The bug of iput() removal from flusher thread?

DISCUSSION OGAWA Hirofumi 4 participants 15 messages Nov 17, 2012 – Nov 21, 2012 VFS

lore

Share & tools

lore

OGAWA Hirofumi

author

2012-11-17 08:42 UTC #0 lore raw

Hi,

In 169ebd90131b2ffca74bb2dbe7eeacd39fb83714 commit, writeback doesn't __iget()/iput() anymore.

This means nobody moves the inode to lru list. I.e.

	new_inode()
	dirty_inode()
	iput_final()
		/* keep inode without adding lru */
	flush indoes
        /* clean inode is not on lru */

I noticed this situation in my FS though, I think the same bug is on all FSes of linus tree too, after this commit.

Am I missing the something?

Thanks.

Signature · 1 line

OGAWA Hirofumi <[email protected]>

OGAWA Hirofumi

author

2012-11-19 08:56 UTC #1 lore raw

Quoted 14 lines · OGAWA Hirofumi

Hi,
In 169ebd90131b2ffca74bb2dbe7eeacd39fb83714 commit, writeback doesn't __iget()/iput() anymore.
This means nobody moves the inode to lru list. I.e.
	new_inode()
	dirty_inode()
	iput_final()
		/* keep inode without adding lru */
	flush indoes
        /* clean inode is not on lru */

Quoted 4 lines

I noticed this situation in my FS though, I think the same bug is on all FSes of linus tree too, after this commit.
Am I missing the something?

This seems to be reproducible by the following,

#!/bin/sh

for i in $(seq -w 1000); do

	for j in $(seq -w 1000); do
        	for k in $(seq -w 1000); do
                	mkdir -p $i/$j
                        echo $i/$j/$k > $i/$j/$k
                        echo 2 > /proc/sys/vm/drop_caches
                done
        done

done

Some inodes never be reclaimed, and ls -l frees those inodes (stat(2) does iget/iput).

Signature · 1 line

OGAWA Hirofumi <[email protected]>

Jan Kara

reviewer

2012-11-19 14:51 UTC #2 lore raw

Quoted 33 lines · OGAWA Hirofumi

Quoted 14 lines · OGAWA Hirofumi
Hi,
In 169ebd90131b2ffca74bb2dbe7eeacd39fb83714 commit, writeback doesn't __iget()/iput() anymore.
This means nobody moves the inode to lru list. I.e.
	new_inode()
	dirty_inode()
	iput_final()
		/* keep inode without adding lru */
	flush indoes
        /* clean inode is not on lru */
Quoted 4 lines
I noticed this situation in my FS though, I think the same bug is on all FSes of linus tree too, after this commit.
Am I missing the something?
This seems to be reproducible by the following,
#!/bin/sh
for i in $(seq -w 1000); do
	for j in $(seq -w 1000); do
        	for k in $(seq -w 1000); do
                	mkdir -p $i/$j
                        echo $i/$j/$k > $i/$j/$k
                        echo 2 > /proc/sys/vm/drop_caches
                done
        done

Quoted 4 lines

done
Some inodes never be reclaimed, and ls -l frees those inodes (stat(2) does iget/iput).

So looking into the code I agree we won't put inode into the LRU when it
is dirty or under writeback and after writeback is done it won't happen either. That's certainly a bug. But I have hard time reproducing your results because on my kernels even dcache doesn't get shrunk thus inodes are pinned in memory by it. Not sure what's going on yet but I'll investigate. Thanks for report!

