public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Zizhi Wo <wozizhi@huawei.com>
To: Gao Xiang <hsiangkao@linux.alibaba.com>, <netfs@lists.linux.dev>,
	<dhowells@redhat.com>, <jlayton@kernel.org>, <brauner@kernel.org>
Cc: <jefflexu@linux.alibaba.com>, <zhujia.zj@bytedance.com>,
	<linux-erofs@lists.ozlabs.org>, <linux-fsdevel@vger.kernel.org>,
	<linux-kernel@vger.kernel.org>, <libaokun1@huawei.com>,
	<yangerkun@huawei.com>, <houtao1@huawei.com>,
	<yukuai3@huawei.com>
Subject: Re: [PATCH 0/8] netfs/cachefiles: Some bugfixes
Date: Thu, 10 Oct 2024 12:08:04 +0800	[thread overview]
Message-ID: <90e546b6-ee8a-436f-890b-0e73cb0b1530@huawei.com> (raw)
In-Reply-To: <15a74197-9b84-4d73-a770-8bfc2fde7742@linux.alibaba.com>



在 2024/10/10 11:31, Gao Xiang 写道:
> Hi Zizhi,
> 
> On 2024/10/10 11:08, Zizhi Wo wrote:
>> Hi!
>>
>> This patchset involves some general cachefiles workflows and the on-
>> demand loading process. For example, the eighth patch fixes a memory
>> ordering issue in cachefiles, and the fifth patch includes some cleanup.
>> These all related to changes in the general cachefiles workflow, and I
>> think these deserve some attention.
>>
>> Additionally, although the current EROFS on-demand loading mode based on
>> cachefiles interaction might be considered for switching to the fanotify
>> mode in the future, I believe the code based on the current cachefiles
>> on-demand loading mode still requires maintenance. The first few patches
>> here are bugfixes specifically for that.
> 
> Yes, I also agree with you.  I pinged David weeks ago, because many
> bugfixes are not only impacted to cachefiles on-demand feature but
> also generic cachefiles, hopefully they could be addressed upstream.
> 

Thank you very much for your support and reply!

Thanks,
Zizhi Wo

> Thanks,
> Gao Xiang
>
>>
>> Therefore, I would greatly appreciate it if anyone could take some time
>> to review these patches. So friendly ping.
>>
>> Thanks,
>> Zizhi Wo
>>
>>
>> 在 2024/8/21 10:42, Zizhi Wo 写道:
>>> Hi!
>>>
>>> We recently discovered some bugs through self-discovery and testing in
>>> erofs ondemand loading mode, and this patchset is mainly used to fix
>>> them. These patches are relatively simple changes, and I would be 
>>> excited
>>> to discuss them together with everyone. Below is a brief introduction to
>>> each patch:
>>>
>>> Patch 1: Fix for wrong block_number calculated in ondemand write.
>>>
>>> Patch 2: Fix for wrong length return value in ondemand write.
>>>
>>> Patch 3: Fix missing position update in ondemand write, for scenarios
>>> involving read-ahead, invoking the write syscall.
>>>
>>> Patch 4: Previously, the last redundant data was cleared during the 
>>> umount
>>> phase. This patch remove unnecessary data in advance.
>>>
>>> Patch 5: Code clean up for cachefiles_commit_tmpfile().
>>>
>>> Patch 6: Modify error return value in cachefiles_daemon_secctx().
>>>
>>> Patch 7: Fix object->file Null-pointer-dereference problem.
>>>
>>> Patch 8: Fix for memory out of order in fscache_create_volume().
>>>
>>>
>>> Zizhi Wo (8):
>>>    cachefiles: Fix incorrect block calculations in
>>>      __cachefiles_prepare_write()
>>>    cachefiles: Fix incorrect length return value in
>>>      cachefiles_ondemand_fd_write_iter()
>>>    cachefiles: Fix missing pos updates in
>>>      cachefiles_ondemand_fd_write_iter()
>>>    cachefiles: Clear invalid cache data in advance
>>>    cachefiles: Clean up in cachefiles_commit_tmpfile()
>>>    cachefiles: Modify inappropriate error return value in
>>>      cachefiles_daemon_secctx()
>>>    cachefiles: Fix NULL pointer dereference in object->file
>>>    netfs/fscache: Add a memory barrier for FSCACHE_VOLUME_CREATING
>>>
>>>   fs/cachefiles/daemon.c    |  2 +-
>>>   fs/cachefiles/interface.c |  3 +++
>>>   fs/cachefiles/io.c        | 10 +++++-----
>>>   fs/cachefiles/namei.c     | 23 +++++++++++++----------
>>>   fs/cachefiles/ondemand.c  | 38 +++++++++++++++++++++++++++++---------
>>>   fs/netfs/fscache_volume.c |  3 +--
>>>   6 files changed, 52 insertions(+), 27 deletions(-)
>>>
> 

