[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <b20a6309-ae31-496d-a1ad-61ea358ad7a9@arm.com>
Date: Tue, 10 Jun 2025 16:09:32 +0530
From: Anshuman Khandual <anshuman.khandual@....com>
To: Leo Yan <leo.yan@....com>, Suzuki K Poulose <suzuki.poulose@....com>
Cc: Mike Leach <mike.leach@...aro.org>, James Clark <james.clark@...aro.org>,
Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>, coresight@...ts.linaro.org,
linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v3 6/9] coresight: Avoid enable programming clock
duplicately
On 09/06/25 10:44 PM, Leo Yan wrote:
> On Mon, Jun 09, 2025 at 05:58:34PM +0100, Suzuki Kuruppassery Poulose wrote:
>
> [...]
>
>>> static inline struct clk *coresight_get_enable_apb_pclk(struct device *dev)
>>> {
>>> - struct clk *pclk;
>>> + struct clk *pclk = NULL;
>>> - pclk = devm_clk_get_enabled(dev, "apb_pclk");
>>> - if (IS_ERR(pclk))
>>> - pclk = devm_clk_get_enabled(dev, "apb");
>>> + if (!dev_is_amba(dev)) {
>>> + pclk = devm_clk_get_enabled(dev, "apb_pclk");
>>> + if (IS_ERR(pclk))
>>> + pclk = devm_clk_get_enabled(dev, "apb");
>>
>> AMBA driver doesn't handle "apb" clock ? So we may need to retain that here
>> ?
>
> Here checks the condition "if (!dev_is_amba(dev))", it means the device
> is not an AMBA device (e.g., a platform device), the APB clock is
> enabled at here.
Just exit early for AMBA devices when 'pclk' clock is still NULL ?
if (dev_is_amba(dev))
return pclk;
>
> Otherwise, for a AMBA device, the AMBA bus layer will help to enable
> APB clock.
>
>> Otherwise looks good to me.
>>
>> Suzuki
>>> + }
>>> return pclk;
>>> }
>>>
>>
Powered by blists - more mailing lists