Honza

Signature · 2 lines

Jan Kara <[email protected]>
SUSE Labs, CR

Jan Kara

reviewer

2012-11-19 19:41 UTC #3 lore raw

Quoted 36 lines · Jan Kara

Quoted 35 lines · OGAWA Hirofumi
Quoted 14 lines · OGAWA Hirofumi
Hi,
In 169ebd90131b2ffca74bb2dbe7eeacd39fb83714 commit, writeback doesn't __iget()/iput() anymore.
This means nobody moves the inode to lru list. I.e.
	new_inode()
	dirty_inode()
	iput_final()
		/* keep inode without adding lru */
	flush indoes
        /* clean inode is not on lru */
Quoted 4 lines
I noticed this situation in my FS though, I think the same bug is on all FSes of linus tree too, after this commit.
Am I missing the something?
This seems to be reproducible by the following,
#!/bin/sh
for i in $(seq -w 1000); do
	for j in $(seq -w 1000); do
        	for k in $(seq -w 1000); do
                	mkdir -p $i/$j
                        echo $i/$j/$k > $i/$j/$k
                        echo 2 > /proc/sys/vm/drop_caches
                done
        done
done

Quoted 8 lines

Quoted 2 lines
Some inodes never be reclaimed, and ls -l frees those inodes (stat(2) does iget/iput).
So looking into the code I agree we won't put inode into the LRU when it
is dirty or under writeback and after writeback is done it won't happen either. That's certainly a bug. But I have hard time reproducing your results because on my kernels even dcache doesn't get shrunk thus inodes are pinned in memory by it. Not sure what's going on yet but I'll investigate. Thanks for report!

OK, that was just reclaim batching code standing in my way. After
figuring that out I could reproduce the issue and test my fix. It is attached.
Honza

Signature · 2 lines

Jan Kara <[email protected]>
SUSE Labs, CR

Quoted 1 line

From 4fdc5d9a66dfe0286ef4f4a7f53fd3b15086470f Mon Sep 17 00:00:00 2001

From: Jan Kara <[email protected]>
Date: Mon, 19 Nov 2012 20:01:16 +0100
Subject: [PATCH] writeback: Put unused inodes to LRU after writeback completion

Commit 169ebd90 removed iget-iput pair from inode writeback. As a side effect, inodes that are dirty during iput_final() call won't be ever added to inode LRU (iput_final() doesn't add dirty inodes to LRU and later when the inode is cleaned there's noone to add the inode there). Thus inodes are effectively unreclaimable until someone looks them up again.

Practical effect of this bug is limited by the fact that inodes are pinned by a dentry for long enough that the inode gets cleaned. But still the bug can have nasty consequences leading up to OOM conditions under certain circumstances. Following can easily reproduce the problem:

for (( i = 0; i < 1000; i++ )); do
mkdir $i
for (( j = 0; j < 1000; j++ )); do
touch $i/$j
echo 2 > /proc/sys/vm/drop_caches
done
done

then one needs to run 'sync; ls -lR' to make inodes reclaimable again.

We fix the issue by inserting unused clean inodes into the LRU after writeback finishes in inode_sync_complete().

CC Al Viro <[email protected]>

Reported-by OGAWA Hirofumi <[email protected]>

Signed-off-by Jan Kara <[email protected]>

---
 fs/fs-writeback.c |    3 +++
 fs/inode.c        |    2 +-
 fs/internal.h     |    1 +
 3 files changed, 5 insertions(+), 1 deletions(-)

fs/fs-writeback.c +3 -0

index 51ea267..ed7613b 100644
@@ -227,6 +227,9 @@ static void requeue_io(struct inode *inode, struct bdi_writeback *wb)
 
 static void inode_sync_complete(struct inode *inode)
 {
+	/* If inode is clean an unused, put it into LRU now.  */
+	if (!(inode->i_state & I_DIRTY) && !atomic_read(&inode->i_count))
+		inode_lru_list_add(inode);
 	inode->i_state &= ~I_SYNC;
 	/* Waiters must see I_SYNC cleared before being woken up */
 	smp_mb();

fs/inode.c +1 -1

index b03c719..275e447 100644
@@ -397,7 +397,7 @@ void ihold(struct inode *inode)
 }
 EXPORT_SYMBOL(ihold);
 
-static void inode_lru_list_add(struct inode *inode)
+void inode_lru_list_add(struct inode *inode)
 {
 	spin_lock(&inode->i_sb->s_inode_lru_lock);
 	if (list_empty(&inode->i_lru)) {

fs/internal.h +1 -0

index 916b7cb..3ecf43d 100644
@@ -110,6 +110,7 @@ extern int open_check_o_direct(struct file *f);
  * inode.c
  */
 extern spinlock_t inode_sb_list_lock;
+extern void inode_lru_list_add(struct inode *inode);
 
