Received-SPF: pass (google.com: domain of linux-kernel+bounces-45231-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Message-ID: <b9af374d-0e6c-447a-8aed-03f68e8f462d@amd.com>
Date: Tue, 30 Jan 2024 13:56:18 -0600
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH Resend 5/8] tools/power x86_energy_perf_policy: add
 nominal and lowest nonlinear perf values showing support
Content-Language: en-US
To: Perry Yuan <perry.yuan@amd.com>, rafael.j.wysocki@intel.com,
 Borislav.Petkov@amd.com, viresh.kumar@linaro.org, Ray.Huang@amd.com,
 gautham.shenoy@amd.com
Cc: Alexander.Deucher@amd.com, Xinmei.Huang@amd.com, Xiaojian.Du@amd.com,
 Li.Meng@amd.com, linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org
References: <cover.1706592301.git.perry.yuan@amd.com>
 <2307c615b12b9d0063ef92b7642fa0708a6b99f9.1706592301.git.perry.yuan@amd.com>
From: Mario Limonciello <mario.limonciello@amd.com>
In-Reply-To: <2307c615b12b9d0063ef92b7642fa0708a6b99f9.1706592301.git.perry.yuan@amd.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk
MIME-Version: 1.0
X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-MessageData-0:
	=?utf-8?B?N2ZGLzFKMlpkRXdTeUc0bGZUckxSRlNEN0g4OFlKQjNmOUFQWk8wRkVOdnVE?=
 =?utf-8?B?LzJRTHplaGcrZ2Ezc3BzOEpobWNzS1c2Tko1QWZTT3RZR3N3Ykc3UTJUeS9t?=
 =?utf-8?B?TnpYM05xMjh3em1ZMThsRmpyUU12NDViTlhGMXUweHVRVUhmTVF0TitHN1ZU?=
 =?utf-8?B?S2l1cklUMENLYjlENlM0dVlrTkU5emh1TE9pSkFkTXVQbnpiQUtXRzJlK2pF?=
 =?utf-8?B?ZVVFWW9EYTBqOHJnL1U2SURISWNjTEtnSzFGYmo2SzJ3Tk1UMkhTZzZJT2xY?=
 =?utf-8?B?dkV4OVA3bTBJblpQcjhha0p4UUhjUDFhOUI0Nk1xdHdUdjFhb2pEK3RSSnpI?=
 =?utf-8?B?WVY3MElXSDZpUWR3bjhzdXEvRkROMDNyMXBPajVad0trUkZwOFRjYWxqV1ov?=
 =?utf-8?B?NUgzeVBCTWpHMm5KdTRSTC9sT0JGSHluckdCRXQ1SjJxRzhsZG52RUxzbmZU?=
 =?utf-8?B?SEhrdVh4YStMUUtvdVVGMXVHNVN2RHg5OWh6OGFpQTVnNnBPOW1GalJab1lr?=
 =?utf-8?B?QlZvQ3dZSnNoaDlMRlovNjl3N3JhRkF4cTlyOXBWOUtVNHdVd3FZWTJEcVRi?=
 =?utf-8?B?aEFhUnl2Vjc0cUR5UXVHdzd2UDZVZFh3MlZ5ZnVjdThpTGQwUG5tYVY2UUEw?=
 =?utf-8?B?QlJQN21xL1dqdUhJK2lmdnI3ZGI5Rld6U1pPd3lnMWlYTVkvQy93bHoyQlRk?=
 =?utf-8?B?Q2xyeGttQ0hvenFZejkvMmo2TGIrUytkYzVYamFRWmFZK1BsL2JCdDd6UmNU?=
 =?utf-8?B?V2J2dGZCZXdxRDZKdWx6WWgvelVhMlNnU0szWkhLWXpqWVhFRmxsRVFiNk9R?=
 =?utf-8?B?YkkyYWRldGlVTHB6MDI5VXlSQ0ZERnVCbHllcXRRMFU2RzIrcnlyUEZtcTZZ?=
 =?utf-8?B?cWJ2M3BnRFlna3V0TEJ0R01CMW9nYWlJdkZGMHZhVWNyTUdaWVNSYWkvL2xo?=
 =?utf-8?B?WFVMYnlEUXp0M0E3V2FNaFRCMDBBZGZocHNHR2xaOVZrWS9pSFl0dXViYU9w?=
 =?utf-8?B?MERQak9CMXVEZ2Uyalh5YmJDWW5OYVlXVDEvUmFrN2ZOSDJKMkFuOC9UZ1dw?=
 =?utf-8?B?QklBTnRIYWIxdXk2c2J6aW9XYXZIeDJoV0Z5clVVZlJSZ2paQS8rc3dYWW8w?=
