linux-kernel - Re: [PATCH] erofs: do sanity check on m->type in z_erofs_load_compact

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: <6edacba6-a97e-460f-af06-96e8acf5546a@kernel.org>
Date: Tue, 8 Jul 2025 15:10:19 +0800
From: Chao Yu <chao@...nel.org>
To: Gao Xiang <hsiangkao@...ux.alibaba.com>, Hongbo Li
 <lihongbo22@...wei.com>, xiang@...nel.org
Cc: chao@...nel.org, linux-erofs@...ts.ozlabs.org,
 linux-kernel@...r.kernel.org, Yue Hu <zbestahu@...il.com>,
 Jeffle Xu <jefflexu@...ux.alibaba.com>, Sandeep Dhavale <dhavale@...gle.com>
Subject: Re: [PATCH] erofs: do sanity check on m->type in
 z_erofs_load_compact_lcluster()

On 7/8/25 10:35, Gao Xiang wrote:
> 
> 
> On 2025/7/8 10:30, Hongbo Li wrote:
>>
>>
>> On 2025/7/7 16:47, Chao Yu wrote:
>>> All below functions will do sanity check on m->type, let's move sanity
>>> check to z_erofs_load_compact_lcluster() for cleanup.
>>> - z_erofs_map_blocks_fo
>>> - z_erofs_get_extent_compressedlen
>>> - z_erofs_get_extent_decompressedlen
>>> - z_erofs_extent_lookback
>>>
>>> Signed-off-by: Chao Yu <chao@...nel.org>
>>> ---
>>>   fs/erofs/zmap.c | 60 ++++++++++++++++++-------------------------------
>>>   1 file changed, 22 insertions(+), 38 deletions(-)
>>>
>>> diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c
>>> index 0bebc6e3a4d7..e530b152e14e 100644
>>> --- a/fs/erofs/zmap.c
>>> +++ b/fs/erofs/zmap.c
>>> @@ -240,6 +240,13 @@ static int z_erofs_load_compact_lcluster(struct z_erofs_maprecorder *m,
>>>   static int z_erofs_load_lcluster_from_disk(struct z_erofs_maprecorder *m,
>>>                          unsigned int lcn, bool lookahead)
>>>   {
>>> +    if (m->type >= Z_EROFS_LCLUSTER_TYPE_MAX) {
>>> +        erofs_err(m->inode->i_sb, "unknown type %u @ lcn %u of nid %llu",
>>> +                m->type, lcn, EROFS_I(m->inode)->nid);
>>> +        DBG_BUGON(1);
>>> +        return -EOPNOTSUPP;
>>> +    }
>>> +
>>
>> Hi, Chao,
>>
>> After moving the condition in here, there is no need to check in z_erofs_extent_lookback, z_erofs_get_extent_compressedlen and z_erofs_get_extent_decompressedlen. Because in z_erofs_map_blocks_fo,
>> the condition has been checked in before. Right?
> 
> I've replied some similar question.
> 
> Because z_erofs_get_extent_compressedlen and z_erofs_get_extent_decompressedlen()
> use the different lcn (lcluster) against z_erofs_map_blocks_fo().
> 
> So if a new lcn(lcluster number) is loaded, we'd check if the type is valid.

Yeah, Xiang has noticed that previously [1], the case is as below, so we'd better check the
condition in z_erofs_load_compact_lcluster() rather than z_erofs_map_blocks_fo():

- z_erofs_extent_lookback
 - z_erofs_load_lcluster_from_disk
  - z_erofs_load_full_lcluster
  : m->type = advise & Z_EROFS_LI_LCLUSTER_TYPE_MASK;
 - z_erofs_load_compact_lcluster
 : m->type = type;

[1] https://lore.kernel.org/linux-erofs/04050888-7abf-40fa-98d6-6215b8ba989e@kernel.org/

Thanks,

> 
> Thanks,
> Gao Xiang
> 
>