 /*
  * fs-writeback.c

Signature · 1 line

1.7.1

OGAWA Hirofumi

author

2012-11-19 20:51 UTC #4 lore raw

Quoted 24 lines · Jan Kara

Quoted 1 line
From 4fdc5d9a66dfe0286ef4f4a7f53fd3b15086470f Mon Sep 17 00:00:00 2001
From: Jan Kara <[email protected]>
Date: Mon, 19 Nov 2012 20:01:16 +0100
Subject: [PATCH] writeback: Put unused inodes to LRU after writeback completion
Commit 169ebd90 removed iget-iput pair from inode writeback. As a side effect, inodes that are dirty during iput_final() call won't be ever added to inode LRU (iput_final() doesn't add dirty inodes to LRU and later when the inode is cleaned there's noone to add the inode there). Thus inodes are effectively unreclaimable until someone looks them up again.
Practical effect of this bug is limited by the fact that inodes are pinned by a dentry for long enough that the inode gets cleaned. But still the bug can have nasty consequences leading up to OOM conditions under certain circumstances. Following can easily reproduce the problem:
for (( i = 0; i < 1000; i++ )); do
mkdir $i
for (( j = 0; j < 1000; j++ )); do
touch $i/$j
echo 2 > /proc/sys/vm/drop_caches
done
done

Quoted 8 lines

then one needs to run 'sync; ls -lR' to make inodes reclaimable again.
We fix the issue by inserting unused clean inodes into the LRU after writeback finishes in inode_sync_complete().
CC Al Viro <[email protected]>
Reported-by OGAWA Hirofumi <[email protected]>
Signed-off-by Jan Kara <[email protected]>

Need to Cc to stable@

Quoted 11 lines

---
 fs/fs-writeback.c |    3 +++
 fs/inode.c        |    2 +-
 fs/internal.h     |    1 +
 3 files changed, 5 insertions(+), 1 deletions(-)

fs/fs-writeback.c +0 -0

index 51ea267..ed7613b 100644

fs/fs-writeback.c static void requeue_io(struct inode *inode, struct bdi_writeback *wb) · line 232

Quoted 6 lines

fs/fs-writeback.c +3 -0

@@ -227,6 +227,9 @@ static void requeue_io(struct inode *inode, struct bdi_writeback *wb)
 
 static void inode_sync_complete(struct inode *inode)
 {
+	/* If inode is clean an unused, put it into LRU now.  */
+	if (!(inode->i_state & I_DIRTY) && !atomic_read(&inode->i_count))
+		inode_lru_list_add(inode);

IMHO, open coding this would be bad idea. And another one is I_REFERENCED. We really want to remove I_REFERENCED?

Thanks.

Signature · 1 line

OGAWA Hirofumi <[email protected]>

Jan Kara

reviewer

2012-11-19 21:24 UTC #5 lore raw

Quoted 28 lines · OGAWA Hirofumi

Quoted 26 lines · Jan Kara
Quoted 1 line
From 4fdc5d9a66dfe0286ef4f4a7f53fd3b15086470f Mon Sep 17 00:00:00 2001
From: Jan Kara <[email protected]>
Date: Mon, 19 Nov 2012 20:01:16 +0100
Subject: [PATCH] writeback: Put unused inodes to LRU after writeback completion
Commit 169ebd90 removed iget-iput pair from inode writeback. As a side effect, inodes that are dirty during iput_final() call won't be ever added to inode LRU (iput_final() doesn't add dirty inodes to LRU and later when the inode is cleaned there's noone to add the inode there). Thus inodes are effectively unreclaimable until someone looks them up again.
Practical effect of this bug is limited by the fact that inodes are pinned by a dentry for long enough that the inode gets cleaned. But still the bug can have nasty consequences leading up to OOM conditions under certain circumstances. Following can easily reproduce the problem:
for (( i = 0; i < 1000; i++ )); do
mkdir $i
for (( j = 0; j < 1000; j++ )); do
touch $i/$j
echo 2 > /proc/sys/vm/drop_caches
done
done
then one needs to run 'sync; ls -lR' to make inodes reclaimable again.

