Received-SPF: pass (google.com: domain of linux-kernel+bounces-41543-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Message-ID: <1162e59e-3f53-4c84-b186-bb2f4642216b@amd.com>
Date: Sat, 27 Jan 2024 23:38:28 -0600
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH] drm/amd/display: add panel_power_savings sysfs entry to
 eDP connectors
Content-Language: en-US
From: Mario Limonciello <mario.limonciello@amd.com>
To: Hamza Mahfooz <hamza.mahfooz@amd.com>, amd-gfx@lists.freedesktop.org
Cc: Harry Wentland <harry.wentland@amd.com>, Leo Li <sunpeng.li@amd.com>,
 Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>,
 Alex Deucher <alexander.deucher@amd.com>,
 =?UTF-8?Q?Christian_K=C3=B6nig?= <christian.koenig@amd.com>,
 "Pan, Xinhui" <Xinhui.Pan@amd.com>, David Airlie <airlied@gmail.com>,
 Daniel Vetter <daniel@ffwll.ch>, Alex Hung <alex.hung@amd.com>,
 Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>,
 Wayne Lin <wayne.lin@amd.com>, dri-devel@lists.freedesktop.org,
 linux-kernel@vger.kernel.org
References: <20240126222300.119292-1-hamza.mahfooz@amd.com>
 <1b8de155-4788-4c12-9c40-d45e508a526c@amd.com>
In-Reply-To: <1b8de155-4788-4c12-9c40-d45e508a526c@amd.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk
MIME-Version: 1.0
X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-MessageData-0:
	=?utf-8?B?MSsrcmdwRDM1eHRySlpjTkZvOHR3WjlYTjRSU2xMTUExWDdpS1pvenIyOTI2?=
 =?utf-8?B?TkV2VkpkdEFOL2xrMGRtOWorc3g4UnRkUXkvOS9lVWZPU0FIeWJQTkRraDlZ?=
 =?utf-8?B?UjZ6S1hxOHBxQlNDSDF0MWR5QzJCN2drR0JMRlV4OEptdEFydXNheFVkRUtV?=
 =?utf-8?B?Q280VkxDV1hBbWRYcURyN2wrUDNUWHVjWlIzUkhtOFdsU29oT3lwV3FsQUp4?=
 =?utf-8?B?dys3VWppTjFHbXhWYStnTmFXT08xSHNkZjZHWG81dm5XTzd4R2dZZEVpQ3dT?=
 =?utf-8?B?c0w3SmdjRFVwVVBwb0YrZWRpQ2dhSHA4U1RwT3NmR0l3SEZySmYwMGJveGpX?=
 =?utf-8?B?ZitkdXc4bU0yb1NPbUxlcGlMWVJLelVhUlpEYVFNZnhJTlR4WklqbzRUUEZp?=
 =?utf-8?B?M0VqNjF3UFV6K255QkxuYTc5ME5jV3JBbFQ5eUg5VFBRVWN6UzV2SVpudnZI?=
 =?utf-8?B?cGRpUDNnSXFoZzhaREExbUVyM05DZk5GTTEwQk1mQjVxTHdLekVEMWtJM25z?=
 =?utf-8?B?cUxscnNkM1c5VStVRGVUUk1RUGRmMERjMGJoLzNnbkNqZ2p6a08rYmFRUzFJ?=
 =?utf-8?B?YitUMTRsaS9wcCtlUUdHSVdXS3ErSmRmMkxja3NJaklXTFdaOGt5NDhIOTYv?=
 =?utf-8?B?akZzblEyRm5UMk40ZFF0cDhHTk95SHplaDNncjJQcGZqVnFGemtQMWRiWTNv?=
 =?utf-8?B?S0k4bnFWL1lhTzNvM3prUldiWGlEQ1VVeit1b3RCUnpEcHJPQVF6eTJzYUwx?=
 =?utf-8?B?UVR6VExMVEMyQTM1OTU5clJDYnNOZVFjcWZnWUYwanBsYUZLY1J5bHlkb2JW?=
 =?utf-8?B?OTYxcjY2VmkybmROdGp1NzF4UjZtL0piZWZ1RDJCVUtxYXJxYXhrNEkzek4y?=
 =?utf-8?B?U3hINFVzdllWUGVoY1R6NmxhZndHU0tQR0dUWHF4QXlCcDh0eTZYN1VBeGxC?=
 =?utf-8?B?L081bHhHUkdYc25lVGhHaFdzajJyZVFkUWRDVHRMdnN0RzNhd1c1dFNhM2hp?=
 =?utf-8?B?QlRjSzBSK2pkSnEzaGpmNkJlQmVGQ0l4VmNjdjlsRUVRRy95TkMyN3hTb2VD?=
 =?utf-8?B?Tm0wWWt2a2tZdHlMWnA3dlNsZllacS9JM3Iyd1NOU0w3eEtmVU9TeHRFdnRj?=
 =?utf-8?B?V2hSbEZCT3c0MGtteW1KT3dpOHpVb2VqL3F6QnFkMW44NExwMDJUTXdxZVF3?=
 =?utf-8?B?YTRRMGliaC9SWWdSdjZiZndsOFBQZHVyQnhwTFdzVnZicnEvTTBubENjZWhP?=
 =?utf-8?B?NWZZbU1FVUMyZzNLWVlkcGpySnorSm5ncTQxdnBrcStNMnArbXpHWHczUjRS?=
 =?utf-8?B?WHNRVVFkdTloUjNYRTVoYnprYzhBRHNLcTNLbU9RNWdlMmlweFUxOUVaKzJK?=
 =?utf-8?B?djY0bVJWcCt6cEZJTEVhMVNVRWMyN0s5cExxR3F5VldENXVYWW9FQWZiZEll?=
 =?utf-8?B?Z0hiTUhEeWQwc1BsUzRwbmgxcysvdWhBaXJ1VFdEWWxVU2JEYkRwaDhqb05E?=
 =?utf-8?B?Njc4NGRNZE1sL2pKTStDUkgvWFFCSXZIYmRZRmlyZW1saXhvL3krc3EvOWxj?=
 =?utf-8?B?Um9EdUlOV2cwRHRId3NzYWZ4V1RCSXptNXVQc0tmNy9OZ293UDlMN0krVXJI?=
 =?utf-8?B?ekR2aWpsSWgvYTVnVVdLQS9CSEZrMXZvR0xDbGU3RXdGaTRrZ0gwbU84U3V6?=
 =?utf-8?B?KzR3Q0JHSXJGNWluNXpLUnM4MkRWOUhUUzJpcTI4YmZUVWFoNTBCc1RZdTk5?=
 =?utf-8?B?N1piNlVndC9PNkVYK0theVBEcGRtQW4vaEtqL1pYVnlDcGNJeE5IbXdZRDJt?=
 =?utf-8?B?WkRPK0xJQkx3dU1acFNSUWJtV0hOTldIM0lCWXpheEdzbTdFaU5melJ0cEVT?=
 =?utf-8?B?VndZcndZTStSMTZRZVc5NERiZHViUDBtZytDdUFmR1E2OVdTN0hud3ZuVXhs?=
 =?utf-8?B?ZXdER1BWbWYrb24rNjQ2Y1ltd0tDM0xNN2pnb1ZxVjFLNkxyanZtVHNhR0NJ?=
 =?utf-8?B?VmowMlVreHZBc25jRS9WeXNYbFRXVjFTTFBWWm5xNGx6VnlIRGZCR2MvbjRy?=
 =?utf-8?B?ODBBRlExQ292SWJBZG5BVG1yVk8wRGkrU0pSNFBTNy9ZTm1zN1JkQ1dBVG9J?=
 =?utf-8?Q?3nLHrHZx5dKcjt2gslCz5cYlr?=
X-OriginatorOrg: amd.com
X-MS-Exchange-CrossTenant-Network-Message-Id: e404e8af-5259-4dce-399c-08dc1fc35cae
X-MS-Exchange-CrossTenant-AuthSource: MN0PR12MB6101.namprd12.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 28 Jan 2024 05:38:31.4962
 (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: y0E2eAWyNJJlpK+9T6P45foWK5KmGVIPVYeJWKoT8KDd6LNyBcLtNjeYnGeIZyx9pP4cJv9w8/GQDK8j5BLSHQ==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: DS7PR12MB6118

