Message-ID: <65ec1f29-2ad2-acd6-feff-7d1d6f40ef8b@amd.com>
Date:   Tue, 14 Feb 2023 11:43:31 +0530
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.4.2
Subject: Re: [PATCH V2 6/8] soundwire: amd: add runtime pm ops for AMD
 soundwire manager driver
Content-Language: en-US
To:     Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>,
        vkoul@kernel.org
Cc:     amadeuszx.slawinski@linux.intel.com, Mario.Limonciello@amd.com,
        Sunil-kumar.Dommati@amd.com, Basavaraj.Hiregoudar@amd.com,
        Mastan.Katragadda@amd.com, Arungopal.kondaveeti@amd.com,
        Bard Liao <yung-chuan.liao@linux.intel.com>,
        Sanyog Kale <sanyog.r.kale@intel.com>,
        "moderated list:SOUNDWIRE SUBSYSTEM" <alsa-devel@alsa-project.org>,
        open list <linux-kernel@vger.kernel.org>
References: <20230213094031.2231058-1-Vijendar.Mukunda@amd.com>
 <20230213094031.2231058-7-Vijendar.Mukunda@amd.com>
 <383a8166-bc60-8557-e76b-f6287c967598@linux.intel.com>
From:   "Mukunda,Vijendar" <vijendar.mukunda@amd.com>
In-Reply-To: <383a8166-bc60-8557-e76b-f6287c967598@linux.intel.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
MIME-Version: 1.0
X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-MessageData-0: =?utf-8?B?Y3VDMEZzWTVickd3NHpJUGM5Rnl0VjNjUnp4OW1iMEprZXN1bVFHVXlSSW1C?=
 =?utf-8?B?SEgwMkVZeGppMzBvMTUvNW5Iejgza2c0RVVxaDVOVWxuVzFsdTFYOG44UDdG?=
 =?utf-8?B?UjU1SkMzVUwvZktuRms5VzJVZnFrTmhzc1I5MXlCVHJNR3JxSnZTdEQ0RFFU?=
 =?utf-8?B?Y1BNMFVIaTJiME9wbkdWdExYbitROVpacGQ3cW5OeTdid0lwTm40ZmVxVVgr?=
 =?utf-8?B?ckIyblBNRXNacndYN0l5TmQ4STJVb3pGTE5naWxOcyt6aXlJT0psTFpYZURK?=
 =?utf-8?B?NDMvTkprYTVBUTA0clBWbUFMdUE4Qm1EbHhIV2k0ZnIwelhaSnRLczlaa2ZR?=
 =?utf-8?B?aXFlSWhJTURsU2pPdU1NMkVLUDVzSUUrVHJVcFk5Ukp3dWQ2UUF4N0R5aStv?=
 =?utf-8?B?NXVQTWdKQ2JCeHhHaXdMdXpEQVN6QVNGOUJJVERhdlpQaUNzMmRpYnVnamhx?=
 =?utf-8?B?YklobEN4bVFRN3lZVXJvVkYzdGw4d2V5NUlWT3U1ZFNyOUZaL0dEc3k2THQ2?=
 =?utf-8?B?ZlVTMktOTmtaTFU3OUdBZ1l5Nk9MeWp2akdSUmZBUDNsS1RJeThSTnhXOEtG?=
 =?utf-8?B?TnBwd0h6MW9KV2ZybFo2UFo2UGdibG5icFdIeUlWZ3dMbHVQbXdnc2g3c2hk?=
 =?utf-8?B?YTVBbHgybFVEUjJwdnplMkZLR2RYbmltcHh6V0tmUFNzTmxqMzdvOEFZNTY4?=
 =?utf-8?B?SHVodTh3bnh4V0J6c1I5L2JRZTNubXZwbmpicW85UDRLOVJiWnFUeVhXWSsw?=
 =?utf-8?B?MGxLMnVScUU4MmhjMkRadEtoeXFCeXk3R29PS0RQVk8wNHhoUjR1T3hJdG9r?=
 =?utf-8?B?b1lQK09hdnJ4amJqdWtIak9sYkFCakxmS2xWTTRVbjIzc2FBYVdwVWp4RTdO?=
