From b21ebf143af219207214c79bc217beb39c43212a Mon Sep 17 00:00:00 2001 From: Harshad Shirwadkar <harshadshirwadkar@gmail.com> Date: Thu, 5 Nov 2020 19:58:51 -0800 Subject: ext4: mark fc ineligible if inode gets evictied due to mem pressure If inode gets evicted due to memory pressure, we have to remove it from the fast commit list. However, that inode may have uncommitted changes that fast commits will lose. So, just fall back to full commits in this case. Also, rename the fast commit ineligiblity reason from "EXT4_FC_REASON_MEM" to "EXT4_FC_REASON_MEM_NOMEM" for better expression. Suggested-by: Jan Kara <jack@suse.cz> Signed-off-by: Harshad Shirwadkar <harshadshirwadkar@gmail.com> Link: https://lore.kernel.org/r/20201106035911.1942128-3-harshadshirwadkar@gmail.com Signed-off-by: Theodore Ts'o <tytso@mit.edu> --- include/trace/events/ext4.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/trace/events/ext4.h b/include/trace/events/ext4.h index b14314fcf732..0f899d3b09d3 100644 --- a/include/trace/events/ext4.h +++ b/include/trace/events/ext4.h @@ -100,7 +100,7 @@ TRACE_DEFINE_ENUM(ES_REFERENCED_B); { EXT4_FC_REASON_XATTR, "XATTR"}, \ { EXT4_FC_REASON_CROSS_RENAME, "CROSS_RENAME"}, \ { EXT4_FC_REASON_JOURNAL_FLAG_CHANGE, "JOURNAL_FLAG_CHANGE"}, \ - { EXT4_FC_REASON_MEM, "NO_MEM"}, \ + { EXT4_FC_REASON_NOMEM, "NO_MEM"}, \ { EXT4_FC_REASON_SWAP_BOOT, "SWAP_BOOT"}, \ { EXT4_FC_REASON_RESIZE, "RESIZE"}, \ { EXT4_FC_REASON_RENAME_DIR, "RENAME_DIR"}, \ @@ -2917,13 +2917,13 @@ TRACE_EVENT(ext4_fc_stats, ), TP_printk("dev %d:%d fc ineligible reasons:\n" - "%s:%d, %s:%d, %s:%d, %s:%d, %s:%d, %s:%d, %s:%d, %s,%d; " + "%s:%d, %s:%d, %s:%d, %s:%d, %s:%d, %s:%d, %s:%d, %s:%d; " "num_commits:%ld, ineligible: %ld, numblks: %ld", MAJOR(__entry->dev), MINOR(__entry->dev), FC_REASON_NAME_STAT(EXT4_FC_REASON_XATTR), FC_REASON_NAME_STAT(EXT4_FC_REASON_CROSS_RENAME), FC_REASON_NAME_STAT(EXT4_FC_REASON_JOURNAL_FLAG_CHANGE), - FC_REASON_NAME_STAT(EXT4_FC_REASON_MEM), + FC_REASON_NAME_STAT(EXT4_FC_REASON_NOMEM), FC_REASON_NAME_STAT(EXT4_FC_REASON_SWAP_BOOT), FC_REASON_NAME_STAT(EXT4_FC_REASON_RESIZE), FC_REASON_NAME_STAT(EXT4_FC_REASON_RENAME_DIR), -- cgit v1.2.3 From ede7dc7fa0af619afc08995776eadb9ff3b0a711 Mon Sep 17 00:00:00 2001 From: Harshad Shirwadkar <harshadshirwadkar@gmail.com> Date: Thu, 5 Nov 2020 19:58:54 -0800 Subject: jbd2: rename j_maxlen to j_total_len and add jbd2_journal_max_txn_bufs The on-disk superblock field sb->s_maxlen represents the total size of the journal including the fast commit area and is no more the max number of blocks available for a transaction. The maximum number of blocks available to a transaction is reduced by the number of fast commit blocks. So, this patch renames j_maxlen to j_total_len to better represent its intent. Also, it adds a function to calculate max number of bufs available for a transaction. Suggested-by: Jan Kara <jack@suse.cz> Signed-off-by: Harshad Shirwadkar <harshadshirwadkar@gmail.com> Link: https://lore.kernel.org/r/20201106035911.1942128-6-harshadshirwadkar@gmail.com Signed-off-by: Theodore Ts'o <tytso@mit.edu> --- fs/ext4/fsmap.c | 2 +- fs/ext4/super.c | 2 +- fs/jbd2/commit.c | 2 +- fs/jbd2/journal.c | 12 ++++++------ fs/jbd2/recovery.c | 6 +++--- fs/ocfs2/journal.c | 2 +- include/linux/jbd2.h | 9 +++++++-- 7 files changed, 20 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/fs/ext4/fsmap.c b/fs/ext4/fsmap.c index b232c2767534..4c2a9fe30067 100644 --- a/fs/ext4/fsmap.c +++ b/fs/ext4/fsmap.c @@ -280,7 +280,7 @@ static int ext4_getfsmap_logdev(struct