linux-ext4 - Re: [PATCH 05/10] ext4: Fix race when reusing xattr blocks

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: <aba0cf2a-57c1-699e-5fc8-dd1bddc240c5@huawei.com>
Date:   Tue, 26 Jul 2022 09:14:09 +0800
From:   Zhihao Cheng <chengzhihao1@...wei.com>
To:     Jan Kara <jack@...e.cz>
CC:     Ted Tso <tytso@....edu>, <linux-ext4@...r.kernel.org>,
        Ritesh Harjani <ritesh.list@...il.com>,
        <stable@...r.kernel.org>
Subject: Re: [PATCH 05/10] ext4: Fix race when reusing xattr blocks

在 2022/7/25 23:23, Jan Kara 写道:

>> So far, we have mb_cache_entry_delete_or_get() and
>> mb_cache_entry_wait_unused(), so used cache entry cannot be concurrently
>> removed. Removing check 'hlist_bl_unhashed(&ce->e_hash_list)' is okay.
>>
>> What's affect of changing the other two checks 'ref >=
>> EXT4_XATTR_REFCOUNT_MAX' and '!ce->e_reusable'? To make code more obvious?
>> eg. To point out the condition of 'ref <= EXT4_XATTR_REFCOUNT_MAX' rather
>> than 'ce->e_reusable', we have checked 'ce->e_reusable' in
>> ext4_xattr_block_cache_find() before?
> 
> Well, ce->e_reusable is set if and only if BHDR(new_bh)->h_refcount <
> EXT4_XATTR_REFCOUNT_MAX. So checking whether the refcount is small enough
> is all that is needed and we don't need the ce->e_reusable check here.
> 

Thanks for replying, I get it, thanks.