 =?utf-8?B?cmJESWpsb1QvM3VtU1RPOG5mUHJ5MldCYTFwNkJVNHVEQ3l4WWp2VmZodS85?=
 =?utf-8?B?UlhjOVJpWDRrMWtzN1FqOVlUUTNiSU05MzhiR3h6eC9zQWZyYWJNMmpPcy96?=
 =?utf-8?B?cWIzQjBwcWNzOENISy8vTDNnbjJSRlhsYW9yZ3o3d1NLSFl6OVMyMVd3Tmds?=
 =?utf-8?B?SjJBSjBQM1ErWTYxQTVtekNFaXlRakl2dGs0dk5EdXI1ZXB1bUt5SHJTdnEw?=
 =?utf-8?B?S3RpdVduQmVlVFRNQTM4TE4vRGdYNC9OYlBrNjNicGs1SkdCS1lJT3hJZE9S?=
 =?utf-8?B?NE9iaHd2L2FzK1hsN3FVVm1kR1JkZFI5TzduUGZ1REE1cmhhdDhwNHRMVmpU?=
 =?utf-8?B?eWM2TzU0eEdWbnh4RlZUSFEra0pSQy9xK1E1RHR5aVVpWWZ3amE0M0FsYUp1?=
 =?utf-8?B?Wkk5ODhMSDZEdUVhcHppNXpqKzBOUjU5bWY0Qk9EU1pGdktSa3RCZTh3QU1D?=
 =?utf-8?B?YmpUWHg2cnUzOExaNHFKUXB2V1l3dFE1aTY1N0ZBNTYzWGlMQmRnTnpVN2ho?=
 =?utf-8?B?emxWZWFCemduT2svaElBWU1jeTN0VS9xZS90S1hUbGRsQmhvdFAvb1YzMlVY?=
 =?utf-8?B?SGI3NHMzcUpHcUZmdCtaMW5DS3d2czZUNXRzU2Z5VjUzb2VOSGJnRitMSGlj?=
 =?utf-8?B?WHJIUkRQejF4OEl5VmFsTEYrQ0JlVkV6eGNUZVRNWGFFYzVRTElqQy91Y0xn?=
 =?utf-8?B?dFIrU0lRODMycHh2NzJ5Q0svSjFGYzh0RFZYS3F1UUUrQStCbytMOHBjenQy?=
 =?utf-8?B?TkZjakpGOVQyd1drdGd3c29UQk1laEpGY1pocEl1Z2lFOTdKdFRSQTE2WDVN?=
 =?utf-8?B?MXQ4UnJVK3I0M2xOQzZnSWkvVldjcGpUT0huTnY2TVBvUjV4WkdzM1ZnbEtX?=
 =?utf-8?B?NHBsUld1L25MWkhXbXdZbUkwaWNkZExOcEtwL0lsWitmMlQ5N3FqdDRmd0Iy?=
 =?utf-8?B?aEh6ZEd0S1A1NU9MTmIvRHBYRWpWR3RLTkhITTNoNVdaQ0J4a3hXVkxSR2RJ?=
 =?utf-8?B?TS9nMXNsK3MzbG11YmRBNnhnalFONFFWdkhCUWdSVndaUVRuVnFWSDlyNjNH?=
 =?utf-8?B?YWNOQ3UxaklxbWFGelVZRjhZUTVYUVBmKzJyTVV4aUIxSlpVVmxqaDlnTVFh?=
 =?utf-8?B?QmhkNFhudTNKQU0xc0tLeDFpNFEwZ1lvNnBXMFZGYzVkT05LU0g4VUZMSzM0?=
 =?utf-8?Q?H55+Sv3vnP+cw4h0binS5Q6rL?=
X-OriginatorOrg: amd.com
X-MS-Exchange-CrossTenant-Network-Message-Id: e171d705-f333-4546-aa20-08db0e5231a7
X-MS-Exchange-CrossTenant-AuthSource: DM6PR12MB4123.namprd12.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 14 Feb 2023 06:10:38.9828
 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-CrossTenant-Id: 3dd8961f-e488-4e60-8e11-a82d994e183d
X-MS-Exchange-CrossTenant-MailboxType: HOSTED
X-MS-Exchange-CrossTenant-UserPrincipalName: ylLClAsM9RInANRp6V6ViHeWUECrWaSl3m28k17UnpIcMHUW11EHvBYODS9Pn8I7taR13PWNxtTAWhSK793JKg==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: DM6PR12MB4284
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On 13/02/23 23:50, Pierre-Louis Bossart wrote:
>
> On 2/13/23 03:40, Vijendar Mukunda wrote:
>> Add support for runtime pm ops for AMD soundwire manager driver.