super_block *sb, struct ext4_fsmap *keys, /* Fabricate an rmap entry for the external log device. */ irec.fmr_physical = journal->j_blk_offset; - irec.fmr_length = journal->j_maxlen; + irec.fmr_length = journal->j_total_len; irec.fmr_owner = EXT4_FMR_OWN_LOG; irec.fmr_flags = 0; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 738a6dd4957d..8a6dd433bb70 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -3976,7 +3976,7 @@ int ext4_calculate_overhead(struct super_block *sb) * loaded or not */ if (sbi->s_journal && !sbi->s_journal_bdev) - overhead += EXT4_NUM_B2C(sbi, sbi->s_journal->j_maxlen); + overhead += EXT4_NUM_B2C(sbi, sbi->s_journal->j_total_len); else if (ext4_has_feature_journal(sb) && !sbi->s_journal && j_inum) { /* j_inum for internal journal is non-zero */ j_inode = ext4_get_journal_inode(sb, j_inum); diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index fa688e163a80..ec516490cb35 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -801,7 +801,7 @@ start_journal_io: if (first_block < journal->j_tail) freed += journal->j_last - journal->j_first; /* Update tail only if we free significant amount of space */ - if (freed < journal->j_maxlen / 4) + if (freed < jbd2_journal_get_max_txn_bufs(journal)) update_tail = 0; } J_ASSERT(commit_transaction->t_state == T_COMMIT); diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 0c7c42bd530f..c3c768248527 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -1348,7 +1348,7 @@ static journal_t *journal_init_common(struct block_device *bdev, journal->j_dev = bdev; journal->j_fs_dev = fs_dev; journal->j_blk_offset = start; - journal->j_maxlen = len; + journal->j_total_len = len; /* We need enough buffers to write out full descriptor block. */ n = journal->j_blocksize / jbd2_min_tag_size(); journal->j_wbufsize = n; @@ -1531,7 +1531,7 @@ static int journal_reset(journal_t *journal) journal->j_commit_sequence = journal->j_transaction_sequence - 1; journal->j_commit_request = journal->j_commit_sequence; - journal->j_max_transaction_buffers = journal->j_maxlen / 4; + journal->j_max_transaction_buffers = jbd2_journal_get_max_txn_bufs(journal); /* * As a special case, if the on-disk copy is already marked as needing @@ -1792,15 +1792,15 @@ static int journal_get_superblock(journal_t *journal) goto out; } - if (be32_to_cpu(sb->s_maxlen) < journal->j_maxlen) - journal->j_maxlen = be32_to_cpu(sb->s_maxlen); - else if (be32_to_cpu(sb->s_maxlen) > journal->j_maxlen) { + if (be32_to_cpu(sb->s_maxlen) < journal->j_total_len) + journal->j_total_len = be32_to_cpu(sb->s_maxlen); + else if (be32_to_cpu(sb->s_maxlen) > journal->j_total_len) { printk(KERN_WARNING "JBD2: journal file too short\n"); goto out; } if (be32_to_cpu(sb->s_first) == 0 || - be32_to_cpu(sb->s_first) >= journal->j_maxlen) { + be32_to_cpu(sb->s_first) >= journal->j_total_len) { printk(KERN_WARNING "JBD2: Invalid start block of journal: %u\n", be32_to_cpu(sb->s_first)); diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index eb2606133cd8..dc0694fcfcd1 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -74,8 +74,8 @@ static int do_readahead(journal_t *journal, unsigned int start) /* Do up to 128K of readahead */ max = start + (128 * 1024 / journal->j_blocksize); - if (max > journal->j_maxlen) - max = journal->j_maxlen; + if (max > journal->j_total_len) + max = journal->j_total_len; /* Do the readahead itself. We'll submit MAXBUF buffer_heads at * a time to the block device IO layer. */ @@ -134,7 +134,7 @@ static int jread(struct buffer_head **bhp, journal_t *journal, *bhp = NULL; - if (offset >= journal->j_maxlen) { + if (offset >= journal->j_total_len) { printk(KERN_ERR "JBD2: corrupted journal superblock\n"); return -EFSCORRUPTED; } diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index b9a9d69dde7e..db52e843002a 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -877,7 +877,7 @@ int ocfs2_journal_init(struct ocfs2_journal *journal, int *dirty) goto done; } - trace_ocfs2_journal_init_maxlen(j_journal->j_maxlen); + trace_ocfs2_journal_init_maxlen(j_journal->j_total_len); *dirty = (le32_to_cpu(di->id1.journal1.ij_flags) & OCFS2_JOURNAL_DIRTY_FL); diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 1d5566af48ac..e0b6b53eae64 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -988,9 +988,9 @@ struct journal_s struct block_device *j_fs_dev; /** - * @j_maxlen: Total maximum capacity of the journal region on disk. + * @j_total_len: Total maximum capacity of the journal region on disk. */ - unsigned int j_maxlen; + unsigned int j_total_len; /** * @j_reserved_credits: @@ -1624,6 +1624,11 @@ int jbd2_wait_inode_data(journal_t *journal, struct jbd2_inode *jinode); int jbd2_fc_wait_bufs(journal_t *journal, int num_blks); int jbd2_fc_release_bufs(journal_t *journal); +static inline int jbd2_journal_get_max_txn_bufs(journal_t *journal) +{ + return (journal->j_total_len - journal->j_fc_wbufsize) / 4; +} + /* * is_journal_abort * -- cgit v1.2.3 From a1e5e465b31d6015fccb359d99053b39e5180466 Mon Sep 17 00:00:00 2001 From: Harshad Shirwadkar <harshadshirwadkar@gmail.com> Date: Thu, 5 Nov 2020 19:58:55 -0800 Subject: ext4: clean up the JBD2 API that initializes fast commits This patch removes jbd2_fc_init() API and its related functions to simplify enabling fast commits. With this change, the number of fast commit blocks to use is solely determined by the JBD2 layer. So, we move the default value for minimum number of fast commit blocks from ext4/fast_commit.h to include/linux/jbd2.h. However, whether or not to use fast commits is determined by the file system. The file system just sets the fast commit feature using jbd2_journal_set_features(). JBD2 layer then determines how many blocks to use for fast commits (based on the value found in the JBD2 superblock). Note that the JBD2 feature flag of fast commits is just an indication that there are fast commit blocks present on disk. It doesn't tell JBD2 layer about the intent of the file system of whether to it wants to use fast commit or not. That's why, we blindly clear the fast commit flag in journal_reset() after the recovery is done. Suggested-by: Jan Kara <jack@suse.cz> Signed-off-by: Harshad Shirwadkar <harshadshirwadkar@gmail.com> Link: https://lore.kernel.org/r/20201106035911.1942128-7-harshadshirwadkar@gmail.com Signed-off-by: Theodore Ts'o <tytso@mit.edu> --- fs/ext4/fast_commit.c | 14 -------- fs/ext4/fast_commit.h | 3 -- fs/ext4/super.c | 8 +++++ fs/jbd2/journal.c | 96 ++++++++++++++++++++++++++++++--------------------- include/linux/jbd2.h | 2 +- 5 files changed, 65 insertions(+), 58 deletions(-) (limited to 'include') diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c index 9399e9cccb7e..bab60c5d5095 100644 --- a/fs/ext4/fast_commit.c +++ b/fs/ext4/fast_commit.c @@ -2091,8 +2091,6 @@ static int ext4_fc_replay(journal_t *journal, struct buffer_head *bh, void ext4_fc_init(struct super_block *sb, journal_t *journal) { - int num_fc_blocks; - /* * We set replay callback even if fast commit disabled because we may * could still have fast commit blocks that need to be replayed even if @@ -2102,18 +2100,6 @@ void ext4_fc_init(struct super_block *sb, journal_t *journal) if (!test_opt2(sb, JOURNAL_FAST_COMMIT)) return; journal->j_fc_cleanup_callback = ext4_fc_cleanup; - if (!buffer_uptodate(journal->j_sb_buffer) - && ext4_read_bh_lock(journal->j_sb_buffer, REQ_META | REQ_PRIO, - true)) { - ext4_msg(sb, KERN_ERR, "I/O error on journal"); - return; - } - num_fc_blocks = be32_to_cpu(journal->j_superblock->s_num_fc_blks); - if (jbd2_fc_init(journal, num_fc_blocks ? num_fc_blocks : - EXT4_NUM_FC_BLKS)) { - pr_warn("Error while enabling fast commits, turning off."); - ext4_clear_feature_fast_commit(sb); - } } const char *fc_ineligible_reasons[] = { diff --git a/fs/ext4/fast_commit.h b/fs/ext4/fast_commit.h index 140fbb6af19e..1d96e0ac8138 100644 --- a/fs/ext4/fast_commit.h +++ b/fs/ext4/fast_commit.h @@ -3,9 +3,6 @@ #ifndef __FAST_COMMIT_H__ #define __FAST_COMMIT_H__ -/* Number of blocks in journal area to allocate for fast commits */ -#define EXT4_NUM_FC_BLKS 256 - /* Fast commit tags */ #define EXT4_FC_TAG_ADD_RANGE 0x0001 #define EXT4_FC_TAG_DEL_RANGE 0x0002 diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 8a6dd433bb70..ba02d7c86fb3 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -4857,6 +4857,14 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) goto failed_mount_wq; } + if (test_opt2(sb, JOURNAL_FAST_COMMIT) && + !jbd2_journal_set_features(EXT4_SB(sb)->s_journal, 0, 0, + JBD2_FEATURE_INCOMPAT_FAST_COMMIT)) { + ext4_msg(sb, KERN_ERR, + "Failed to set fast commit journal feature"); + goto failed_mount_wq; + } + /* We have now updated the journal if required, so we can * validate the data journaling mode. */ switch (test_opt(sb, DATA_FLAGS)) { diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index c3c768248527..500152f0421a 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -1352,19 +1352,12 @@ static journal_t *journal_init_common(struct block_device *bdev, /* We need enough buffers to write out full descriptor block. */ n = journal->j_blocksize / jbd2_min_tag_size(); journal->j_wbufsize = n; + journal->j_fc_wbuf = NULL; journal->j_wbuf = kmalloc_array(n, sizeof(struct buffer_head *), GFP_KERNEL); if (!journal->j_wbuf) goto err_cleanup; - if (journal->j_fc_wbufsize > 0) { - journal->j_fc_wbuf = kmalloc_array(journal->j_fc_wbufsize, - sizeof(struct buffer_head *), - GFP_KERNEL); - if (!journal->j_fc_wbuf) - goto err_cleanup; - } - bh = getblk_unmovable(journal->j_dev, start, journal->j_blocksize); if (!bh) { pr_err("%s: Cannot get buffer for journal superblock\n", @@ -1378,23 +1371,11 @@ static journal_t *journal_init_common(struct block_device *bdev, err_cleanup: kfree(journal->j_wbuf); - kfree(journal->j_fc_wbuf); jbd2_journal_destroy_revoke(journal); kfree(journal); return NULL; } -int jbd2_fc_init(journal_t *journal, int num_fc_blks) -{ - journal->j_fc_wbufsize = num_fc_blks; - journal->j_fc_wbuf = kmalloc_array(journal->j_fc_wbufsize, - sizeof(struct buffer_head *), GFP_KERNEL); - if (!journal->j_fc_wbuf) - return -ENOMEM; - return 0; -} -EXPORT_SYMBOL(jbd2_fc_init); - /* jbd2_journal_init_dev and jbd2_journal_init_inode: * * Create a journal structure assigned some fixed set of disk blocks to @@ -1512,16 +1493,7 @@ static int journal_reset(journal_t *journal) } journal->j_first = first; - - if (jbd2_has_feature_fast_commit(journal) && - journal->j_fc_wbufsize > 0) { - journal->j_fc_last = last; - journal->j_last = last - journal->j_fc_wbufsize; - journal->j_fc_first = journal->j_last + 1; - journal->j_fc_off = 0; - } else { - journal->j_last = last; - } + journal->j_last = last; journal->j_head = journal->j_first; journal->j_tail = journal->j_first; @@ -1533,6 +1505,13 @@ static int journal_reset(journal_t *journal) journal->j_max_transaction_buffers = jbd2_journal_get_max_txn_bufs(journal); + /* + * Now that journal recovery is done, turn fast commits off here. This + * way, if fast commit was enabled before the crash but if now FS has + * disabled it, we don't enable fast commits. + */ + jbd2_clear_feature_fast_commit(journal); + /* * As a special case, if the on-disk copy is already marked as needing * no recovery (s_start == 0), then we can safely defer the superblock @@ -1872,6 +1851,7 @@ static int load_superblock(journal_t *journal) { int err; journal_superblock_t *sb; + int num_fc_blocks; err = journal_get_superblock(journal); if (err) @@ -1883,15 +1863,17 @@ static int load_superblock(journal_t *journal) journal->j_tail = be32_to_cpu(sb->s_start); journal->j_first = be32_to_cpu(sb->s_first); journal->j_errno = be32_to_cpu(sb->s_errno); + journal->j_last = be32_to_cpu(sb->s_maxlen); - if (jbd2_has_feature_fast_commit(journal) && - journal->j_fc_wbufsize > 0) { + if (jbd2_has_feature_fast_commit(journal)) { journal->j_fc_last = be32_to_cpu(sb->s_maxlen); - journal->j_last = journal->j_fc_last - journal->j_fc_wbufsize; + num_fc_blocks = be32_to_cpu(sb->s_num_fc_blks); + if (!num_fc_blocks) + num_fc_blocks = JBD2_MIN_FC_BLOCKS; + if (journal->j_last - num_fc_blocks >= JBD2_MIN_JOURNAL_BLOCKS) + journal->j_last = journal->j_fc_last - num_fc_blocks; journal->j_fc_first = journal->j_last + 1; journal->j_fc_off = 0; - } else { - journal->j_last = be32_to_cpu(sb->s_maxlen); } return 0; @@ -1954,9 +1936,6 @@ int jbd2_journal_load(journal_t *journal) */ journal->j_flags &= ~JBD2_ABORT; - if (journal->j_fc_wbufsize > 0) - jbd2_journal_set_features(journal, 0, 0, - JBD2_FEATURE_INCOMPAT_FAST_COMMIT); /* OK, we've finished with the dynamic journal bits: * reinitialise the dynamic contents of the superblock in memory * and reset them on disk. */ @@ -2040,8 +2019,7 @@ int jbd2_journal_destroy(journal_t *journal) jbd2_journal_destroy_revoke(journal); if (journal->j_chksum_driver) crypto_free_shash(journal->j_chksum_driver); - if (journal->j_fc_wbufsize > 0) - kfree(journal->j_fc_wbuf); + kfree(journal->j_fc_wbuf); kfree(journal->j_wbuf); kfree(journal); @@ -2116,6 +2094,37 @@ int jbd2_journal_check_available_features(journal_t *journal, unsigned long comp return 0; } +static int +jbd2_journal_initialize_fast_commit(journal_t *journal) +{ + journal_superblock_t *sb = journal->j_superblock; + unsigned long long num_fc_blks; + + num_fc_blks = be32_to_cpu(sb->s_num_fc_blks); + if (num_fc_blks == 0) + num_fc_blks = JBD2_MIN_FC_BLOCKS; + if (journal->j_last - num_fc_blks < JBD2_MIN_JOURNAL_BLOCKS) + return -ENOSPC; + + /* Are we called twice? */ + WARN_ON(journal->j_fc_wbuf != NULL); + journal->j_fc_wbuf = kmalloc_array(num_fc_blks, + sizeof(struct buffer_head *), GFP_KERNEL); + if (!journal->j_fc_wbuf) + return -ENOMEM; + + journal->j_fc_wbufsize = num_fc_blks; + journal->j_fc_last = journal->j_last; + journal->j_last = journal->j_fc_last - num_fc_blks; + journal->j_fc_first = journal->j_last + 1; + journal->j_fc_off = 0; + journal->j_free = journal->j_last - journal->j_first; + journal->j_max_transaction_buffers = + jbd2_journal_get_max_txn_bufs(journal); + + return 0; +} + /** * int jbd2_journal_set_features() - Mark a given journal feature in the superblock * @journal: Journal to act on. @@ -2159,6 +2168,13 @@ int jbd2_journal_set_features(journal_t *journal, unsigned long compat, sb = journal->j_superblock; + if (incompat & JBD2_FEATURE_INCOMPAT_FAST_COMMIT) { + if (jbd2_journal_initialize_fast_commit(journal)) { + pr_err("JBD2: Cannot enable fast commits.