Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH] clocksource/drivers/timer-ti-dm: Fix suspend and resume
 for am3 and am4
To:     Tony Lindgren <tony@atomide.com>
CC:     Daniel Lezcano <daniel.lezcano@linaro.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        <linux-kernel@vger.kernel.org>, <linux-omap@vger.kernel.org>,
        <linux-arm-kernel@lists.infradead.org>,
        Carlos Hernandez <ceh@ti.com>
References: <20200713162601.6829-1-tony@atomide.com>
 <1972bace-e9d1-b901-eb33-b4081a4b175d@ti.com>
 <20200715173211.GA10993@atomide.com>
From:   Grygorii Strashko <grygorii.strashko@ti.com>
Message-ID: <cc613f22-5a22-a840-f4e2-cd4650cbd237@ti.com>
Date:   Wed, 15 Jul 2020 21:43:47 +0300
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.8.0
MIME-Version: 1.0
In-Reply-To: <20200715173211.GA10993@atomide.com>
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On 15/07/2020 20:32, Tony Lindgren wrote:
> * Grygorii Strashko <grygorii.strashko@ti.com> [200715 10:17]:
>> On 13/07/2020 19:26, Tony Lindgren wrote:
>>> @@ -653,8 +679,8 @@ static int __init dmtimer_clocksource_init(struct device_node *np)
>>>    	dev->mask = CLOCKSOURCE_MASK(32);
>>>    	dev->flags = CLOCK_SOURCE_IS_CONTINUOUS;
>>> -	if (of_device_is_compatible(np, "ti,am33xx") ||
>>> -	    of_device_is_compatible(np, "ti,am43")) {
>>> +	/* Unlike for clockevent, legacy code sets suspend only for am4 */
>>> +	if (of_machine_is_compatible("ti,am43")) {
>>>    		dev->suspend = dmtimer_clocksource_suspend;
>>>    		dev->resume = dmtimer_clocksource_resume;
>>>    	}
>>>
>>
>> It might be better to use SOC_BUS infra here, which is available on OMAP platforms by default,
>> instead if DT. What do you think?
> 
> Well we have time_init() run very early, and some SoCs have only external
> timers. So timer-ti-dm-systimer must be initialized early. And I think
> soc_device_match() depends on soc_bus_register() that runs later at
> core_initcall().

Ah. right it's early init. I revoke my proposal then.

> 
> If you are thinking of something else, or we can use soc_device_match()
> early, sure :) But sounds like that change should be a separate patch
> then.
> 
> The reason for why we need to check for the SoC, we have the compatible
> for ti,am335x-timer already in use on am3/4 and dm814x, so we can't use
> TIMER_OF_DECLARE alone here.


-- 
Best regards,
grygorii