Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII;
 format=flowed
Content-Transfer-Encoding: 7bit
Date:   Mon, 31 Aug 2020 10:38:54 -0700
From:   nguyenb@codeaurora.org
To:     Bart Van Assche <bvanassche@acm.org>
Cc:     cang@codeaurora.org, asutoshd@codeaurora.org,
        martin.petersen@oracle.com, linux-scsi@vger.kernel.org,
        linux-arm-msm@vger.kernel.org,
        Alim Akhtar <alim.akhtar@samsung.com>,
        Avri Altman <avri.altman@wdc.com>,
        "James E.J. Bottomley" <jejb@linux.ibm.com>,
        Stanley Chu <stanley.chu@mediatek.com>,
        Nitin Rawat <nitirawa@codeaurora.org>,
        Bean Huo <beanhuo@micron.com>,
        open list <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v1 1/1] scsi: ufshcd: Allow zero value setting to
 Auto-Hibernate Timer
In-Reply-To: <56c8bde3-2457-402f-0ad2-94fc1fe12cd5@acm.org>
References: <b141cfcd7998b8933635828b56fbb64f8ad4d175.1598661071.git.nguyenb@codeaurora.org>
 <56c8bde3-2457-402f-0ad2-94fc1fe12cd5@acm.org>
Message-ID: <0914db1a7aaa8b3b528f2298eb213b3c@codeaurora.org>
User-Agent: Roundcube Webmail/1.3.9
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 2020-08-28 20:13, Bart Van Assche wrote:
> On 2020-08-28 18:05, Bao D. Nguyen wrote:
>>  static ssize_t auto_hibern8_show(struct device *dev,
>>  				 struct device_attribute *attr, char *buf)
>>  {
>> +	u32 ahit;
>>  	struct ufs_hba *hba = dev_get_drvdata(dev);
> 
> Although not strictly required for SCSI code, how about following the 
> "reverse
> christmas tree" convention and adding "u32 ahit" below the "hba" 
> declaration?
Thanks for your comment. I will change it.
>>  	if (!ufshcd_is_auto_hibern8_supported(hba))
>>  		return -EOPNOTSUPP;
>> 
>> -	return snprintf(buf, PAGE_SIZE, "%d\n", 
>> ufshcd_ahit_to_us(hba->ahit));
>> +	pm_runtime_get_sync(hba->dev);
>> +	ufshcd_hold(hba, false);
>> +	ahit = ufshcd_readl(hba, REG_AUTO_HIBERNATE_IDLE_TIMER);
>> +	ufshcd_release(hba);
>> +	pm_runtime_put_sync(hba->dev);
>> +
>> +	return scnprintf(buf, PAGE_SIZE, "%d\n", ufshcd_ahit_to_us(ahit));
>>  }
> 
> Why the pm_runtime_get_sync()/pm_runtime_put_sync() and
> ufshcd_hold()/ufshcd_release() calls? I don't think these are necessary 
> here.
We may try to access the hardware register during runtime suspend or UFS 
clock is gated.
UFS clock gating can happen even during runtime resume. Here we are 
trying to prevent NoC error
due to unclocked access.
> Thanks,
> 
> Bart.