linux-kernel - Re: [PATCH] staging: erofs: don't check special inode layout

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 for Android: free password hash cracker in your pocket

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

Message-ID: <a3743d00-a5c8-6e2a-7b1b-f5111ca59009@huawei.com>
Date:   Sat, 29 Jun 2019 16:38:39 +0800
From:   Chao Yu <yuchao0@...wei.com>
To:     Yue Hu <zbestahu@...il.com>, Gao Xiang <gaoxiang25@...wei.com>
CC:     <gregkh@...uxfoundation.org>, <linux-erofs@...ts.ozlabs.org>,
        <devel@...verdev.osuosl.org>, <linux-kernel@...r.kernel.org>,
        <huyue2@...ong.com>, Miao Xie <miaoxie@...wei.com>
Subject: Re: [PATCH] staging: erofs: don't check special inode layout

On 2019/6/28 12:19, Yue Hu wrote:
> On Fri, 28 Jun 2019 11:50:21 +0800
> Gao Xiang <gaoxiang25@...wei.com> wrote:
> 
>> Hi Yue,
>>
>> On 2019/6/28 11:42, Yue Hu wrote:
>>> From: Yue Hu <huyue2@...ong.com>
>>>
>>> Currently, we will check if inode layout is compression or inline if
>>> the inode is special in fill_inode(). Also set ->i_mapping->a_ops for
>>> it. That is pointless since the both modes won't be set for special
>>> inode when creating EROFS filesystem image. So, let's avoid it.
>>>
>>> Signed-off-by: Yue Hu <huyue2@...ong.com>  
>>
>> Have you test this patch with some actual image with legacy mkfs since
>> new mkfs framework have not supported special inode...
> 
> Hi Xiang,
> 
> I'm studying the testing :)
> 
> However, already check the code handling for special inode in leagcy mkfs as below:
> 
> ```c
>                 break;
>         case EROFS_FT_BLKDEV:
>         case EROFS_FT_CHRDEV:
>         case EROFS_FT_FIFO:
>         case EROFS_FT_SOCK:
>                 mkfs_rank_inode(d);
>                 break;
> 
>         default:
>                 erofs_err("inode[%s] file_type error =%d",
>                           d->i_fullpath,
> ```
> 
> No special inode layout operations, so this change should be fine.
> 
> Thx.
> 
>>
>> I think that is fine in priciple, however, in case to introduce some potential
>> issues, I will test this patch later. I will give a Reviewed-by tag after I tested
>> this patch.

This patch looks good to me, if this won't fail any tests from Xiang, you can add:

Reviewed-by: Chao Yu <yuchao0@...wei.com>

Thanks,

> 
> Thanks.
> 
>>
>> Thanks,
>> Gao Xiang
>>
>>> ---
>>>  drivers/staging/erofs/inode.c | 1 +
>>>  1 file changed, 1 insertion(+)
>>>
>>> diff --git a/drivers/staging/erofs/inode.c b/drivers/staging/erofs/inode.c
>>> index 1433f25..2fe0f6d 100644
>>> --- a/drivers/staging/erofs/inode.c
>>> +++ b/drivers/staging/erofs/inode.c
>>> @@ -205,6 +205,7 @@ static int fill_inode(struct inode *inode, int isdir)
>>>  			S_ISFIFO(inode->i_mode) || S_ISSOCK(inode->i_mode)) {
>>>  			inode->i_op = &erofs_generic_iops;
>>>  			init_special_inode(inode, inode->i_mode, inode->i_rdev);
>>> +			goto out_unlock;
>>>  		} else {
>>>  			err = -EIO;
>>>  			goto out_unlock;
>>>   
> 
> .
>