linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Roel Kluin <12o3l@tiscali.nl>
To: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: alex@clusterfs.com, sct@redhat.com, akpm@linux-foundation.org,
	adilger@clusterfs.com, linux-ext4@vger.kernel.org,
	lkml <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2] mballoc: fix hot spins after err_freebuddy and	err_freemeta
Date: Fri, 18 Apr 2008 14:13:20 +0200	[thread overview]
Message-ID: <48089060.70603@tiscali.nl> (raw)
In-Reply-To: <20080418052103.GB10568@skywalker>

Aneesh Kumar K.V wrote:
> On Fri, Apr 18, 2008 at 10:44:14AM +0530, Aneesh Kumar K.V wrote:
>> On Thu, Apr 17, 2008 at 08:29:18PM +0200, Roel Kluin wrote:
>>> Aneesh Kumar K.V wrote:
>>>
>>>> The function needs more changes. For ex:
>>>>
>>>> 2279     if (meta_group_info[j] == NULL) {
>>>> 2280               printk(KERN_ERR "EXT4-fs: can't allocate buddy mem\n");
>>>> 2281               i--;
>>>> 2282               goto err_freebuddy;
>>>> 2283     }
>>>>
>>>> That decrement  i--; could result in bad value if i == 0;.

>> Won't this also have a memory corruption ?  Let's say we fail in the first
>> loop itslef. That's with i = 0, and since we are using kmalloc.
>> we may find sbi->s_group_info[0] having some random values. So the
>> kfree can crash. Why not a simple change like below ?
>>
> 
> Updated one.
> 

Won't this work as well?
---
In ext4_mb_init_backend() 'i' is of type ext4_group_t. Since unsigned, i >= 0 is
always true, so fix hot spins after err_freebuddy: and -meta: and prevent
decrements when zero.

Signed-off-by: Roel Kluin <12o3l@tiscali.nl>
---
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index ef97f19..054cd33 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -2572,13 +2572,13 @@ static int ext4_mb_init_backend(struct super_block *sb)
 		meta_group_info[j] = kzalloc(len, GFP_KERNEL);
 		if (meta_group_info[j] == NULL) {
 			printk(KERN_ERR "EXT4-fs: can't allocate buddy mem\n");
-			i--;
 			goto err_freebuddy;
 		}
 		desc = ext4_get_group_desc(sb, i, NULL);
 		if (desc == NULL) {
 			printk(KERN_ERR
 				"EXT4-fs: can't read descriptor %lu\n", i);
+			i++;
 			goto err_freebuddy;
 		}
 		memset(meta_group_info[j], 0, len);
@@ -2618,13 +2618,11 @@ static int ext4_mb_init_backend(struct super_block *sb)
 	return 0;
 
 err_freebuddy:
-	while (i >= 0) {
+	while (i-- > 0)
 		kfree(ext4_get_group_info(sb, i));
-		i--;
-	}
 	i = num_meta_group_infos;
 err_freemeta:
-	while (--i >= 0)
+	while (i-- > 0)
 		kfree(sbi->s_group_info[i]);
 	iput(sbi->s_buddy_cache);
 err_freesgi:

      reply	other threads:[~2008-04-18 12:13 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-04-17 16:09 [PATCH] mballoc: fix hot spins after err_freebuddy and err_freemeta Roel Kluin
2008-04-17 17:58 ` Aneesh Kumar K.V
2008-04-17 18:29   ` [PATCH v2] " Roel Kluin
2008-04-18  5:14     ` Aneesh Kumar K.V
2008-04-18  5:21       ` Aneesh Kumar K.V
2008-04-18 12:13         ` Roel Kluin [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=48089060.70603@tiscali.nl \
    --to=12o3l@tiscali.nl \
    --cc=adilger@clusterfs.com \
    --cc=akpm@linux-foundation.org \
    --cc=alex@clusterfs.com \
    --cc=aneesh.kumar@linux.vnet.ibm.com \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sct@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).