Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Sender: seevalam=codeaurora.org@mg.codeaurora.org
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII;
 format=flowed
Content-Transfer-Encoding: 7bit
Date:   Wed, 22 Sep 2021 21:23:32 +0530
From:   Seevalamuthu Mariappan <seevalam@codeaurora.org>
To:     Kalle Valo <kvalo@codeaurora.org>
Cc:     Jouni Malinen <jouni@codeaurora.org>, ath11k@lists.infradead.org,
        linux-wireless@vger.kernel.org, Miles Hu <milehu@codeaurora.org>,
        Vasanthakumar Thiagarajan <vthiagar@codeaurora.org>
Subject: Re: [PATCH 2/3] ath11k: add separate APIs for monitor mode
In-Reply-To: <87tuidudmi.fsf@codeaurora.org>
References: <20210721162053.46290-1-jouni@codeaurora.org>
 <20210721162053.46290-3-jouni@codeaurora.org>
 <877dfaukjq.fsf@codeaurora.org> <87tuidudmi.fsf@codeaurora.org>
Message-ID: <58b100804acf0afc1c8baf4840ed65d8@codeaurora.org>
User-Agent: Roundcube Webmail/1.3.9
Precedence: bulk

On 2021-09-21 21:25, Kalle Valo wrote:
> Kalle Valo <kvalo@codeaurora.org> writes:
> 
>>> +vdev_stop:
>>> +	reinit_completion(&ar->vdev_setup_done);
>>> +
>>> +	ret = ath11k_wmi_vdev_stop(ar, vdev_id);
>>> +	if (ret) {
>>> +		ath11k_warn(ar->ab, "failed to stop monitor vdev %i after start 
>>> failure: %d\n",
>>> +			    vdev_id, ret);
>>> +		return ret;
>>> +	}
>>> +
>>> +	ret = ath11k_mac_vdev_setup_sync(ar);
>>> +	if (ret)
>>> +		ath11k_warn(ar->ab, "failed to synchronize setup for vdev %i stop: 
>>> %d\n",
>>> +			    vdev_id, ret);
>> 
>> I added return ret here for consistency..
>> 
>>> +	return ret;
>> 
>> I don't thinks this is right, in an error path (vdev_stop label) we
>> return 0? I changed this to -EIO.
>> 
>>> +static int ath11k_mac_monitor_vdev_stop(struct ath11k *ar)
>>> +{
>>> +	int ret;
>>> +
>>> +	lockdep_assert_held(&ar->conf_mutex);
>>> +
>>> +	reinit_completion(&ar->vdev_setup_done);
>>> +
>>> +	ret = ath11k_wmi_vdev_stop(ar, ar->monitor_vdev_id);
>>> +	if (ret)
>>> +		ath11k_warn(ar->ab, "failed to request monitor vdev %i stop: 
>>> %d\n",
>>> +			    ar->monitor_vdev_id, ret);
>>> +
>>> +	ret = ath11k_mac_vdev_setup_sync(ar);
>>> +	if (ret)
>>> + ath11k_warn(ar->ab, "failed to synchronize monitor vdev %i stop:
>>> %d\n",
>>> +			    ar->monitor_vdev_id, ret);
>>> +
>>> +	ret = ath11k_wmi_vdev_down(ar, ar->monitor_vdev_id);
>>> +	if (ret)
>>> +		ath11k_warn(ar->ab, "failed to put down monitor vdev %i: %d\n",
>>> +			    ar->monitor_vdev_id, ret);
>>> +
>>> +	ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac monitor vdev %i stopped\n",
>>> +		   ar->monitor_vdev_id);
>>> +	return ret;
>>> +}
>> 
>> I was not sure what's the idea of error path handling here, we print
>> warnings but still continue the normal execution. I changed this so 
>> that
>> we bail out in the first error and if everything goes well we return 
>> 0.
> 
> I found quite a few missing error checks, too many to list here but
> fixed in the pending branch:
> 
> https://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git/commit/?h=pending&id=8b2f8d11422e7909ff02db456cda41728f621de4
> 
> https://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git/commit/?h=pending&id=5ff318be206b3d2a0bfdcfaf0ac52cc3b4ecdeae
> 
> Please double check, compile tested only.

Thanks for the fixes Kalle. It looks fine and tested the same.

Regards,
Seevalamuthu M