Quoted 8 lines

Quoted 6 lines
We fix the issue by inserting unused clean inodes into the LRU after writeback finishes in inode_sync_complete().
CC Al Viro <[email protected]>
Reported-by OGAWA Hirofumi <[email protected]>
Signed-off-by Jan Kara <[email protected]>
Need to Cc to stable@

Yup, I'll add that. Thanks for reminder.

Quoted 12 lines

---
 fs/fs-writeback.c |    3 +++
 fs/inode.c        |    2 +-
 fs/internal.h     |    1 +
 3 files changed, 5 insertions(+), 1 deletions(-)

fs/fs-writeback.c +0 -0

index 51ea267..ed7613b 100644
@@ -227,6 +227,9 @@ static void requeue_io(struct inode *inode, struct bdi_writeback *wb)

Quoted 7 lines

fs/fs-writeback.c static void requeue_io(struct inode *inode, struct bdi_writeback *wb) · line 232
Quoted 5 lines
fs/fs-writeback.c +3 -0
 static void inode_sync_complete(struct inode *inode)
 {
+	/* If inode is clean an unused, put it into LRU now.  */
+	if (!(inode->i_state & I_DIRTY) && !atomic_read(&inode->i_count))
+		inode_lru_list_add(inode);
IMHO, open coding this would be bad idea.

Do you mean creating a separate function for the above two lines?

Quoted 1 line

And another one is I_REFERENCED. We really want to remove I_REFERENCED?

We don't want I_REFERENCED set - noone used the inode. But looking into
the code with fresh eyes, the fix isn't as simple as I thought. First I need to check MS_ACTIVE and second I need to check I_FREEING... So the condition will be complex enough to warrant a separate function.

Honza

Signature · 2 lines

Jan Kara <[email protected]>
SUSE Labs, CR

OGAWA Hirofumi

author

2012-11-19 21:53 UTC #6 lore raw

Quoted 8 lines · Jan Kara

Quoted 7 lines
fs/fs-writeback.c static void requeue_io(struct inode *inode, struct bdi_writeback *wb) · line 232
Quoted 5 lines
static void inode_sync_complete(struct inode *inode) {
fs/fs-writeback.c +3 -0
+	/* If inode is clean an unused, put it into LRU now.  */
+	if (!(inode->i_state & I_DIRTY) && !atomic_read(&inode->i_count))
+		inode_lru_list_add(inode);
IMHO, open coding this would be bad idea.
Do you mean creating a separate function for the above two lines?

Yes. And the intent is to consolidate "when adds inode to LRU" with iput_final()'s one.

Quoted 5 lines

Quoted 1 line
And another one is I_REFERENCED. We really want to remove I_REFERENCED?
We don't want I_REFERENCED set - noone used the inode. But looking into
the code with fresh eyes, the fix isn't as simple as I thought. First I need to check MS_ACTIVE and second I need to check I_FREEING... So the condition will be complex enough to warrant a separate function.

I can't see the issue (sync_filesystem() will wait I_DIRTY before MS_ACTIVE, and I_DIRTY prevents I_FREEING) though, it may be possible.

Signature · 1 line

OGAWA Hirofumi <[email protected]>

Jan Kara

reviewer

2012-11-21 01:11 UTC #7 lore raw

Quoted 14 lines · OGAWA Hirofumi

Quoted 8 lines · Jan Kara
Quoted 7 lines
fs/fs-writeback.c static void requeue_io(struct inode *inode, struct bdi_writeback *wb) · line 232
Quoted 5 lines
static void inode_sync_complete(struct inode *inode) {
fs/fs-writeback.c +3 -0
+	/* If inode is clean an unused, put it into LRU now.  */
+	if (!(inode->i_state & I_DIRTY) && !atomic_read(&inode->i_count))
+		inode_lru_list_add(inode);
IMHO, open coding this would be bad idea.
Do you mean creating a separate function for the above two lines?
Yes. And the intent is to consolidate "when adds inode to LRU" with iput_final()'s one.

