext4_lock_group(sb, group);
 	if (ext4_has_group_desc_csum(sb) &&
 	    (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT))) {
@@ -3981,12 +4002,14 @@ ext4_mb_mark_context(struct super_block *sb, bool state, ext4_group_t group,
 			ext4_free_clusters_after_init(sb, group, gdp));
 	}
 
-	already = 0;
-	for (i = 0; i < len; i++)
-		if (mb_test_bit(blkoff + i, bitmap_bh->b_data) ==
-				state)
-			already++;
-	changed = len - already;
+	if (flags & EXT4_MB_BITMAP_MARKED_CHECK) {
+		already = 0;
+		for (i = 0; i < len; i++)
+			if (mb_test_bit(blkoff + i, bitmap_bh->b_data) ==
+					state)
+				already++;
+		changed = len - already;
+	}
 
 	if (state) {
 		mb_set_bits(bitmap_bh->b_data, blkoff, len);
@@ -4001,6 +4024,8 @@ ext4_mb_mark_context(struct super_block *sb, bool state, ext4_group_t group,
 	ext4_block_bitmap_csum_set(sb, gdp, bitmap_bh);
 	ext4_group_desc_csum_set(sb, group, gdp);
 	ext4_unlock_group(sb, group);
+	if (ret_changed)
+		*ret_changed = changed;
 
 	if (sbi->s_log_groups_per_flex) {
 		ext4_group_t flex_group = ext4_flex_group(sbi, group);
@@ -4013,15 +4038,17 @@ ext4_mb_mark_context(struct super_block *sb, bool state, ext4_group_t group,
 			atomic64_add(changed, &fg->free_clusters);
 	}
 
-	err = ext4_handle_dirty_metadata(NULL, NULL, bitmap_bh);
+	err = ext4_handle_dirty_metadata(handle, NULL, bitmap_bh);
 	if (err)
 		goto out_err;
-	err = ext4_handle_dirty_metadata(NULL, NULL, gdp_bh);
+	err = ext4_handle_dirty_metadata(handle, NULL, gdp_bh);
 	if (err)
 		goto out_err;
 
-	sync_dirty_buffer(bitmap_bh);
-	sync_dirty_buffer(gdp_bh);
+	if (flags & EXT4_MB_SYNC_UPDATE) {
+		sync_dirty_buffer(bitmap_bh);
+		sync_dirty_buffer(gdp_bh);
+	}
 
 out_err:
 	brelse(bitmap_bh);
@@ -4181,7 +4208,11 @@ void ext4_mb_mark_bb(struct super_block *sb, ext4_fsblk_t block,
 			break;
 		}
 
-		err = ext4_mb_mark_context(sb, state, group, blkoff, clen);
+		err = ext4_mb_mark_context(NULL, sb, state,
+					   group, blkoff, clen,
+					   EXT4_MB_BITMAP_MARKED_CHECK |
+					   EXT4_MB_SYNC_UPDATE,
+					   NULL);
 		if (err)
 			break;
 
@@ -6398,7 +6429,10 @@ static void ext4_free_blocks_simple(struct inode *inode, ext4_fsblk_t block,
 	ext4_grpblk_t blkoff;
 
 	ext4_get_group_no_and_offset(sb, block, &group, &blkoff);
-	ext4_mb_mark_context(sb, false, group, blkoff, count);
+	ext4_mb_mark_context(NULL, sb, false, group, blkoff, count,
+			     EXT4_MB_BITMAP_MARKED_CHECK |
+			     EXT4_MB_SYNC_UPDATE,
+			     NULL);
 }
 
 /**
-- 
2.30.0


  parent reply	other threads:[~2023-09-28  8:05 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-09-28 16:03 [PATCH v8 00/12] cleanups and unit test for mballoc Kemeng Shi
2023-09-28 16:03 ` [PATCH v8 01/12] ext4: make state in ext4_mb_mark_bb to be bool Kemeng Shi
2023-09-28  8:31   ` Ritesh Harjani
2023-09-28 16:03 ` [PATCH v8 02/12] ext4: factor out codes to update block bitmap and group descriptor on disk from ext4_mb_mark_bb Kemeng Shi
2023-09-28 16:03 ` [PATCH v8 03/12] ext4: call ext4_mb_mark_context in ext4_free_blocks_simple Kemeng Shi
2023-09-28 16:03 ` Kemeng Shi [this message]
2023-09-28 16:04 ` [PATCH v8 05/12] ext4: call ext4_mb_mark_context in ext4_mb_mark_diskspace_used Kemeng Shi
2023-09-28 16:04 ` [PATCH v8 06/12] ext4: Separate block bitmap and buddy bitmap freeing in ext4_mb_clear_bb() Kemeng Shi
2023-09-28 16:04 ` [PATCH v8 07/12] ext4: call ext4_mb_mark_context in ext4_mb_clear_bb Kemeng Shi
2023-09-28 16:04 ` [PATCH v8 08/12] ext4: Separate block bitmap and buddy bitmap freeing in ext4_group_add_blocks() Kemeng Shi
2023-09-28 16:04 ` [PATCH v8 09/12] ext4: call ext4_mb_mark_context " Kemeng Shi
2023-09-28 16:04 ` [PATCH v8 10/12] ext4: add some kunit stub for mballoc kunit test Kemeng Shi
2023-09-28 16:04 ` [PATCH v8 11/12] ext4: add first unit test for ext4_mb_new_blocks_simple in mballoc Kemeng Shi
2023-09-28 16:04 ` [PATCH v8 12/12] ext4: run mballoc test with different layouts setting Kemeng Shi
2023-10-06 18:06 ` [PATCH v8 00/12] cleanups and unit test for mballoc Theodore Ts'o

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=20230928160407.142069-5-shikemeng@huaweicloud.com \
    --to=shikemeng@huaweicloud.com \
    --cc=adilger.kernel@dilger.ca \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=ojaswin@linux.ibm.com \
    --cc=ritesh.list@gmail.com \
    --cc=tytso@mit.edu \
    /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).