>>
>> Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
>> Signed-off-by: Mastan Katragadda <Mastan.Katragadda@amd.com>
>> ---
>>  drivers/soundwire/amd_manager.c   | 163 ++++++++++++++++++++++++++++++
>>  drivers/soundwire/amd_manager.h   |   3 +
>>  include/linux/soundwire/sdw_amd.h |  16 +++
>>  3 files changed, 182 insertions(+)
>>
>> diff --git a/drivers/soundwire/amd_manager.c b/drivers/soundwire/amd_manager.c
>> index 87f9a987d93a..eced189ba6e0 100644
>> --- a/drivers/soundwire/amd_manager.c
>> +++ b/drivers/soundwire/amd_manager.c
>> @@ -14,6 +14,7 @@
>>  #include <linux/slab.h>
>>  #include <linux/soundwire/sdw.h>
>>  #include <linux/soundwire/sdw_registers.h>
>> +#include <linux/pm_runtime.h>
>>  #include <linux/wait.h>
>>  #include <sound/pcm_params.h>
>>  #include <sound/soc.h>
>> @@ -185,6 +186,15 @@ static void amd_disable_sdw_interrupts(struct amd_sdw_manager *amd_manager)
>>  	acp_reg_writel(0x00, amd_manager->mmio + ACP_SW_ERROR_INTR_MASK);
>>  }
>>  
>> +static int amd_deinit_sdw_manager(struct amd_sdw_manager *amd_manager)
>> +{
>> +	int ret;
>> +
>> +	amd_disable_sdw_interrupts(amd_manager);
>> +	ret = amd_disable_sdw_manager(amd_manager);
>> +	return ret;
>> +}
>> +
>>  static void amd_sdw_set_frameshape(struct amd_sdw_manager *amd_manager)
>>  {
>>  	u32 frame_size;
>> @@ -1043,6 +1053,12 @@ static int amd_sdw_manager_probe(struct platform_device *pdev)
>>  	INIT_WORK(&amd_manager->amd_sdw_work, amd_sdw_update_slave_status_work);
>>  	INIT_WORK(&amd_manager->probe_work, amd_sdw_probe_work);
>>  	schedule_work(&amd_manager->probe_work);
>> +	/* Enable runtime PM */
>> +	pm_runtime_set_autosuspend_delay(dev, AMD_SDW_MASTER_SUSPEND_DELAY_MS);
>> +	pm_runtime_use_autosuspend(dev);
>> +	pm_runtime_mark_last_busy(dev);
>> +	pm_runtime_set_active(dev);
>> +	pm_runtime_enable(dev);
> that doesn't sound good to me, why do this here and not in the work
> function? That creates a racy case where the device might suspend before
> being initialized.
This can be moved to work function.
>
>>  	return 0;
>>  }
>>  
>> @@ -1057,14 +1073,161 @@ static int amd_sdw_manager_remove(struct platform_device *pdev)
>>  	amd_disable_sdw_interrupts(amd_manager);
>>  	sdw_bus_master_delete(&amd_manager->bus);
>>  	ret = amd_disable_sdw_manager(amd_manager);
>> +	pm_runtime_disable(&pdev->dev);
> shouldn't you do the pm_runtime_disable first?
I agree. Will fix it.
>
>>  	return ret;
>>  }
>> +/* AMD pm_runtime quirk definitions */
>> +
>> +/*
>> + * Force the clock to stop(ClockStopMode0) when suspend callback
>> + * is invoked.
>> + */
>> +#define AMD_SDW_CLK_STOP_MODE		1
>> +
>> +/*
>> + * Stop the bus when runtime suspend/system level suspend callback
>> + * is invoked. If set, a complete bus reset and re-enumeration will
>> + * be performed when the bus restarts.
>> + */
>> +#define AMD_SDW_POWER_OFF_MODE		2
> You need to clarify this mode, can you deal with device in-band wakes if
> the power is off?
On the Current platform, in-band wakes are not supported if the power is off.


>
>>  #define ACP_SDW0	0
>>  #define ACP_SDW1	1
>>  
>> @@ -57,6 +71,7 @@ struct sdw_amd_dai_runtime {
>>   * @instance: soundwire manager instance
>>   * @quirks: soundwire manager quirks
>>   * @wake_en_mask: wake enable mask per soundwire manager
>> + * @clk_stopped: flag set to true when clock is stopped
>>   * @power_mode_mask: flag interprets amd soundwire manager power mode
>>   * @dai_runtime_array: dai runtime array
>>   */
>> @@ -86,6 +101,7 @@ struct amd_sdw_manager {
>>  	u32 quirks;
>>  	u32 wake_en_mask;
>>  	u32 power_mode_mask;
>> +	bool clk_stopped;
>>  
>>  	struct sdw_amd_dai_runtime **dai_runtime_array;
>>  };