Quoted 8 lines

Quoted 5 lines
Quoted 1 line
And another one is I_REFERENCED. We really want to remove I_REFERENCED?
We don't want I_REFERENCED set - noone used the inode. But looking into
the code with fresh eyes, the fix isn't as simple as I thought. First I need to check MS_ACTIVE and second I need to check I_FREEING... So the condition will be complex enough to warrant a separate function.
I can't see the issue (sync_filesystem() will wait I_DIRTY before MS_ACTIVE, and I_DIRTY prevents I_FREEING) though, it may be possible.

E.g. when inode is deleted it can be both I_DIRTY (and flusher thread
can be working on it) while it is also marked as I_FREEING. In such case we must avoid adding the inode to the LRU.

Regarding MS_ACTIVE - you are right that sync_filesystem() should clean
all dirty inodes but some filesystems dirty their internal inodes during umount so it's better to make flusher thread safe and not add such inodes to the LRU during umount.

Honza

Signature · 2 lines

Jan Kara <[email protected]>
SUSE Labs, CR

Jan Kara

reviewer

2012-11-21 01:48 UTC #8 lore raw

Quoted 23 lines · Jan Kara

Quoted 14 lines · OGAWA Hirofumi
Quoted 8 lines · Jan Kara
Quoted 7 lines
Review
Quoted 5 lines
static void inode_sync_complete(struct inode *inode) {
(quoted diff) +3 -0
+	/* If inode is clean an unused, put it into LRU now.  */
+	if (!(inode->i_state & I_DIRTY) && !atomic_read(&inode->i_count))
+		inode_lru_list_add(inode);
IMHO, open coding this would be bad idea.
Do you mean creating a separate function for the above two lines?
Yes. And the intent is to consolidate "when adds inode to LRU" with iput_final()'s one.
Quoted 8 lines
Quoted 5 lines
Quoted 1 line
And another one is I_REFERENCED. We really want to remove I_REFERENCED?
We don't want I_REFERENCED set - noone used the inode. But looking into
the code with fresh eyes, the fix isn't as simple as I thought. First I need to check MS_ACTIVE and second I need to check I_FREEING... So the condition will be complex enough to warrant a separate function.
I can't see the issue (sync_filesystem() will wait I_DIRTY before MS_ACTIVE, and I_DIRTY prevents I_FREEING) though, it may be possible.

Quoted 8 lines

E.g. when inode is deleted it can be both I_DIRTY (and flusher thread
can be working on it) while it is also marked as I_FREEING. In such case we must avoid adding the inode to the LRU.
Regarding MS_ACTIVE - you are right that sync_filesystem() should clean
all dirty inodes but some filesystems dirty their internal inodes during umount so it's better to make flusher thread safe and not add such inodes to the LRU during umount.

Here's the patch I currently have BTW.

Honza

Signature · 2 lines

Jan Kara <[email protected]>
SUSE Labs, CR

Quoted 1 line

From 00c9878ec690bb8e493582f0109e9aa6ee734ecb Mon Sep 17 00:00:00 2001

From: Jan Kara <[email protected]>
Date: Mon, 19 Nov 2012 20:01:16 +0100
Subject: [PATCH v2] writeback: Put unused inodes to LRU after writeback completion

for (( i = 0; i < 1000; i++ )); do
mkdir $i
for (( j = 0; j < 1000; j++ )); do
touch $i/$j
echo 2 > /proc/sys/vm/drop_caches
done
done

then one needs to run 'sync; ls -lR' to make inodes reclaimable again.

We fix the issue by inserting unused clean inodes into the LRU after writeback finishes in inode_sync_complete().

