Received-SPF: pass (google.com: domain of linux-kernel+bounces-47587-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Message-ID: <fd3d7ea4-b1f8-429b-ba0d-588c2bbe25a3@quicinc.com>
Date: Thu, 1 Feb 2024 11:36:18 +0800
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH v1] scsi: ufs: core: Remove the ufshcd_release in
 ufshcd_err_handling_prepare
To: Bart Van Assche <bvanassche@acm.org>, SEO HOYOUNG <hy50.seo@samsung.com>,
        <linux-scsi@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
        <alim.akhtar@samsung.com>, <avri.altman@wdc.com>, <jejb@linux.ibm.com>,
        <martin.petersen@oracle.com>, <beanhuo@micron.com>,
        <kwangwon.min@samsung.com>, <kwmad.kim@samsung.com>,
        <sh425.lee@samsung.com>, <sc.suh@samsung.com>,
        <quic_nguyenb@quicinc.com>, <cpgs@samsung.com>,
        <grant.jung@samsung.com>, <junwoo80.lee@samsung.com>
References: <CGME20240122083204epcas2p26a1bca522e201972ca072e0b24d23a52@epcas2p2.samsung.com>
 <20240122083324.11797-1-hy50.seo@samsung.com>
 <0ae6c12e-3b55-4d90-b042-e8c2b3a2c4a7@acm.org>
Content-Language: en-US
From: Can Guo <quic_cang@quicinc.com>
In-Reply-To: <0ae6c12e-3b55-4d90-b042-e8c2b3a2c4a7@acm.org>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 8bit


On 2/1/2024 1:44 AM, Bart Van Assche wrote:
> On 1/22/24 00:33, SEO HOYOUNG wrote:
>> This is because ufshcd_errhandling_prepare() and
>> ufshcd_err_handling_unprepare() repeatedly release calls.
> 
> It would have been much more clear if it would have been mentioned that
> ufshcd_err_handling_prepare() should call ufshcd_hold() once and
> also that ufshcd_err_handling_unprepare() should call ufshcd_release()
> once.
> 
> Additionally, a Fixes: tag is missing. Is this patch perhaps a fix for 
> commit
> c72e79c0ad2b ("scsi: ufs: Recover HBA runtime PM error in error handler")?
> Can Guo, since you wrote that patch, can you please take a look at this 
> patch?
> 
>> diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c
>> index 7c59d7a02243..423e83074a20 100644
>> --- a/drivers/ufs/core/ufshcd.c
>> +++ b/drivers/ufs/core/ufshcd.c
>> @@ -6351,7 +6351,6 @@ static void ufshcd_err_handling_prepare(struct 
>> ufs_hba *hba)
>>           ufshcd_hold(hba);
>>           if (!ufshcd_is_clkgating_allowed(hba))
>>               ufshcd_setup_clocks(hba, true);
>> -        ufshcd_release(hba);

When we are here, it means runtime resume has failed.

When I wrote the code (in older kernel), ufshcd_runtime_resume(), if 
fails, bails without calling ufshcd_release(), eventually 
ufshcd_err_handling_unprepare() would call ufshcd_release() to balance.

But now, I see that if __ufshcd_wl_resume() fails, ufshcd_release() is 
called anyways, so ufshcd_release() is not required here. Hence,

Reviewed-by: Can Guo <quic_cang@quicinc.com>

>>           pm_op = hba->is_sys_suspended ? UFS_SYSTEM_PM : UFS_RUNTIME_PM;
>>           ufshcd_vops_resume(hba, pm_op);
>>       } else {
> 
> Reviewed-by: Bart Van Assche <bvanassche@acm.org>