\n"); + return 0; + } + } + /* Load the checksum driver if necessary */ if ((journal->j_chksum_driver == NULL) && INCOMPAT_FEATURE_ON(JBD2_FEATURE_INCOMPAT_CSUM_V3)) { diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index e0b6b53eae64..b2caf7bbd8e5 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -68,6 +68,7 @@ extern void *jbd2_alloc(size_t size, gfp_t flags); extern void jbd2_free(void *ptr, size_t size); #define JBD2_MIN_JOURNAL_BLOCKS 1024 +#define JBD2_MIN_FC_BLOCKS 256 #ifdef __KERNEL__ @@ -1614,7 +1615,6 @@ extern void __jbd2_journal_drop_transaction(journal_t *, transaction_t *); extern int jbd2_cleanup_journal_tail(journal_t *); /* Fast commit related APIs */ -int jbd2_fc_init(journal_t *journal, int num_fc_blks); int jbd2_fc_begin_commit(journal_t *journal, tid_t tid); int jbd2_fc_end_commit(journal_t *journal); int jbd2_fc_end_commit_fallback(journal_t *journal, tid_t tid); -- cgit v1.2.3 From c460e5edc85a063ec9cb60addff93d00ed378701 Mon Sep 17 00:00:00 2001 From: Harshad Shirwadkar <harshadshirwadkar@gmail.com> Date: Thu, 5 Nov 2020 19:58:57 -0800 Subject: jbd2: don't use state lock during commit path Variables journal->j_fc_off, journal->j_fc_wbuf are accessed during commit path. Since today we allow only one process to perform a fast commit, there is no need take state lock before accessing these variables. This patch removes these locks and adds comments to describe this. Suggested-by: Jan Kara <jack@suse.cz> Signed-off-by: Harshad Shirwadkar <harshadshirwadkar@gmail.com> Link: https://lore.kernel.org/r/20201106035911.1942128-9-harshadshirwadkar@gmail.com Signed-off-by: Theodore Ts'o <tytso@mit.edu> --- fs/jbd2/journal.c | 6 ------ include/linux/jbd2.h | 10 ++++++---- 2 files changed, 6 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 500152f0421a..778ea50fc8d1 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -865,7 +865,6 @@ int jbd2_fc_get_buf(journal_t *journal, struct buffer_head **bh_out) int fc_off; *bh_out = NULL; - write_lock(&journal->j_state_lock); if (journal->j_fc_off + journal->j_fc_first < journal->j_fc_last) { fc_off = journal->j_fc_off; @@ -874,7 +873,6 @@ int jbd2_fc_get_buf(journal_t *journal, struct buffer_head **bh_out) } else { ret = -EINVAL; } - write_unlock(&journal->j_state_lock); if (ret) return ret; @@ -909,9 +907,7 @@ int jbd2_fc_wait_bufs(journal_t *journal, int num_blks) struct buffer_head *bh; int i, j_fc_off; - read_lock(&journal->j_state_lock); j_fc_off = journal->j_fc_off; - read_unlock(&journal->j_state_lock); /* * Wait in reverse order to minimize chances of us being woken up before @@ -939,9 +935,7 @@ int jbd2_fc_release_bufs(journal_t *journal) struct buffer_head *bh; int i, j_fc_off; - read_lock(&journal->j_state_lock); j_fc_off = journal->j_fc_off; - read_unlock(&journal->j_state_lock); /* * Wait in reverse order to minimize chances of us being woken up before diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index b2caf7bbd8e5..5f0ef6380b0c 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -945,8 +945,9 @@ struct journal_s /** * @j_fc_off: * - * Number of fast commit blocks currently allocated. - * [j_state_lock]. + * Number of fast commit blocks currently allocated. Accessed only + * during fast commit. Currently only process can do fast commit, so + * this field is not protected by any lock. */ unsigned long j_fc_off; @@ -1109,8 +1110,9 @@ struct journal_s struct buffer_head **j_wbuf; /** - * @j_fc_wbuf: Array of fast commit bhs for - * jbd2_journal_commit_transaction. + * @j_fc_wbuf: Array of fast commit bhs for fast commit. Accessed only + * during a fast commit. Currently only process can do fast commit, so + * this field is not protected by any lock. */ struct buffer_head **j_fc_wbuf; -- cgit v1.2.3 From 0bce577bf9cae13ae32d391432d0030e3f67fc1d Mon Sep 17 00:00:00 2001 From: Harshad Shirwadkar <harshadshirwadkar@gmail.com> Date: Thu, 5 Nov 2020 19:58:58 -0800 Subject: jbd2: don't pass tid to jbd2_fc_end_commit_fallback() In jbd2_fc_end_commit_fallback(), we know which tid to commit. There's no need for caller to pass it. Suggested-by: Jan Kara <jack@suse.cz> Signed-off-by: Harshad Shirwadkar <harshadshirwadkar@gmail.com> Link: https://lore.kernel.org/r/20201106035911.1942128-10-harshadshirwadkar@gmail.com Signed-off-by: Theodore Ts'o <tytso@mit.edu> --- fs/ext4/fast_commit.c | 2 +- fs/jbd2/journal.c | 12 +++++++++--- include/linux/jbd2.h | 2 +- 3 files changed, 11 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c index bab60c5d5095..e69c580fa91e 100644 --- a/fs/ext4/fast_commit.c +++ b/fs/ext4/fast_commit.c @@ -1143,7 +1143,7 @@ out: "Fast commit ended with blks = %d, reason = %d, subtid - %d", nblks, reason, subtid); if (reason == EXT4_FC_REASON_FC_FAILED) - return jbd2_fc_end_commit_fallback(journal, commit_tid); + return jbd2_fc_end_commit_fallback(journal); if (reason == EXT4_FC_REASON_FC_START_FAILED || reason == EXT4_FC_REASON_INELIGIBLE) return jbd2_complete_transaction(journal, commit_tid); diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 778ea50fc8d1..59166e299cde 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -777,13 +777,19 @@ static int __jbd2_fc_end_commit(journal_t *journal, tid_t tid, bool fallback) int jbd2_fc_end_commit(journal_t *journal) { - return __jbd2_fc_end_commit(journal, 0, 0); + return __jbd2_fc_end_commit(journal, 0, false); } EXPORT_SYMBOL(jbd2_fc_end_commit); -int jbd2_fc_end_commit_fallback(journal_t *journal, tid_t tid) +int jbd2_fc_end_commit_fallback(journal_t *journal) { - return __jbd2_fc_end_commit(journal, tid, 1); + tid_t tid; + + read_lock(&journal->j_state_lock); + tid = journal->j_running_transaction ? + journal->j_running_transaction->t_tid : 0; + read_unlock(&journal->j_state_lock); + return __jbd2_fc_end_commit(journal, tid, true); } EXPORT_SYMBOL(jbd2_fc_end_commit_fallback); diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 5f0ef6380b0c..1c49fd62ff2e 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1619,7 +1619,7 @@ extern int jbd2_cleanup_journal_tail(journal_t *); /* Fast commit related APIs */ int jbd2_fc_begin_commit(journal_t *journal, tid_t tid); int jbd2_fc_end_commit(journal_t *journal); -int jbd2_fc_end_commit_fallback(journal_t *journal, tid_t tid); +int jbd2_fc_end_commit_fallback(journal_t *journal); int jbd2_fc_get_buf(journal_t *journal, struct buffer_head **bh_out); int jbd2_submit_inode_data(struct jbd2_inode *jinode); int jbd2_wait_inode_data(journal_t *journal, struct jbd2_inode *jinode); -- cgit v1.2.3 From 556e0319fbb8eee3fa19fdcc27c8bcb4af1c7211 Mon Sep 17 00:00:00 2001 From: Harshad Shirwadkar <harshadshirwadkar@gmail.com> Date: Thu, 5 Nov 2020 19:59:07 -0800 Subject: ext4: disable fast commit with data journalling Fast commits don't work with data journalling. This patch disables the fast commit support when data journalling is turned on. Suggested-by: Jan Kara <jack@suse.cz> Signed-off-by: Harshad