CC Al Viro <[email protected]>

CC OGAWA Hirofumi <[email protected]>

CC [email protected] # >= 3.5

Reported-by OGAWA Hirofumi <[email protected]>

Signed-off-by Jan Kara <[email protected]>

---
 fs/fs-writeback.c |    2 ++
 fs/inode.c        |   16 ++++++++++++++--
 fs/internal.h     |    1 +
 3 files changed, 17 insertions(+), 2 deletions(-)

fs/fs-writeback.c +2 -0

index 51ea267..3e3422f 100644
@@ -228,6 +228,8 @@ static void requeue_io(struct inode *inode, struct bdi_writeback *wb)
 static void inode_sync_complete(struct inode *inode)
 {
 	inode->i_state &= ~I_SYNC;
+	/* If inode is clean an unused, put it into LRU now... */
+	inode_add_lru(inode);
 	/* Waiters must see I_SYNC cleared before being woken up */
 	smp_mb();
 	wake_up_bit(&inode->i_state, __I_SYNC);

fs/inode.c +14 -2

index b03c719..8f6396f 100644
@@ -408,6 +408,19 @@ static void inode_lru_list_add(struct inode *inode)
 	spin_unlock(&inode->i_sb->s_inode_lru_lock);
 }
 
+/*
+ * Add inode to LRU if needed (inode is unused and clean).
+ *
+ * Needs inode->i_lock held.
+ */
+void inode_add_lru(struct inode *inode)
+{
+	if (!(inode->i_state & (I_DIRTY | I_FREEING | I_SYNC)) &&
+	    !atomic_read(&inode->i_count) && inode->i_sb->s_flags & MS_ACTIVE)
+		inode_lru_list_add(inode);
+}
+
+
 static void inode_lru_list_del(struct inode *inode)
 {
 	spin_lock(&inode->i_sb->s_inode_lru_lock);

@@ -1390,8 +1403,7 @@ static void iput_final(struct inode *inode)
 
 	if (!drop && (sb->s_flags & MS_ACTIVE)) {
 		inode->i_state |= I_REFERENCED;
-		if (!(inode->i_state & (I_DIRTY|I_SYNC)))
-			inode_lru_list_add(inode);
+		inode_add_lru(inode);
 		spin_unlock(&inode->i_lock);
 		return;
 	}

fs/internal.h +1 -0

index 916b7cb..2f6af7f 100644
@@ -110,6 +110,7 @@ extern int open_check_o_direct(struct file *f);
  * inode.c
  */
 extern spinlock_t inode_sb_list_lock;
+extern void inode_add_lru(struct inode *inode);
 
 /*
  * fs-writeback.c

Signature · 1 line

1.7.1

Dave Chinner

2012-11-21 02:44 UTC #9 lore raw

Quoted 22 lines · Jan Kara

fs/fs-writeback.c +2 -0

index 51ea267..3e3422f 100644
@@ -228,6 +228,8 @@ static void requeue_io(struct inode *inode, struct bdi_writeback *wb)
 static void inode_sync_complete(struct inode *inode)
 {
 	inode->i_state &= ~I_SYNC;
+	/* If inode is clean an unused, put it into LRU now... */
+	inode_add_lru(inode);
 	/* Waiters must see I_SYNC cleared before being woken up */
 	smp_mb();
 	wake_up_bit(&inode->i_state, __I_SYNC);

fs/inode.c +0 -0

index b03c719..8f6396f 100644
@@ -408,6 +408,19 @@ static void inode_lru_list_add(struct inode *inode)
 	spin_unlock(&inode->i_sb->s_inode_lru_lock);
 }

fs/inode.c static void inode_lru_list_add(struct inode *inode) · line 420

Quoted 10 lines

fs/inode.c +10 -0

+/*
+ * Add inode to LRU if needed (inode is unused and clean).
+ *
+ * Needs inode->i_lock held.
+ */
+void inode_add_lru(struct inode *inode)
+{
+	if (!(inode->i_state & (I_DIRTY | I_FREEING | I_SYNC)) &&
+	    !atomic_read(&inode->i_count) && inode->i_sb->s_flags & MS_ACTIVE)
+		inode_lru_list_add(inode);

Needs to avoid I_WILL_FREE as well. There's no point putting it on the LRU if we are writing from iput_final()....

