linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jia Zhu <zhujia.zj@bytedance.com>
To: libaokun@huaweicloud.com, netfs@lists.linux.dev
Cc: dhowells@redhat.com, jlayton@kernel.org,
	jefflexu@linux.alibaba.com, linux-cachefs@redhat.com,
	linux-erofs@lists.ozlabs.org, linux-fsdevel@vger.kernel.org,
	linux-kernel@vger.kernel.org, Hou Tao <houtao1@huawei.com>,
	Baokun Li <libaokun1@huawei.com>,
	zhujia.zj@bytedance.com
Subject: Re: [PATCH 3/5] cachefiles: flush ondemand_object_worker during clean object
Date: Thu, 25 Apr 2024 13:41:20 +0800	[thread overview]
Message-ID: <8572a732-ca12-48d7-817c-d8218d536c0c@bytedance.com> (raw)
In-Reply-To: <20240424033409.2735257-4-libaokun@huaweicloud.com>

Thanks for catching this. How about adding a Fixes tag.

Reviewed-by: Jia Zhu <zhujia.zj@bytedance.com>


在 2024/4/24 11:34, libaokun@huaweicloud.com 写道:
> From: Hou Tao <houtao1@huawei.com>
> 
> When queuing ondemand_object_worker() to re-open the object,
> cachefiles_object is not pinned. The cachefiles_object may be freed when
> the pending read request is completed intentionally and the related
> erofs is umounted. If ondemand_object_worker() runs after the object is
> freed, it will incur use-after-free problem as shown below.
> 
> process A  processs B  process C  process D
> 
> cachefiles_ondemand_send_req()
> // send a read req X
> // wait for its completion
> 
>             // close ondemand fd
>             cachefiles_ondemand_fd_release()
>             // set object as CLOSE
> 
>                         cachefiles_ondemand_daemon_read()
>                         // set object as REOPENING
>                         queue_work(fscache_wq, &info->ondemand_work)
> 
>                                  // close /dev/cachefiles
>                                  cachefiles_daemon_release
>                                  cachefiles_flush_reqs
>                                  complete(&req->done)
> 
> // read req X is completed
> // umount the erofs fs
> cachefiles_put_object()
> // object will be freed
> cachefiles_ondemand_deinit_obj_info()
> kmem_cache_free(object)
>                         // both info and object are freed
>                         ondemand_object_worker()
> 
> When dropping an object, it is no longer necessary to reopen the object,
> so use cancel_work_sync() to cancel or wait for ondemand_object_worker()
> to complete.
> 
> Signed-off-by: Hou Tao <houtao1@huawei.com>
> Signed-off-by: Baokun Li <libaokun1@huawei.com>
> ---
>   fs/cachefiles/ondemand.c | 3 +++
>   1 file changed, 3 insertions(+)
> 
> diff --git a/fs/cachefiles/ondemand.c b/fs/cachefiles/ondemand.c
> index d24bff43499b..f6440b3e7368 100644
> --- a/fs/cachefiles/ondemand.c
> +++ b/fs/cachefiles/ondemand.c
> @@ -589,6 +589,9 @@ void cachefiles_ondemand_clean_object(struct cachefiles_object *object)
>   		}
>   	}
>   	xa_unlock(&cache->reqs);
> +
> +	/* Wait for ondemand_object_worker() to finish to avoid UAF. */
> +	cancel_work_sync(&object->ondemand->ondemand_work);
>   }
>   
>   int cachefiles_ondemand_init_obj_info(struct cachefiles_object *object,

  reply	other threads:[~2024-04-25  5:41 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-24  3:34 [PATCH 0/5] cachefiles: some bugfixes for clean object/send req/poll libaokun
2024-04-24  3:34 ` [PATCH 1/5] cachefiles: stop sending new request when dropping object libaokun
2024-04-24  3:34 ` [PATCH 2/5] cachefiles: flush all requests for the object that is being dropped libaokun
2024-04-24  3:34 ` [PATCH 3/5] cachefiles: flush ondemand_object_worker during clean object libaokun
2024-04-25  5:41   ` Jia Zhu [this message]
2024-04-25  6:53     ` Baokun Li
2024-04-24  3:34 ` [PATCH 4/5] cachefiles: cyclic allocation of msg_id to avoid reuse libaokun
2024-04-24  3:34 ` [PATCH 5/5] cachefiles: add missing lock protection when polling libaokun
2024-04-24  4:29   ` Gao Xiang
2024-04-24  6:23     ` Baokun Li
2024-04-24  5:46   ` Jia Zhu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=8572a732-ca12-48d7-817c-d8218d536c0c@bytedance.com \
    --to=zhujia.zj@bytedance.com \
    --cc=dhowells@redhat.com \
    --cc=houtao1@huawei.com \
    --cc=jefflexu@linux.alibaba.com \
    --cc=jlayton@kernel.org \
    --cc=libaokun1@huawei.com \
    --cc=libaokun@huaweicloud.com \
    --cc=linux-cachefs@redhat.com \
    --cc=linux-erofs@lists.ozlabs.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netfs@lists.linux.dev \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).