linux-kernel - Re: [PATCH] ath10k: allocate dummy net

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]

Message-ID: <c9ed5980-c01b-4d20-a5c2-e2ee97373dca@quicinc.com>
Date: Tue, 19 Mar 2024 11:46:42 -0700
From: Jeff Johnson <quic_jjohnson@...cinc.com>
To: Breno Leitao <leitao@...ian.org>
CC: Kalle Valo <kvalo@...nel.org>, Jeff Johnson <jjohnson@...nel.org>,
        <kuba@...nel.org>, <keescook@...omium.org>,
        "open list:NETWORKING DRIVERS
 (WIRELESS)" <linux-wireless@...r.kernel.org>,
        "open list:QUALCOMM ATHEROS
 ATH10K WIRELESS DRIVER" <ath10k@...ts.infradead.org>,
        open list
	<linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] ath10k: allocate dummy net_device dynamically

On 3/19/2024 10:15 AM, Breno Leitao wrote:
> On Tue, Mar 19, 2024 at 09:05:24AM -0700, Jeff Johnson wrote:
>> On 3/19/2024 3:47 AM, Breno Leitao wrote:
>>> @@ -3687,6 +3690,8 @@ struct ath10k *ath10k_core_create(size_t priv_size, struct device *dev,
>>>  
>>>  err_free_coredump:
>>>  	ath10k_coredump_destroy(ar);
>>> +err_free_netdev:
>>> +	free_netdev(ar->napi_dev);
>>>  err_free_tx_complete:
>>>  	destroy_workqueue(ar->workqueue_tx_complete);
>>>  err_free_aux_wq:
>>> @@ -3708,6 +3713,7 @@ void ath10k_core_destroy(struct ath10k *ar)
>>>  
>>>  	destroy_workqueue(ar->workqueue_tx_complete);
>>>  
>>> +	free_netdev(ar->napi_dev);
>>>  	ath10k_debug_destroy(ar);
>>>  	ath10k_coredump_destroy(ar);
>>>  	ath10k_htt_tx_destroy(&ar->htt);
>>
>> looks like there is a pre-existing issue that the order of operations in
>> _destroy() doesn't match the order of operations in the _create() error path.
> 
> Right. I found it weird as well. Basically "ath10k_coredump" and
> "ath10k_debug" operations are swapped between ath10k_core_create() and
> ath10k_core_destroy().
> 
> If you wish, I can submit a patch ordering it properly.

Don't bother. I'll queue that up to fix separately myself

/jeff