Re: [PATCH] ceph: init the i_list/g_list for cap flush

[Xiubo Li <xiubli@redhat.com>]

On 8/25/21 10:31 PM, Ilya Dryomov wrote:
> On Wed, Aug 25, 2021 at 12:50 PM Jeff Layton <jlayton@kernel.org> wrote:
>> On Wed, 2021-08-25 at 13:22 +0800, xiubli@redhat.com wrote:
>>> From: Xiubo Li <xiubli@redhat.com>
>>>
>>> Always init the i_list/g_list in the begining to make sure it won't
>>> crash the kernel if someone want to delete the cap_flush from the
>>> lists.
>>>
>>> Cc: stable@vger.kernel.org
>>> URL: https://tracker.ceph.com/issues/52401
>>> Signed-off-by: Xiubo Li <xiubli@redhat.com>
>>> ---
>>>   fs/ceph/caps.c | 2 +-
>>>   fs/ceph/snap.c | 2 ++
>>>   2 files changed, 3 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
>>> index 4f0dbc640b0b..60f60260cf42 100644
>>> --- a/fs/ceph/caps.c
>>> +++ b/fs/ceph/caps.c
>>> @@ -3666,7 +3666,7 @@ static void handle_cap_flush_ack(struct inode *inode, u64 flush_tid,
>>>        while (!list_empty(&to_remove)) {
>>>                cf = list_first_entry(&to_remove,
>>>                                      struct ceph_cap_flush, i_list);
>>> -             list_del(&cf->i_list);
>>> +             list_del_init(&cf->i_list);
>>>                if (!cf->is_capsnap)
>>>                        ceph_free_cap_flush(cf);
>>>        }
>>> diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
>>> index 62fab59bbf96..b41e6724c591 100644
>>> --- a/fs/ceph/snap.c
>>> +++ b/fs/ceph/snap.c
>>> @@ -488,6 +488,8 @@ static void ceph_queue_cap_snap(struct ceph_inode_info *ci)
>>>                return;
>>>        }
>>>        capsnap->cap_flush.is_capsnap = true;
>>> +     INIT_LIST_HEAD(&capsnap->cap_flush.i_list);
>>> +     INIT_LIST_HEAD(&capsnap->cap_flush.g_list);
>>>
>>>        spin_lock(&ci->i_ceph_lock);
>>>        used = __ceph_caps_used(ci);
>> I'm not certain the second hunk is strictly needed. These either end up
>> on the list or they just get freed. That said, they shouldn't hurt
>> anything and it is more consistent. Merged into testing.
>>
>> Ilya, since this is marked for stable, this probably ought to go to
>> Linus in the last v5.14 pile.
> I'm inclined to fold this into "ceph: correctly handle releasing an
> embedded cap flush" which is already queued for 5.14.

Yeah, that's will be better.


>
> Thanks,
>
>                  Ilya
>