linux-kernel - Re: [PATCH v1 3/9] exfat: reuse cache to improve exfat_get

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID:
 <PUZPR04MB6316842140CBEDFC8E30CDE081B9A@PUZPR04MB6316.apcprd04.prod.outlook.com>
Date: Sun, 4 Jan 2026 07:56:20 +0000
From: "Yuezhang.Mo@...y.com" <Yuezhang.Mo@...y.com>
To: Chi Zhiling <chizhiling@....com>
CC: "brauner@...nel.org" <brauner@...nel.org>,
        "chizhiling@...inos.cn"
	<chizhiling@...inos.cn>,
        "jack@...e.cz" <jack@...e.cz>,
        "linkinjeon@...nel.org" <linkinjeon@...nel.org>,
        "linux-fsdevel@...r.kernel.org" <linux-fsdevel@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "sj1557.seo@...sung.com" <sj1557.seo@...sung.com>,
        "viro@...iv.linux.org.uk"
	<viro@...iv.linux.org.uk>,
        "willy@...radead.org" <willy@...radead.org>
Subject: Re: [PATCH v1 3/9] exfat: reuse cache to improve exfat_get_cluster

> On 12/30/25 17:05, Yuezhang.Mo@...y.com wrote:
>>> -             if (exfat_ent_get(sb, *dclus, &content, NULL))
>>> -                     return -EIO;
>>> +             if (exfat_ent_get(sb, *dclus, &content, &bh))
>>> +                     goto err;
>>
>> As you commented,  the buffer_head needs release if no error return.
>> Here, an error was returned, buffer_head had been released.
>
>
> I mean, it seems like a duplicate release in there, but in fact it's not.
> 
> When exfat_ent_get return an error, *bh is released and set to NULL. So
> the second brelse() call in exfat_get_cluster() does nothing.
>
> ~~~
> int exfat_ent_get(struct super_block *sb, unsigned int loc,
>                  unsigned int *content, struct buffer_head **last)
> {
>
>        ...
>
> err:
>          if (last) {
>                 brelse(*last);
> 
>                 /* Avoid double release */
>                 *last = NULL;
>          }
>          return -EIO;
> }
> ~~~
> 
> The reason using "goto err" is that I want to handle all errors in the
> same way. Although this does seem a bit strange and confusing with the
> comment in exfat_ent_get.

I don't think it's necessary to handle all errors in the same way.

- Only the following error handling requires calling brelse.
    /* prevent the infinite loop of cluster chain */
    if (fclus > limit) {
            exfat_fs_error(sb,
                    "detected the cluster chain loop (i_pos %u)",
                    fclus);
            goto err;
    }
- This makes confused with the comment in exfat_ent_get.
- Unnecessary code modifications are avoided.

>>
>>>
>>>                *last_dclus = *dclus;
>>>                *dclus = content;
>>> @@ -299,7 +300,7 @@ int exfat_get_cluster(struct inode *inode, unsigned int cluster,
>>>                                exfat_fs_error(sb,
>>>                                       "invalid cluster chain (i_pos %u, last_clus 0x%08x is EOF)",
>>>                                       *fclus, (*last_dclus));
>>> -                             return -EIO;
>>> +                             goto err;
>>>                        }
>>>
>>>                        break;
>>> @@ -309,6 +310,10 @@ int exfat_get_cluster(struct inode *inode, unsigned int cluster,
>>>                        cache_init(&cid, *fclus, *dclus);
>>>        }
>>>
>>> +     brelse(bh);
>>>        exfat_cache_add(inode, &cid);
>>>        return 0;
>>> +err:
>>> +     brelse(bh);
>>> +     return -EIO;
>>>   }
>>
>>