Otherwise looks OK.

Cheers,

Dave.

Signature · 2 lines

Dave Chinner
[email protected]

Jan Kara

reviewer

2012-11-21 17:08 UTC #10 lore raw

Quoted 23 lines · Dave Chinner

Quoted 22 lines · Jan Kara

fs/fs-writeback.c +2 -0

index 51ea267..3e3422f 100644
@@ -228,6 +228,8 @@ static void requeue_io(struct inode *inode, struct bdi_writeback *wb)
 static void inode_sync_complete(struct inode *inode)
 {
 	inode->i_state &= ~I_SYNC;
+	/* If inode is clean an unused, put it into LRU now... */
+	inode_add_lru(inode);
 	/* Waiters must see I_SYNC cleared before being woken up */
 	smp_mb();
 	wake_up_bit(&inode->i_state, __I_SYNC);

fs/inode.c +0 -0

index b03c719..8f6396f 100644
@@ -408,6 +408,19 @@ static void inode_lru_list_add(struct inode *inode)
 	spin_unlock(&inode->i_sb->s_inode_lru_lock);
 }

Quoted 13 lines

fs/inode.c static void inode_lru_list_add(struct inode *inode) · line 420
Quoted 10 lines
fs/inode.c +10 -0
+/*
+ * Add inode to LRU if needed (inode is unused and clean).
+ *
+ * Needs inode->i_lock held.
+ */
+void inode_add_lru(struct inode *inode)
+{
+	if (!(inode->i_state & (I_DIRTY | I_FREEING | I_SYNC)) &&
+	    !atomic_read(&inode->i_count) && inode->i_sb->s_flags & MS_ACTIVE)
+		inode_lru_list_add(inode);
Needs to avoid I_WILL_FREE as well. There's no point putting it on the LRU if we are writing from iput_final()....

Yeah, it won't cause any real problems but you are right it's not useful
to put such inodes into the LRU. I'll send v3 in a moment, now as a proper patch submission since we seem to agree...

Honza

Signature · 2 lines

Jan Kara <[email protected]>
SUSE Labs, CR

Andrew Morton

2012-11-21 08:05 UTC #11 lore raw

In reply to Jan Kara

fs/inode.c static void inode_lru_list_add(struct inode *inode) · line 421

Quoted 11 lines · On Wed, 21 Nov 2012 02:48:51 +0100 Jan Kara <[email protected]> wrote:

fs/inode.c +11 -0

+/*
+ * Add inode to LRU if needed (inode is unused and clean).
+ *
+ * Needs inode->i_lock held.
+ */
+void inode_add_lru(struct inode *inode)
+{
+	if (!(inode->i_state & (I_DIRTY | I_FREEING | I_SYNC)) &&
+	    !atomic_read(&inode->i_count) && inode->i_sb->s_flags & MS_ACTIVE)
+		inode_lru_list_add(inode);
+}

Is i_lock sufficient to stabilise i_count?

Is evict_inodes() wrong to test i_count outside i_lock?

invalidate_inodes() looks better.

can_unuse() must be called under i_lock, and is. Apparently this requirement was sufficiently obvious to not meed documenting.

prune_icache_sb() gets it right.

iput() gets it right.

So to answer my own question: yes, it is sufficient. But a) the comment for inode.i_lock is out of date and b) evict_inodes() looks fishy.

Dave Chinner

2012-11-21 08:22 UTC #12 lore raw

Quoted 16 lines · Andrew Morton

fs/inode.c static void inode_lru_list_add(struct inode *inode) · line 421
Quoted 11 lines · On Wed, 21 Nov 2012 02:48:51 +0100 Jan Kara <[email protected]> wrote:
fs/inode.c +11 -0
+/*
+ * Add inode to LRU if needed (inode is unused and clean).
+ *
+ * Needs inode->i_lock held.
+ */
+void inode_add_lru(struct inode *inode)
+{
+	if (!(inode->i_state & (I_DIRTY | I_FREEING | I_SYNC)) &&
+	    !atomic_read(&inode->i_count) && inode->i_sb->s_flags & MS_ACTIVE)
+		inode_lru_list_add(inode);
+}
Is i_lock sufficient to stabilise i_count?