      reply	other threads:[~2024-10-10  4:08 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-08-21  2:42 [PATCH 0/8] netfs/cachefiles: Some bugfixes Zizhi Wo
2024-08-21  2:42 ` [PATCH 1/8] cachefiles: Fix incorrect block calculations in __cachefiles_prepare_write() Zizhi Wo
2024-10-10 10:34   ` David Howells
2024-10-10 11:11     ` Zizhi Wo
2024-10-10 11:36       ` David Howells
2024-10-10 12:17         ` Zizhi Wo
2024-10-10 13:15         ` Zizhi Wo
2024-08-21  2:42 ` [PATCH 2/8] cachefiles: Fix incorrect length return value in cachefiles_ondemand_fd_write_iter() Zizhi Wo
2024-08-21  2:42 ` [PATCH 3/8] cachefiles: Fix missing pos updates " Zizhi Wo
2024-10-10 11:26   ` David Howells
2024-08-21  2:42 ` [PATCH 4/8] cachefiles: Clear invalid cache data in advance Zizhi Wo
2024-10-10 11:16   ` David Howells
2024-08-21  2:42 ` [PATCH 5/8] cachefiles: Clean up in cachefiles_commit_tmpfile() Zizhi Wo
2024-10-10 11:23   ` David Howells
2024-08-21  2:42 ` [PATCH 6/8] cachefiles: Modify inappropriate error return value in cachefiles_daemon_secctx() Zizhi Wo
2024-10-10 11:31   ` David Howells
2024-10-10 11:47     ` Zizhi Wo
2024-08-21  2:43 ` [PATCH 7/8] cachefiles: Fix NULL pointer dereference in object->file Zizhi Wo
2024-10-10 11:26   ` David Howells
2024-10-10 12:04     ` Zizhi Wo
2024-10-10 14:52       ` David Howells
2024-10-11  1:31         ` Zizhi Wo
2024-08-21  2:43 ` [PATCH 8/8] netfs/fscache: Add a memory barrier for FSCACHE_VOLUME_CREATING Zizhi Wo
2024-10-10 11:24   ` David Howells
2024-10-10  3:08 ` [PATCH 0/8] netfs/cachefiles: Some bugfixes Zizhi Wo
2024-10-10  3:31   ` Gao Xiang
2024-10-10  4:08     ` Zizhi Wo [this message]

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=90e546b6-ee8a-436f-890b-0e73cb0b1530@huawei.com \
    --to=wozizhi@huawei.com \
    --cc=brauner@kernel.org \
    --cc=dhowells@redhat.com \
    --cc=houtao1@huawei.com \
    --cc=hsiangkao@linux.alibaba.com \
    --cc=jefflexu@linux.alibaba.com \
    --cc=jlayton@kernel.org \
    --cc=libaokun1@huawei.com \
    --cc=linux-erofs@lists.ozlabs.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netfs@lists.linux.dev \
    --cc=yangerkun@huawei.com \
    --cc=yukuai3@huawei.com \
    --cc=zhujia.zj@bytedance.com \
    /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