Shirwadkar <harshadshirwadkar@gmail.com> Reviewed-by: Jan Kara <jack@suse.cz> Link: https://lore.kernel.org/r/20201106035911.1942128-19-harshadshirwadkar@gmail.com Signed-off-by: Theodore Ts'o <tytso@mit.edu> --- fs/ext4/fast_commit.c | 7 +++++++ fs/ext4/fast_commit.h | 1 + fs/ext4/super.c | 3 ++- include/trace/events/ext4.h | 6 ++++-- 4 files changed, 14 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c index 6b963e09af2c..fb9b4e9d82b2 100644 --- a/fs/ext4/fast_commit.c +++ b/fs/ext4/fast_commit.c @@ -483,6 +483,12 @@ void ext4_fc_track_inode(handle_t *handle, struct inode *inode) if (S_ISDIR(inode->i_mode)) return; + if (ext4_should_journal_data(inode)) { + ext4_fc_mark_ineligible(inode->i_sb, + EXT4_FC_REASON_INODE_JOURNAL_DATA); + return; + } + ret = ext4_fc_track_template(handle, inode, __track_inode, NULL, 1); trace_ext4_fc_track_inode(inode, ret); } @@ -2102,6 +2108,7 @@ const char *fc_ineligible_reasons[] = { "Resize", "Dir renamed", "Falloc range op", + "Data journalling", "FC Commit Failed" }; diff --git a/fs/ext4/fast_commit.h b/fs/ext4/fast_commit.h index 1d96e0ac8138..3a6e5a1fa1b8 100644 --- a/fs/ext4/fast_commit.h +++ b/fs/ext4/fast_commit.h @@ -102,6 +102,7 @@ enum { EXT4_FC_REASON_RESIZE, EXT4_FC_REASON_RENAME_DIR, EXT4_FC_REASON_FALLOC_RANGE, + EXT4_FC_REASON_INODE_JOURNAL_DATA, EXT4_FC_COMMIT_FAILED, EXT4_FC_REASON_MAX }; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index ba02d7c86fb3..9e2e06ed2c45 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -4340,9 +4340,10 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) #endif if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) { - printk_once(KERN_WARNING "EXT4-fs: Warning: mounting with data=journal disables delayed allocation, dioread_nolock, and O_DIRECT support!\n"); + printk_once(KERN_WARNING "EXT4-fs: Warning: mounting with data=journal disables delayed allocation, dioread_nolock, O_DIRECT and fast_commit support!\n"); /* can't mount with both data=journal and dioread_nolock. */ clear_opt(sb, DIOREAD_NOLOCK); + clear_opt2(sb, JOURNAL_FAST_COMMIT); if (test_opt2(sb, EXPLICIT_DELALLOC)) { ext4_msg(sb, KERN_ERR, "can't mount with " "both data=journal and delalloc"); diff --git a/include/trace/events/ext4.h b/include/trace/events/ext4.h index 0f899d3b09d3..70ae5497b73a 100644 --- a/include/trace/events/ext4.h +++ b/include/trace/events/ext4.h @@ -104,7 +104,8 @@ TRACE_DEFINE_ENUM(ES_REFERENCED_B); { EXT4_FC_REASON_SWAP_BOOT, "SWAP_BOOT"}, \ { EXT4_FC_REASON_RESIZE, "RESIZE"}, \ { EXT4_FC_REASON_RENAME_DIR, "RENAME_DIR"}, \ - { EXT4_FC_REASON_FALLOC_RANGE, "FALLOC_RANGE"}) + { EXT4_FC_REASON_FALLOC_RANGE, "FALLOC_RANGE"}, \ + { EXT4_FC_REASON_INODE_JOURNAL_DATA, "INODE_JOURNAL_DATA"}) TRACE_EVENT(ext4_other_inode_update_time, TP_PROTO(struct inode *inode, ino_t orig_ino), @@ -2917,7 +2918,7 @@ TRACE_EVENT(ext4_fc_stats, ), TP_printk("dev %d:%d fc ineligible reasons:\n" - "%s:%d, %s:%d, %s:%d, %s:%d, %s:%d, %s:%d, %s:%d, %s:%d; " + "%s:%d, %s:%d, %s:%d, %s:%d, %s:%d, %s:%d, %s:%d, %s:%d, %s:%d; " "num_commits:%ld, ineligible: %ld, numblks: %ld", MAJOR(__entry->dev), MINOR(__entry->dev), FC_REASON_NAME_STAT(EXT4_FC_REASON_XATTR), @@ -2928,6 +2929,7 @@ TRACE_EVENT(ext4_fc_stats, FC_REASON_NAME_STAT(EXT4_FC_REASON_RESIZE), FC_REASON_NAME_STAT(EXT4_FC_REASON_RENAME_DIR), FC_REASON_NAME_STAT(EXT4_FC_REASON_FALLOC_RANGE), + FC_REASON_NAME_STAT(EXT4_FC_REASON_INODE_JOURNAL_DATA), __entry->sbi->s_fc_stats.fc_num_commits, __entry->sbi->s_fc_stats.fc_ineligible_commits, __entry->sbi->s_fc_stats.fc_numblks) -- cgit v1.2.3