[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240818040356.241684-4-harshadshirwadkar@gmail.com>
Date: Sun, 18 Aug 2024 04:03:49 +0000
From: Harshad Shirwadkar <harshadshirwadkar@...il.com>
To: linux-ext4@...r.kernel.org
Cc: tytso@....edu,
jack@...e.cz,
harshads@...gle.com,
Harshad Shirwadkar <harshadshirwadkar@...il.com>
Subject: [PATCH v7 2/9] ext4: for committing inode, make ext4_fc_track_inode wait
If the inode that's being requested to track using ext4_fc_track_inode
is being committed, then wait until the inode finishes the
commit. Also, add calls to ext4_fc_track_inode at the right places.
With this patch, now calling ext4_reserve_inode_write() results in
inode being tracked for next fast commit. A subtle lock ordering
requirement with i_data_sem (which is documented in the code) requires
that ext4_fc_track_inode() be called before grabbing i_data_sem. So,
this patch also adds explicit ext4_fc_track_inode() calls in places
where i_data_sem grabbed.
Signed-off-by: Harshad Shirwadkar <harshadshirwadkar@...il.com>
---
fs/ext4/fast_commit.c | 33 +++++++++++++++++++++++++++++++++
fs/ext4/inline.c | 3 +++
fs/ext4/inode.c | 4 ++++
3 files changed, 40 insertions(+)
diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c
index 8731728cc..dfa999913 100644
--- a/fs/ext4/fast_commit.c
+++ b/fs/ext4/fast_commit.c
@@ -12,6 +12,7 @@
#include "ext4_extents.h"
#include "mballoc.h"
+#include <linux/lockdep.h>
/*
* Ext4 Fast Commits
* -----------------
@@ -581,6 +582,8 @@ static int __track_inode(struct inode *inode, void *arg, bool update)
void ext4_fc_track_inode(handle_t *handle, struct inode *inode)
{
+ struct ext4_inode_info *ei = EXT4_I(inode);
+ wait_queue_head_t *wq;
int ret;
if (S_ISDIR(inode->i_mode))
@@ -598,6 +601,36 @@ void ext4_fc_track_inode(handle_t *handle, struct inode *inode)
if (ext4_test_mount_flag(inode->i_sb, EXT4_MF_FC_INELIGIBLE))
return;
+ if (!list_empty(&ei->i_fc_list))
+ return;
+
+#ifdef CONFIG_LOCKDEP
+ /*
+ * If we come here, we may sleep while waiting for the inode to
+ * commit. We shouldn't be holding i_data_sem when we go to sleep since
+ * the commit path needs to grab the lock while committing the inode.
+ */
+ WARN_ON(lockdep_is_held(&ei->i_data_sem));
+#endif
+
+ while (ext4_test_inode_state(inode, EXT4_STATE_FC_COMMITTING)) {
+#if (BITS_PER_LONG < 64)
+ DEFINE_WAIT_BIT(wait, &ei->i_state_flags,
+ EXT4_STATE_FC_COMMITTING);
+ wq = bit_waitqueue(&ei->i_state_flags,
+ EXT4_STATE_FC_COMMITTING);
+#else
+ DEFINE_WAIT_BIT(wait, &ei->i_flags,
+ EXT4_STATE_FC_COMMITTING);
+ wq = bit_waitqueue(&ei->i_flags,
+ EXT4_STATE_FC_COMMITTING);
+#endif
+ prepare_to_wait(wq, &wait.wq_entry, TASK_UNINTERRUPTIBLE);
+ if (ext4_test_inode_state(inode, EXT4_STATE_FC_COMMITTING))
+ schedule();
+ finish_wait(wq, &wait.wq_entry);
+ }
+
ret = ext4_fc_track_template(handle, inode, __track_inode, NULL, 1);
trace_ext4_fc_track_inode(handle, inode, ret);
}
diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c
index e7a09a998..534a5c1bf 100644
--- a/fs/ext4/inline.c
+++ b/fs/ext4/inline.c
@@ -596,6 +596,7 @@ static int ext4_convert_inline_data_to_extent(struct address_space *mapping,
goto out;
}
+ ext4_fc_track_inode(handle, inode);
ret = ext4_destroy_inline_data_nolock(handle, inode);
if (ret)
goto out;
@@ -696,6 +697,7 @@ int ext4_try_to_write_inline_data(struct address_space *mapping,
goto convert;
}
+ ext4_fc_track_inode(handle, inode);
ret = ext4_journal_get_write_access(handle, inode->i_sb, iloc.bh,
EXT4_JTR_NONE);
if (ret)
@@ -948,6 +950,7 @@ int ext4_da_write_inline_data_begin(struct address_space *mapping,
if (ret < 0)
goto out_release_page;
}
+ ext4_fc_track_inode(handle, inode);
ret = ext4_journal_get_write_access(handle, inode->i_sb, iloc.bh,
EXT4_JTR_NONE);
if (ret)
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 941c1c0d5..e11f00ff8 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -636,6 +636,7 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode,
*/
map->m_flags &= ~EXT4_MAP_FLAGS;
+ ext4_fc_track_inode(handle, inode);
/*
* New blocks allocate and/or writing to unwritten extent
* will possibly result in updating i_data, so we take
@@ -4057,6 +4058,7 @@ int ext4_punch_hole(struct file *file, loff_t offset, loff_t length)
if (stop_block > first_block) {
ext4_lblk_t hole_len = stop_block - first_block;
+ ext4_fc_track_inode(handle, inode);
down_write(&EXT4_I(inode)->i_data_sem);
ext4_discard_preallocations(inode);
@@ -4210,6 +4212,7 @@ int ext4_truncate(struct inode *inode)
if (err)
goto out_stop;
+ ext4_fc_track_inode(handle, inode);
down_write(&EXT4_I(inode)->i_data_sem);
ext4_discard_preallocations(inode);
@@ -5806,6 +5809,7 @@ ext4_reserve_inode_write(handle_t *handle, struct inode *inode,
brelse(iloc->bh);
iloc->bh = NULL;
}
+ ext4_fc_track_inode(handle, inode);
}
ext4_std_error(inode->i_sb, err);
return err;
--
2.46.0.184.g6999bdac58-goog
Powered by blists - more mailing lists