Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <4ccefff5-a207-8832-c94f-058757ecf8e8@linaro.org>
Date:   Mon, 16 Jan 2023 14:06:52 +0100
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.6.1
Subject: Re: [PATCH v2 02/10] interconnect: qcom: rpm: make QoS INVALID
 default, separate out driver data
Content-Language: en-US
From:   Konrad Dybcio <konrad.dybcio@linaro.org>
To:     Bryan O'Donoghue <bryan.odonoghue@linaro.org>,
        linux-arm-msm@vger.kernel.org, andersson@kernel.org,
        agross@kernel.org, krzysztof.kozlowski@linaro.org
Cc:     marijn.suijten@somainline.org, Georgi Djakov <djakov@kernel.org>,
        Evan Green <evgreen@chromium.org>,
        Jun Nie <jun.nie@linaro.org>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Brian Masney <masneyb@onstation.org>,
        Yassine Oudjana <y.oudjana@protonmail.com>,
        Dmitry Baryshkov <dmitry.baryshkov@linaro.org>,
        linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org
References: <20230110132202.956619-1-konrad.dybcio@linaro.org>
 <20230110132202.956619-3-konrad.dybcio@linaro.org>
 <d8a7e477-f612-19cb-8573-5cc0449df0fa@linaro.org>
 <22d4bd3b-1b36-8ad4-ca19-157597949a21@linaro.org>
In-Reply-To: <22d4bd3b-1b36-8ad4-ca19-157597949a21@linaro.org>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Precedence: bulk


On 11.01.2023 00:47, Konrad Dybcio wrote:
> 
> 
> On 11.01.2023 00:13, Bryan O'Donoghue wrote:
>> On 10/01/2023 13:21, Konrad Dybcio wrote:
>>> +#define NOC_QOS_MODE_INVALID_VAL    -1
>>> +#define NOC_QOS_MODE_FIXED_VAL        0x0
>>> +#define NOC_QOS_MODE_BYPASS_VAL        0x2
>>
>> The basic fix you are applying here makes sense to me.
>>
>> But why bother with an additional _VAL defintion, you have your enum.
> Thinking about it, I was probably confused by MODE_INVALID checks in
> qcom_icc_set_bimc_qos and only now realized that it's not even called
> with MODE_INVALID.. Will surely fix!
Actually, no.. qcom_icc_set_noc_qos() writes the _VAL to
NOC_QOS_MODEn_ADDR(), so it does matter.

Konrad
> 
> Konrad
>>
>> +enum qos_mode {
>> +    NOC_QOS_MODE_INVALID = 0,
>> +    NOC_QOS_MODE_FIXED,
>> +    NOC_QOS_MODE_BYPASS,
>> +};
>>
>> ---
>> bod