linux-kernel - Re: [PATCH 3/3] mmc: sdhci_am654: Add Initial Support for AM654 SDHCI driver

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <94794998-2c22-1b7f-946c-7887116ee55b@ti.com>
Date:   Mon, 10 Dec 2018 17:34:20 +0530
From:   Faiz Abbas <faiz_abbas@...com>
To:     Adrian Hunter <adrian.hunter@...el.com>,
        <linux-kernel@...r.kernel.org>,
        <linux-arm-kernel@...ts.infradead.org>,
        <devicetree@...r.kernel.org>, <linux-mmc@...r.kernel.org>
CC:     <ulf.hansson@...aro.org>, <robh+dt@...nel.org>,
        <mark.rutland@....com>, <michal.simek@...inx.com>, <kishon@...com>
Subject: Re: [PATCH 3/3] mmc: sdhci_am654: Add Initial Support for AM654 SDHCI
 driver

Hi Adrian,

On 07/12/18 7:02 PM, Adrian Hunter wrote:
> On 5/12/18 5:07 PM, Faiz Abbas wrote:
>> Hi Adrian,
>>
>> On 05/12/18 7:12 PM, Adrian Hunter wrote:
>>> On 29/11/18 6:15 PM, Faiz Abbas wrote:
>>>> The host controllers on TI's AM654 SOCs are not compatible with
>>>> the phy and consumer model of the sdhci-of-arasan driver. It turns out
>>>> that for optimal operation at higher speeds, a special tuning procedure
>>>> needs to be implemented which involves configuration of platform
>>>> specific phy registers.
>>>>
>>>> Therefore, branch out to a new sdhci_am654 driver and add the phy
>>>> register space with all phy configurations to it. Populate AM654
>>>> specific callbacks to sdhci_ops and add SDHCI_QUIRKS wherever
>>>> applicable.
>>>>
>>>> Only add support for upto High Speed for SD card and upto DDR52 speed
>>>> mode for eMMC. Higher speeds will be added in subsequent patches.
>>>>
...
>>>> +
>>>> +	sdhci_am654->clk_ahb = devm_clk_get(dev, "clk_ahb");
>>>> +	if (IS_ERR(sdhci_am654->clk_ahb)) {
>>>> +		dev_err(dev, "clk_ahb clock not found.\n");
>>>> +		ret = PTR_ERR(sdhci_am654->clk_ahb);
>>>> +		goto err_pltfm_free;
>>>> +	}
>>>
>>> Did you intend not to enable clks?
>>
>> Yes. Clocks get enabled as a part of pm_runtime calls.
> 
> Ok, but that could use an explanatory comment.  Also why get a reference to
> clk_ahb if that reference is never used?
> 

You're right. It was being used in sdhci-of-arasan because other users
needed to call enable() and disable(). I missed out on removing it when
porting over. Will remove it and add the comment.

Thanks,
Faiz