Patchwork ext4: Free resources in the error path of ext4_mb_init.

login
register
mail settings
Submitter Tao Ma
Date Aug. 19, 2011, 3:28 p.m.
Message ID <1313767689-2786-1-git-send-email-tm@tao.ma>
Download mbox | patch
Permalink /patch/110690/
State Accepted
Headers show

Comments

Tao Ma - Aug. 19, 2011, 3:28 p.m.
From: Tao Ma <boyu.mt@taobao.com>

In commit 79a77c5ac, we move ext4_mb_init_backend after the allocation
of s_locality_group to avoid memory leak in error path, but there are
still some other error paths in ext4_mb_init that need to do the same
work. So this patch adds all the error patch for ext4_mb_init. And all
the pointers are reset to NULL in case the caller may double free them.

Signed-off-by: Tao Ma <boyu.mt@taobao.com>
---
 fs/ext4/mballoc.c |   23 +++++++++++++++--------
 1 files changed, 15 insertions(+), 8 deletions(-)
Theodore Ts'o - Oct. 6, 2011, 2:22 p.m.
On Fri, Aug 19, 2011 at 11:28:08PM +0800, Tao Ma wrote:
> From: Tao Ma <boyu.mt@taobao.com>
> 
> In commit 79a77c5ac, we move ext4_mb_init_backend after the allocation
> of s_locality_group to avoid memory leak in error path, but there are
> still some other error paths in ext4_mb_init that need to do the same
> work. So this patch adds all the error patch for ext4_mb_init. And all
> the pointers are reset to NULL in case the caller may double free them.
> 
> Signed-off-by: Tao Ma <boyu.mt@taobao.com>

but I had to reorder the cleanup code for
"out_free_groupinfo_slab" and "out_free_locality_groups" in the
following patch hunk....


> @@ -2513,11 +2512,19 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery)
>  
>  	if (sbi->s_journal)
>  		sbi->s_journal->j_commit_callback = release_blocks_on_commit;
> +
> +	return 0;
> +
> +out_free_groupinfo_slab:
> +	ext4_groupinfo_destroy_slabs();
> +out_free_locality_groups:
> +	free_percpu(sbi->s_locality_groups);
> +	sbi->s_locality_groups = NULL;

Since we first allocate the groupinfo slabs, and then the locality
groups, so the cleanup paths need to do things in the opposite
order.

						- Ted
--
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

Patch

diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 17a5a57..e7d64d8 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -2490,7 +2490,7 @@  int ext4_mb_init(struct super_block *sb, int needs_recovery)
 	sbi->s_locality_groups = alloc_percpu(struct ext4_locality_group);
 	if (sbi->s_locality_groups == NULL) {
 		ret = -ENOMEM;
-		goto out;
+		goto out_free_groupinfo_slab;
 	}
 	for_each_possible_cpu(i) {
 		struct ext4_locality_group *lg;
@@ -2503,9 +2503,8 @@  int ext4_mb_init(struct super_block *sb, int needs_recovery)
 
 	/* init file for buddy data */
 	ret = ext4_mb_init_backend(sb);
-	if (ret != 0) {
-		goto out;
-	}
+	if (ret != 0)
+		goto out_free_locality_groups;
 
 	if (sbi->s_proc)
 		proc_create_data("mb_groups", S_IRUGO, sbi->s_proc,
@@ -2513,11 +2512,19 @@  int ext4_mb_init(struct super_block *sb, int needs_recovery)
 
 	if (sbi->s_journal)
 		sbi->s_journal->j_commit_callback = release_blocks_on_commit;
+
+	return 0;
+
+out_free_groupinfo_slab:
+	ext4_groupinfo_destroy_slabs();
+out_free_locality_groups:
+	free_percpu(sbi->s_locality_groups);
+	sbi->s_locality_groups = NULL;
 out:
-	if (ret) {
-		kfree(sbi->s_mb_offsets);
-		kfree(sbi->s_mb_maxs);
-	}
+	kfree(sbi->s_mb_offsets);
+	sbi->s_mb_offsets = NULL;
+	kfree(sbi->s_mb_maxs);
+	sbi->s_mb_maxs = NULL;
 	return ret;
 }