linux-kernel - Re: [PATCH for-next 2/3] RDMA/rtrs: Fix rxe_dealloc

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 for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:   Mon, 17 Apr 2023 11:08:09 +0800
From:   Guoqing Jiang <guoqing.jiang@...ux.dev>
To:     "Zhijian Li (Fujitsu)" <lizhijian@...itsu.com>,
        Leon Romanovsky <leon@...nel.org>
Cc:     "haris.iqbal@...os.com" <haris.iqbal@...os.com>,
        "jinpu.wang@...os.com" <jinpu.wang@...os.com>,
        "jgg@...pe.ca" <jgg@...pe.ca>,
        "linux-rdma@...r.kernel.org" <linux-rdma@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH for-next 2/3] RDMA/rtrs: Fix rxe_dealloc_pd warning



On 4/14/23 18:09, Zhijian Li (Fujitsu) wrote:
>
> On 14/04/2023 14:04, Guoqing Jiang wrote:
>>
>> On 4/14/23 13:37, Zhijian Li (Fujitsu) wrote:
>>> On 14/04/2023 11:40, Guoqing Jiang wrote:
>>>> On 4/13/23 16:12, Zhijian Li (Fujitsu) wrote:
>>>>> On 13/04/2023 15:35, Guoqing Jiang wrote:
>>>>>> Hi,
>>>>>>
>>>>>> I take a closer look today.
>>>>>>
>>>>>> On 4/12/23 09:15, Zhijian Li (Fujitsu) wrote:
>>>>>>> On 11/04/2023 20:26, Leon Romanovsky wrote:
>>>>>>>> On Tue, Apr 11, 2023 at 02:43:46AM +0000, Zhijian Li (Fujitsu) wrote:
>>>>>>>>> On 10/04/2023 21:10, Guoqing Jiang wrote:
>>>>>>>>>> On 4/10/23 20:08, Leon Romanovsky wrote:
>>>>>>>>>>> On Mon, Apr 10, 2023 at 06:43:03AM +0000, Li Zhijian wrote:
>>>>>>>>>>>> The warning occurs when destroying PD whose reference count is not zero.
>>>>>>>>>>>>
>>>>>>>>>>>> Precodition: clt_path->s.con_num is 2.
>>>>>>>>>>>> So 2 cm connection will be created as below:
>>>>>>>>>>>> CPU0                                              CPU1
>>>>>>>>>>>> init_conns {                              |
>>>>>>>>>>>>         create_cm() // a. con[0] created        |
>>>>>>>>>>>>                                                 |  a'. rtrs_clt_rdma_cm_handler() {
>>>>>>>>>>>>                                                 |    rtrs_rdma_addr_resolved()
>>>>>>>>>>>>                                                 |      create_con_cq_qp(con); << con[0]
>>>>>>>>>>>>                                                 |  }
>>>>>>>>>>>>                                                 | in this moment, refcnt of PD was increased to 2+
>>>>>> What do you mean "refcnt of PD"? usecnt in struct ib_pd or dev_ref.
>>>>> I mean usecnt in struct ib_pd
>>>>>
>>>>>>>>>>>>                                                 |
>>>>>>>>>>>>         create_cm() // b. cid = 1, failed       |
>>>>>>>>>>>>           destroy_con_cq_qp()                   |
>>>>>>>>>>>>             rtrs_ib_dev_put()                   |
>>>>>>>>>>>>               dev_free()                        |
>>>>>>>>>>>>                 ib_dealloc_pd(dev->ib_pd) << PD |
>>>>>>>>>>>>                  is destroyed, but refcnt is    |
>>>>>>>>>>>>                  still greater than 0           |
>>>>>> Assuming you mean "pd->usecnt". We only allocate pd in con[0] by rtrs_ib_dev_find_or_add,
>>>>>> if con[1] failed to create cm, then alloc_path_reqs -> ib_alloc_mr -> atomic_inc(&pd->usecnt)
>>>> The above can't be invoked, right?
>>>>
>>>>>> can't be triggered. Is there other places could increase the refcnt?
>>>>> Yes, when create a qp, it will also associate to this PD, that also mean refcnt of PD will be increased.
>>>>>
>>>>> When con[0](create_con_cq_qp) succeeded, refcnt of PD will be 2. and then when con[1] failed, since
>>>>> QP didn't create, refcnt of PD is still 2. con[1]'s cleanup will destroy the PD(ib_dealloc_pd) since dev_ref = 1, after that its
>>>>> refcnt is still 1.
>>>> I can see the path increase usecnt to 1.
>>>>
>>>> rtrs_cq_qp_create -> create_qp
>>>>                        -> rdma_create_qp
>>>>                            -> ib_create_qp
>>>>                                -> create_qp
>>>>                                -> ib_qp_usecnt_inc which increases pd->usecnt
>>>>
>>>> Where is another place to increase usecnt to 2?
>>> It should be
>>> ib_create_qp　...
>>>      -> rxe_create_qp
>>>        -> rxe_qp_from_init
>>>           -> rxe_get(pd) <<< pd's refcnt will be increased.
>> Isn't rxe_get just increase elem->ref_cnt?
> Yes, that's true.

I am confused, does increase ref_cnt equal to increase usecnt?
If not, then where is another place to increase usecnt to 2?

BTW, I traced with 6.3-rc5, seems pd's usecnt is only increase once
after create one connection.

[ 6941.525088] in init_conns 2353 con_num=3
[ 6941.525732] in create_con_cq_qp 1648
[ 6941.525944] in rtrs_cq_qp_create 311 con->cid=0 
path->dev->ib_pd->usecnt=1
[ 6941.532460] in create_con_cq_qp 1648
[ 6941.532746] in rtrs_cq_qp_create 311 con->cid=1 
path->dev->ib_pd->usecnt=2
[ 6941.533183] in create_con_cq_qp 1648
[ 6941.533464] in rtrs_cq_qp_create 311 con->cid=2 
path->dev->ib_pd->usecnt=3
[ 6941.533685] in init_conns 2365, clt_path->s.dev->ib_pd->usecnt=3
[ 6941.535680] in init_conns 2371, clt_path->s.dev->ib_pd->usecnt=515

Thanks,
Guoqing