Subject: Re: [PATCH 2/2] spi: tegra210-quad: add acpi support
To:     Mark Brown <broonie@kernel.org>,
        Krishna Yarlagadda <kyarlagadda@nvidia.com>
Cc:     Philipp Zabel <p.zabel@pengutronix.de>,
        Laxman Dewangan <ldewangan@nvidia.com>,
        Thierry Reding <thierry.reding@gmail.com>,
        Jonathan Hunter <jonathanh@nvidia.com>,
        Sowjanya Komatineni <skomatineni@nvidia.com>,
        "linux-tegra@vger.kernel.org" <linux-tegra@vger.kernel.org>,
        "linux-spi@vger.kernel.org" <linux-spi@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
References: <1637834152-32093-1-git-send-email-kyarlagadda@nvidia.com>
 <1637834152-32093-2-git-send-email-kyarlagadda@nvidia.com>
 <YZ+ImY1LrvB5a5iF@sirena.org.uk>
 <BN6PR12MB124973BF5CBB4AB35CC59B8AC3669@BN6PR12MB1249.namprd12.prod.outlook.com>
 <YaTHKuohUNt/hVLq@sirena.org.uk>
 <BN6PR12MB124927C4F4FAF53B59C2A23CC3679@BN6PR12MB1249.namprd12.prod.outlook.com>
 <YaYhtFnHlZob9s0J@sirena.org.uk>
From:   Dmitry Osipenko <digetx@gmail.com>
Message-ID: <f6c68e0a-eaa9-12a8-cc44-84b13592c1d9@gmail.com>
Date:   Tue, 30 Nov 2021 17:14:07 +0300
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.14.0
MIME-Version: 1.0
In-Reply-To: <YaYhtFnHlZob9s0J@sirena.org.uk>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Precedence: bulk

30.11.2021 16:05, Mark Brown пишет:
> On Tue, Nov 30, 2021 at 01:50:07AM +0000, Krishna Yarlagadda wrote:
> 
>>> That said frankly I'd expect this handling of ACPI reset to be pushed into the
>>> reset code, it's obviously not good to be open coding this in drivers when this
>>> looks like it's completely generic to any ACPI object so shouldn't be being
>>> open coded in individual driers especially with the ifdefery.  Shouldn't the
>>> reset API be able to figure out that an object with _RST has a reset control
>>> and provide access to it through the reset API?
> 
>> Common reset apis are not handling _RST. Each driver is implementing
>> _RST method in ACPI and calling from drivers.

I see only two or three other drivers in kernel which do that.

> I can see that.  What I'm saying is that this seems bad and we should
> instead be implementing this in common code.
> 

The ifdefery, that this patch has, shouldn't be needed. We have a
similar ACPI patch for Tegra I2C [1] and it doesn't have that.

[1]
https://patchwork.ozlabs.org/project/linux-tegra/patch/1637859224-5179-1-git-send-email-akhilrajeev@nvidia.com/

I assume this patch could be reworked similarly to the I2C patch.

Agree that should be better to have a common reset driver for ACPI
instead of polluting each driver with a boilerplate code.