linux-kernel - Re: [PATCH v2] f2fs: fix potential .flags overflow on 32bit architecture

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]

Date:   Mon, 23 Mar 2020 09:07:04 +0800
From:   Chao Yu <yuchao0@...wei.com>
To:     Joe Perches <joe@...ches.com>, Chao Yu <chao@...nel.org>,
        <jaegeuk@...nel.org>
CC:     <linux-f2fs-devel@...ts.sourceforge.net>,
        <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2] f2fs: fix potential .flags overflow on 32bit
 architecture

On 2020/3/22 22:56, Joe Perches wrote:
> On Sun, 2020-03-22 at 21:56 +0800, Chao Yu wrote:
>> From: Chao Yu <yuchao0@...wei.com>
>>
>> f2fs_inode_info.flags is unsigned long variable, it has 32 bits
>> in 32bit architecture, since we introduced FI_MMAP_FILE flag
>> when we support data compression, we may access memory cross
>> the border of .flags field, corrupting .i_sem field, result in
>> below deadlock.
>>
>> To fix this issue, let's expand .flags as an array to grab enough
>> space to store new flags.
> []
>> +static inline void __set_inode_flag(struct inode *inode, int flag)
>> +{
>> +	if (!test_bit(flag % BITS_PER_LONG,
>> +			&F2FS_I(inode)->flags[BIT_WORD(flag)]))
>> +		set_bit(flag % BITS_PER_LONG,
>> +			&F2FS_I(inode)->flags[BIT_WORD(flag)]);
>> +}
> 
> I believe you don't need to do anything like this
> but just let test_bit and set_bit do the indexing.
> 
> 	if (!test_bit(flg, F2FS_I(inode->flags)))
> 		set_bit(flag, F2FS_I(inode->flags));
> 
> And there already is a function called test_and_set_bit()

Yup, I've cleaned up with test_and_{set,clear}_bit(), thanks for your
reminder.

Thanks,

> 
> 
> .
>