On 1/26/2024 16:34, Mario Limonciello wrote:
> On 1/26/2024 16:22, Hamza Mahfooz wrote:
>> We want programs besides the compositor to be able to enable or disable
>> panel power saving features. However, since they are currently only
>> configurable through DRM properties, that isn't possible. So, to remedy
>> that issue introduce a new "panel_power_savings" sysfs attribute.
>>
>> Cc: Mario Limonciello <mario.limonciello@amd.com>
>> Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
>> ---
>>   .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 59 +++++++++++++++++++
>>   1 file changed, 59 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c 
>> b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
>> index cd98b3565178..b3fcd833015d 100644
>> --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
>> +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
>> @@ -6534,6 +6534,58 @@ 
>> amdgpu_dm_connector_atomic_duplicate_state(struct drm_connector 
>> *connector)
>>       return &new_state->base;
>>   }
>> +static ssize_t panel_power_savings_show(struct device *device,
>> +                    struct device_attribute *attr,
>> +                    char *buf)
>> +{
>> +    struct drm_connector *connector = dev_get_drvdata(device);
>> +    struct drm_device *dev = connector->dev;
>> +    ssize_t val;
>> +
>> +    drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
>> +    val = to_dm_connector_state(connector->state)->abm_level;
>> +    drm_modeset_unlock(&dev->mode_config.connection_mutex);
>> +
>> +    return sysfs_emit(buf, "%lu\n", val);

