Received-SPF: pass (google.com: domain of linux-kernel+bounces-92190-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Message-ID: <6670e21a-46d7-db1f-3830-73e97786cde8@quicinc.com>
Date: Tue, 5 Mar 2024 16:24:32 +0530
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:102.0) Gecko/20100101
 Thunderbird/102.13.0
Subject: Re: [PATCH v12 4/9] firmware: qcom: scm: Rework dload mode
 availability check
Content-Language: en-US
To: Bjorn Andersson <andersson@kernel.org>
CC: <konrad.dybcio@linaro.org>, <linux-arm-msm@vger.kernel.org>,
        <linux-kernel@vger.kernel.org>, <linus.walleij@linaro.org>,
        <linux-gpio@vger.kernel.org>, Elliot Berman <quic_eberman@quicinc.com>
References: <20240227155308.18395-1-quic_mojha@quicinc.com>
 <20240227155308.18395-5-quic_mojha@quicinc.com>
 <bmwswdvzqldse4dgbcev7hcyktkzffur3pcs6ogwnekxrklgiu@hridh2l2hhgk>
From: Mukesh Ojha <quic_mojha@quicinc.com>
In-Reply-To: <bmwswdvzqldse4dgbcev7hcyktkzffur3pcs6ogwnekxrklgiu@hridh2l2hhgk>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 7bit


On 3/3/2024 12:46 AM, Bjorn Andersson wrote:
> On Tue, Feb 27, 2024 at 09:23:03PM +0530, Mukesh Ojha wrote:
>> QCOM_SCM_BOOT_SET_DLOAD_MODE was only valid for very older
>> target and firmware and for recent targets there is dload
>> mode tcsr registers available to set the download mode.
>>
> 
> I presume this implies that it will always return false, so what's the
> actual problem with that? Presumably you want this because it takes
> unnecessary time to make that call, if so please say so.

Correct, will add in commit description.

Also, __qcom_scm_is_call_available() usage legacy mode of setting DLOAD 
mode setting via command which is deprecated long back from Trust zone 
firmware also it takes or wait statically global mutex lock at the lower 
level which unnecessary adds time which is of no use.
> 
> 
> Content of the patch looks good.

Thanks.

-Mukesh

> 
> Regards,
> Bjorn
> 
>> So, it is better to keep it as fallback check instead of
>> checking its availability and failing it always.
>>
>> Signed-off-by: Mukesh Ojha <quic_mojha@quicinc.com>
>> Reviewed-by: Elliot Berman <quic_eberman@quicinc.com>
>> ---
>>   drivers/firmware/qcom/qcom_scm.c | 12 +++++-------
>>   1 file changed, 5 insertions(+), 7 deletions(-)
>>
>> diff --git a/drivers/firmware/qcom/qcom_scm.c b/drivers/firmware/qcom/qcom_scm.c
>> index bd6bfdf2d828..3fd89cddba3b 100644
>> --- a/drivers/firmware/qcom/qcom_scm.c
>> +++ b/drivers/firmware/qcom/qcom_scm.c
>> @@ -540,18 +540,16 @@ static int __qcom_scm_set_dload_mode(struct device *dev, bool enable)
>>   static void qcom_scm_set_download_mode(bool enable)
>>   {
>>   	u32 val = enable ? QCOM_DLOAD_FULLDUMP : QCOM_DLOAD_NODUMP;
>> -	bool avail;
>>   	int ret = 0;
>>   
>> -	avail = __qcom_scm_is_call_available(__scm->dev,
>> -					     QCOM_SCM_SVC_BOOT,
>> -					     QCOM_SCM_BOOT_SET_DLOAD_MODE);
>> -	if (avail) {
>> -		ret = __qcom_scm_set_dload_mode(__scm->dev, enable);
>> -	} else if (__scm->dload_mode_addr) {
>> +	if (__scm->dload_mode_addr) {
>>   		ret = qcom_scm_io_rmw(__scm->dload_mode_addr,
>>   				      QCOM_DLOAD_MASK,
>>   				      FIELD_PREP(QCOM_DLOAD_MASK, val));
>> +	} else if (__qcom_scm_is_call_available(__scm->dev,
>> +						QCOM_SCM_SVC_BOOT,
>> +						QCOM_SCM_BOOT_SET_DLOAD_MODE)) {
>> +		ret = __qcom_scm_set_dload_mode(__scm->dev, enable);
>>   	} else {
>>   		dev_err(__scm->dev,
>>   			"No available mechanism for setting download mode\n");
>> -- 
>> 2.43.0.254.ga26002b62827
>>