linux-kernel - Re: [PATCH net 1/2] octeontx2-pf: Fix possible memory leak in otx2

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]

Date:   Fri, 2 Dec 2022 09:44:30 +0800
From:   "Ziyang Xuan (William)" <william.xuanziyang@...wei.com>
To:     Maciej Fijalkowski <maciej.fijalkowski@...el.com>,
        <sunil.kovvuri@...il.com>, <sgoutham@...vell.com>
CC:     <gakula@...vell.com>, <sbhatta@...vell.com>, <hkelam@...vell.com>,
        <davem@...emloft.net>, <edumazet@...gle.com>, <kuba@...nel.org>,
        <pabeni@...hat.com>, <netdev@...r.kernel.org>,
        <naveenm@...vell.com>, <rsaladi2@...vell.com>,
        <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH net 1/2] octeontx2-pf: Fix possible memory leak in
 otx2_probe()

>> On Thu, Nov 24, 2022 at 09:56:43AM +0800, Ziyang Xuan wrote:
>>> In otx2_probe(), there are several possible memory leak bugs
>>> in exception paths as follows:
>>> 1. Do not release pf->otx2_wq when excute otx2_init_tc() failed.
>>> 2. Do not shutdown tc when excute otx2_register_dl() failed.
>>> 3. Do not unregister devlink when initialize SR-IOV failed.
>>>
>>> Fixes: 1d4d9e42c240 ("octeontx2-pf: Add tc flower hardware offload on ingress traffic")
>>> Fixes: 2da489432747 ("octeontx2-pf: devlink params support to set mcam entry count")
>>> Signed-off-by: Ziyang Xuan <william.xuanziyang@...wei.com>
>>> ---
>>>  drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c | 5 ++++-
>>>  1 file changed, 4 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
>>> index 303930499a4c..8d7f2c3b0cfd 100644
>>> --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
>>> +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
>>> @@ -2900,7 +2900,7 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
>>>  
>>>  	err = otx2_register_dl(pf);
>>>  	if (err)
>>> -		goto err_mcam_flow_del;
>>> +		goto err_register_dl;
>>>  
>>>  	/* Initialize SR-IOV resources */
>>>  	err = otx2_sriov_vfcfg_init(pf);
>>> @@ -2919,8 +2919,11 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
>>>  	return 0;
>>
>> If otx2_dcbnl_set_ops() fails at the end then shouldn't we also call
>> otx2_sriov_vfcfg_cleanup() ?
> 
> I think it does not need. This is the probe process. PF and VF are all not ready to work,
> so pf->vf_configs[i].link_event_work does not scheduled. And pf->vf_configs memory resource will
> be freed by devm subsystem if probe failed. There are not memory leak and other problems.
> 
Hello Sunil Goutham, Maciej Fijalkowski,

What do you think about my analysis? Look forward to your reply.

Thank you!

> @Sunil Goutham, Please help to confirm.
> 
> Thanks.
> 
>>
>>>  
>>>  err_pf_sriov_init:
>>> +	otx2_unregister_dl(pf);
>>> +err_register_dl:
>>>  	otx2_shutdown_tc(pf);
>>>  err_mcam_flow_del:
>>> +	destroy_workqueue(pf->otx2_wq);
>>>  	otx2_mcam_flow_del(pf);
>>>  err_unreg_netdev:
>>>  	unregister_netdev(netdev);
>>> -- 
>>> 2.25.1
>>>
>> .
>>
> .
>