When the system is first booted up with nothing on the kernel command 
line, this emits the value for ABM_LEVEL_IMMEDIATE_DISABLE, which isn't 
something we normally should be showing to people.

I think you should be special casing it similar to how the store special 
case works.

>> +}
>> +
>> +static ssize_t panel_power_savings_store(struct device *device,
>> +                     struct device_attribute *attr,
>> +                     const char *buf, size_t count)
>> +{
>> +    struct drm_connector *connector = dev_get_drvdata(device);
>> +    struct drm_device *dev = connector->dev;
>> +    long val;
>> +    int ret;
>> +
>> +    ret = kstrtol(buf, 0, &val);
>> +
>> +    if (ret)
>> +        return ret;
>> +
>> +    if (val < 0 || val > 4)
>> +        return -EINVAL;
>> +
>> +    drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
>> +    to_dm_connector_state(connector->state)->abm_level = val ?:
>> +        ABM_LEVEL_IMMEDIATE_DISABLE;
>> +    drm_modeset_unlock(&dev->mode_config.connection_mutex);
>> +

It appears that the ABM doesn't actually take effect by writing this 
value until the next modeset.
This is also reported by Dominik Förderer.  I suggested he try to change 
resolutions in his DE and that activated ABM.

So I think it's missing another call to flush out to the hardware.

>> +    return count;
>> +}
>> +
> 
> To make this more discoverable I think you want some kerneldoc.
> 
>> +static DEVICE_ATTR_RW(panel_power_savings);
>> +
>> +static struct attribute *amdgpu_attrs[] = {
>> +    &dev_attr_panel_power_savings.attr,
>> +    NULL
>> +};
>> +
>> +static const struct attribute_group amdgpu_group = {
>> +    .name = "amdgpu",
>> +    .attrs = amdgpu_attrs
>> +};
>> +
>>   static int
>>   amdgpu_dm_connector_late_register(struct drm_connector *connector)
>>   {
>> @@ -6541,6 +6593,13 @@ amdgpu_dm_connector_late_register(struct 
>> drm_connector *connector)
>>           to_amdgpu_dm_connector(connector);
>>       int r;
>> +    if (connector->connector_type == DRM_MODE_CONNECTOR_eDP) {
>> +        r = sysfs_create_group(&connector->kdev->kobj,
>> +                       &amdgpu_group);
>> +        if (r)
>> +            return r;
>> +    }
>> +
> 
> I think there should be some matching code sysfs_remove_group(), maybe 
> in early_unregister() callback.
> 
>>       amdgpu_dm_register_backlight_device(amdgpu_dm_connector);
>>       if ((connector->connector_type == 
>> DRM_MODE_CONNECTOR_DisplayPort) ||
>