[PATCH] ext4: Don't set s_buddy_cache->i_ino to EXT4_BAD_INO

[PATCH] ext4: Don't set s_buddy_cache->i_ino to EXT4_BAD_INO

[Tao Ma]

From: Tao Ma <boyu.mt@taobao.com>

In 85fe4025, Christoph removed get_next_ino() from new_inode() and
do s_buddy_cache->i_ino = get_next_ino();
And then 48e6061b gives EXT4_BAD_INO to this inode. But actually
s_buddy_cache is used directly and we never get it from an inode number.
So it should be safe for us to not set i_ino at all and I guess that
is the case Christoph described in his commit log of 85fe4025
"For a few more filesystems we can avoid assigning any inode number given
that they aren't user visible".

Cc: Yu Jian <yujian@whamcloud.com>
Cc: Andreas Dilger <adilger@whamcloud.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Tao Ma <boyu.mt@taobao.com>
---
 fs/ext4/mballoc.c |    5 -----
 1 files changed, 0 insertions(+), 5 deletions(-)

diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 17a5a57..982f783 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -2342,11 +2342,6 @@ static int ext4_mb_init_backend(struct super_block *sb)
 		ext4_msg(sb, KERN_ERR, "can't get new inode");
 		goto err_freesgi;
 	}
-	/* To avoid potentially colliding with an valid on-disk inode number,
-	 * use EXT4_BAD_INO for the buddy cache inode number.  This inode is
-	 * not in the inode hash, so it should never be found by iget(), but
-	 * this will avoid confusion if it ever shows up during debugging. */
-	sbi->s_buddy_cache->i_ino = EXT4_BAD_INO;
 	EXT4_I(sbi->s_buddy_cache)->i_disksize = 0;
 	for (i = 0; i < ngroups; i++) {
 		desc = ext4_get_group_desc(sb, i, NULL);
-- 
1.7.0.4

Re: [PATCH] ext4: Don't set s_buddy_cache->i_ino to EXT4_BAD_INO

[Andreas Dilger]

On 2011-08-17, at 2:20 AM, Tao Ma <tm@tao.ma> wrote:
> In 85fe4025, Christoph removed get_next_ino() from new_inode() and
> do s_buddy_cache->i_ino = get_next_ino();
> And then 48e6061b gives EXT4_BAD_INO to this inode. But actually
> s_buddy_cache is used directly and we never get it from an inode number.
> So it should be safe for us to not set i_ino at all and I guess that
> is the case Christoph described in his commit log of 85fe4025
> "For a few more filesystems we can avoid assigning any inode number given
> that they aren't user visible".

I don't see that setting i_ino is harmful here (only done once per filesystem mount) and the comment clearly indicates that it is for precautionary reasons only. 

If this inode happens to appear in a crashdump for some reason, I'd prefer it has an inode number that can be identified easily rather than some random value left over from a previously allocated and freed inode, because inode_init_always() does not zero out i_ino. 

> Cc: Yu Jian <yujian@whamcloud.com>
> Cc: Andreas Dilger <adilger@whamcloud.com>
> Cc: "Theodore Ts'o" <tytso@mit.edu>
> Signed-off-by: Tao Ma <boyu.mt@taobao.com>
> ---
> fs/ext4/mballoc.c |    5 -----
> 1 files changed, 0 insertions(+), 5 deletions(-)
> 
> diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
> index 17a5a57..982f783 100644
> --- a/fs/ext4/mballoc.c
> +++ b/fs/ext4/mballoc.c
> @@ -2342,11 +2342,6 @@ static int ext4_mb_init_backend(struct super_block *sb)
>        ext4_msg(sb, KERN_ERR, "can't get new inode");
>        goto err_freesgi;
>    }
> -    /* To avoid potentially colliding with an valid on-disk inode number,
> -     * use EXT4_BAD_INO for the buddy cache inode number.  This inode is
> -     * not in the inode hash, so it should never be found by iget(), but
> -     * this will avoid confusion if it ever shows up during debugging. */
> -    sbi->s_buddy_cache->i_ino = EXT4_BAD_INO;
>    EXT4_I(sbi->s_buddy_cache)->i_disksize = 0;
>    for (i = 0; i < ngroups; i++) {
>        desc = ext4_get_group_desc(sb, i, NULL);
> -- 
> 1.7.0.4
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH] ext4: Don't set s_buddy_cache->i_ino to EXT4_BAD_INO

[Tao Ma]

On 08/17/2011 11:31 PM, Andreas Dilger wrote:
> On 2011-08-17, at 2:20 AM, Tao Ma <tm@tao.ma> wrote:
>> In 85fe4025, Christoph removed get_next_ino() from new_inode() and
>> do s_buddy_cache->i_ino = get_next_ino();
>> And then 48e6061b gives EXT4_BAD_INO to this inode. But actually
>> s_buddy_cache is used directly and we never get it from an inode number.
>> So it should be safe for us to not set i_ino at all and I guess that
>> is the case Christoph described in his commit log of 85fe4025
>> "For a few more filesystems we can avoid assigning any inode number given
>> that they aren't user visible".
> 
> I don't see that setting i_ino is harmful here (only done once per filesystem mount) and the comment clearly indicates that it is for precautionary reasons only. 
> 
> If this inode happens to appear in a crashdump for some reason, I'd prefer it has an inode number that can be identified easily rather than some random value left over from a previously allocated and freed inode, because inode_init_always() does not zero out i_ino. 
Fair enough. Thanks for the explanation.

Thanks
Tao
> 
>> Cc: Yu Jian <yujian@whamcloud.com>
>> Cc: Andreas Dilger <adilger@whamcloud.com>
>> Cc: "Theodore Ts'o" <tytso@mit.edu>
>> Signed-off-by: Tao Ma <boyu.mt@taobao.com>
>> ---
>> fs/ext4/mballoc.c |    5 -----
>> 1 files changed, 0 insertions(+), 5 deletions(-)
>>
>> diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
>> index 17a5a57..982f783 100644
>> --- a/fs/ext4/mballoc.c
>> +++ b/fs/ext4/mballoc.c
>> @@ -2342,11 +2342,6 @@ static int ext4_mb_init_backend(struct super_block *sb)
>>        ext4_msg(sb, KERN_ERR, "can't get new inode");
>>        goto err_freesgi;
>>    }
>> -    /* To avoid potentially colliding with an valid on-disk inode number,
>> -     * use EXT4_BAD_INO for the buddy cache inode number.  This inode is
>> -     * not in the inode hash, so it should never be found by iget(), but
>> -     * this will avoid confusion if it ever shows up during debugging. */
>> -    sbi->s_buddy_cache->i_ino = EXT4_BAD_INO;
>>    EXT4_I(sbi->s_buddy_cache)->i_disksize = 0;
>>    for (i = 0; i < ngroups; i++) {
>>        desc = ext4_get_group_desc(sb, i, NULL);
>> -- 
>> 1.7.0.4
>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html