Quoted 8 lines

<looks at the code a bit>
Is evict_inodes() wrong to test i_count outside i_lock?
invalidate_inodes() looks better.
can_unuse() must be called under i_lock, and is. Apparently this requirement was sufficiently obvious to not meed documenting.

It is documented. can_unuse looks at i_state and i_count, and both are documented as requiring the i_lock at the top of the file in the locking rules section. Also, see __iget(), also mentioned in the locking rules....

Quoted 6 lines

prune_icache_sb() gets it right.
iput() gets it right.
So to answer my own question: yes, it is sufficient. But a) the comment for inode.i_lock is out of date

If you means the one in fs.h, then yeah, it's way out of date....

Quoted 3 lines

and b) evict_inodes() looks
fishy.

As I understand it, evict_inodes() is special - it's only called from generic_shutdown_super() after the MS_ACTIVE flag has been removed from the filesytem, the dcache has been pruned and all the inodes cleaned. So there should be no new references to the inodes occurring, and hence we don't need to hold the lock to serialise against new references being taken....

Cheers,

Dave.

Signature · 2 lines

Dave Chinner
[email protected]

Dave Chinner

2012-11-20 22:37 UTC #13 lore raw

In reply to Jan Kara

fs/fs-writeback.c static void requeue_io(struct inode *inode, struct bdi_writeback *wb) · line 232

Quoted 11 lines · Jan Kara

fs/fs-writeback.c +3 -0

index 51ea267..ed7613b 100644
@@ -227,6 +227,9 @@ static void requeue_io(struct inode *inode, struct bdi_writeback *wb)
 
 static void inode_sync_complete(struct inode *inode)
 {
+	/* If inode is clean an unused, put it into LRU now.  */
+	if (!(inode->i_state & I_DIRTY) && !atomic_read(&inode->i_count))
+		inode_lru_list_add(inode);

Unsafe. Has to avoid I_FREEING|I_WILL_FREE (see warnings at start of writeback_single_inode()).

Also, if you do this after clearing the I_SYNC flag, you can use exactly the same logic as iput_final(). A helper function could be used for both cases at that point, and....

fs/inode.c void ihold(struct inode *inode) · line 400

Quoted 8 lines

fs/inode.c +1 -1

@@ -397,7 +397,7 @@ void ihold(struct inode *inode)
 }
 EXPORT_SYMBOL(ihold);
 
-static void inode_lru_list_add(struct inode *inode)
+void inode_lru_list_add(struct inode *inode)

.... the inode lru list function can stay static.

Cheers,

Dave.

Signature · 2 lines

Dave Chinner
[email protected]

Jan Kara

reviewer

2012-11-21 01:30 UTC #14 lore raw

Quoted 16 lines · Dave Chinner

fs/fs-writeback.c static void requeue_io(struct inode *inode, struct bdi_writeback *wb) · line 232
Quoted 11 lines · Jan Kara
fs/fs-writeback.c +3 -0
index 51ea267..ed7613b 100644
@@ -227,6 +227,9 @@ static void requeue_io(struct inode *inode, struct bdi_writeback *wb)
 
 static void inode_sync_complete(struct inode *inode)
 {
+	/* If inode is clean an unused, put it into LRU now.  */
+	if (!(inode->i_state & I_DIRTY) && !atomic_read(&inode->i_count))
+		inode_lru_list_add(inode);
Unsafe. Has to avoid I_FREEING|I_WILL_FREE (see warnings at start of writeback_single_inode()).

Quoted 3 lines

Also, if you do this after clearing the I_SYNC flag, you can use exactly the same logic as iput_final(). A helper function could be used for both cases at that point, and....

Agreed. That's what I do in the current version of the patch (not posted
yet).

Honza

Signature · 2 lines

Jan Kara <[email protected]>
SUSE Labs, CR