linux-kernel - Re: [PATCH v2 2/2] staging: erofs: fix i

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] [day] [month] [year] [list]

Message-ID: <8bbeb607-a18b-aeee-1668-501ad65ba230@huawei.com>
Date:   Tue, 28 May 2019 17:45:25 +0800
From:   Chao Yu <yuchao0@...wei.com>
To:     Dan Carpenter <dan.carpenter@...cle.com>
CC:     Gao Xiang <gaoxiang25@...wei.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        <devel@...verdev.osuosl.org>, <linux-erofs@...ts.ozlabs.org>,
        Chao Yu <chao@...nel.org>, LKML <linux-kernel@...r.kernel.org>,
        <weidu.du@...wei.com>, Fang Wei <fangwei1@...wei.com>,
        Miao Xie <miaoxie@...wei.com>
Subject: Re: [PATCH v2 2/2] staging: erofs: fix i_blocks calculation

On 2019/5/28 14:57, Dan Carpenter wrote:
> On Tue, May 28, 2019 at 11:02:12AM +0800, Chao Yu wrote:
>> On 2019/5/28 10:36, Gao Xiang wrote:
>>> For compressed files, i_blocks should not be calculated
>>> by using i_size. i_u.compressed_blocks is used instead.
>>>
>>> In addition, i_blocks was miscalculated for non-compressed
>>> files previously, fix it as well.
>>>
>>> Signed-off-by: Gao Xiang <gaoxiang25@...wei.com>
>>> ---
>>> change log v2:
>>>  - fix description in commit message
>>>  - fix to 'inode->i_blocks = nblks << LOG_SECTORS_PER_BLOCK'
>>>
>>> Thanks,
>>> Gao Xiang
>>>
>>>  drivers/staging/erofs/inode.c | 14 ++++++++++++--
>>>  1 file changed, 12 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/staging/erofs/inode.c b/drivers/staging/erofs/inode.c
>>> index 8da144943ed6..6e67e018784e 100644
>>> --- a/drivers/staging/erofs/inode.c
>>> +++ b/drivers/staging/erofs/inode.c
>>> @@ -20,6 +20,7 @@ static int read_inode(struct inode *inode, void *data)
>>>  	struct erofs_vnode *vi = EROFS_V(inode);
>>>  	struct erofs_inode_v1 *v1 = data;
>>>  	const unsigned int advise = le16_to_cpu(v1->i_advise);
>>> +	erofs_blk_t nblks = 0;
>>>  
>>>  	vi->data_mapping_mode = __inode_data_mapping(advise);
>>>  
>>> @@ -60,6 +61,10 @@ static int read_inode(struct inode *inode, void *data)
>>>  			le32_to_cpu(v2->i_ctime_nsec);
>>>  
>>>  		inode->i_size = le64_to_cpu(v2->i_size);
>>> +
>>> +		/* total blocks for compressed files */
>>> +		if (vi->data_mapping_mode == EROFS_INODE_LAYOUT_COMPRESSION)
>>> +			nblks = v2->i_u.compressed_blocks;
>>
>> Xiang,
>>
>> It needs to use le32_to_cpu(). ;)
>>
> 
> I wonder it the kbuild bot is going to send an email about that...

0-day may do this a little later.

> Hopefully these sorts of bugs get detected with Sparse CF=-D__CHECK_ENDIAN__

Thanks, Dan, let's use this sparse flag more frequently to avoid such issue.

Thanks,

> 
> regards,
> dan carpenter
> 
> .
>