 =?utf-8?B?aVl0Y2VsV2VjS0tYcDhvMi9aR0psaW90QlVncjg0VC9JU2lsWnF0QTY4VXJ3?=
 =?utf-8?B?UlVtcmlZbkFsSkhONjB0TngwUVpNai9lRE9xcG1oYkFlbEZsekV4N3BreUZF?=
 =?utf-8?B?a3JhWlpadGhlcG45T0RRWnV1VU4zSkJwc0JYY3p5dTFkVFdjVkZwbDhNdks3?=
 =?utf-8?B?M3A4RVR0bEJyaXZhbUlvQ0tmblV0a0kweWRhcmQyN2hWeWs4QmVYZDJWc0Zo?=
 =?utf-8?B?RVFtNXRodEV2cFlwemhCVjBERTFjS01LaGpnOEhWazhjZzA0SVFiZkg4THFE?=
 =?utf-8?B?QzFjcSt4U1BoS0xKZGtLWWlwWUs5cXFMWWNsUURPTDY2V2JQVG1laGJ4Q20z?=
 =?utf-8?B?d3dIem1LUmk5MStvRlVsWDRDc2MreDNRV2JoT1hERWdoUVpoT0pxUVNGcFhC?=
 =?utf-8?B?ajRPRkRwbGZta0pNNDNxdEgzczRPdkZxQ2F3TWcyRzY3clpobitMQ1lJSUlC?=
 =?utf-8?B?NElwTnI5Yk1mNDY3dU9nM09qUnlBYk95VCs1L1ExT3BhWFQvS08xaFB5dytk?=
 =?utf-8?B?SUtLVDhSZk54TWUxUVVqdmdNVjRINlg0ZTBvTUkwUnpISXliOVFESVVOMExH?=
 =?utf-8?B?c0hncERIMGNGQU8wUDVtbDVUa3JaWU53U0JCVjA2UWtmM0xBeHdGZ0JWWHRu?=
 =?utf-8?B?NlFiekJtQmxYbCtBSEJyNVVDeUl5M0o2R3ErZFJGUXAvaEhUamFsWi9FMUR0?=
 =?utf-8?B?enNIWmM2cm5zT0RWMExnUUd2SWM5MjJpelBRdWVHWVU2R3RZS3pTUk9KSVAz?=
 =?utf-8?B?djRNSENKKzVzT29NRDRMNWtySjE2R2c0amhpTi9vVVFvZU1LeW5HMzlaelNO?=
 =?utf-8?B?Q1JBb3U1L0lkT09KK2txZFZqYjFRKytudVdhQ3dwZkFveWFyMEFubkhkTzNo?=
 =?utf-8?B?NzJDOStTeVRlZmxxWDBqOTYwVEExNkZDRFA2bGhwZVdES0lZb0o5RG5TT2Rj?=
 =?utf-8?Q?GL2Ing7m1R0SGP1wLFuSac2vL?=
X-OriginatorOrg: amd.com
X-MS-Exchange-CrossTenant-Network-Message-Id: 318f8f8e-7d7a-4d1a-39f2-08dc21cd86f1
X-MS-Exchange-CrossTenant-AuthSource: MN0PR12MB6101.namprd12.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 30 Jan 2024 19:56:19.7042
 (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: zSQpXY0U3tqfdsLtvnI2dfWLmx/28ZLE2o8krRHKB94Wmh/bBhhVyUBqtJ4hUd2sdoSJYuDY9WUPXXfQCShGbw==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: IA1PR12MB7709

On 1/29/2024 23:56, Perry Yuan wrote:
> Add nominal_perf and lowest_nonlinear_perf read and print the values. It
> will help user to check each core all perf values of CPPC and change the
> EPP profile as need.
> 
> cpu0: [AMD HWP_CAP]: low 12 lowest_non 86 nominal 121 highest 196
> cpu1: [AMD HWP_CAP]: low 12 lowest_non 86 nominal 121 highest 176
> cpu2: [AMD HWP_CAP]: low 12 lowest_non 86 nominal 121 highest 201
> cpu3: [AMD HWP_CAP]: low 12 lowest_non 86 nominal 121 highest 181
> 
> Signed-off-by: Perry Yuan <perry.yuan@amd.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
> ---
>   .../x86_energy_perf_policy/x86_energy_perf_policy.c | 13 +++++++++++--
>   1 file changed, 11 insertions(+), 2 deletions(-)
> 
> diff --git a/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.c b/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.c
> index a08403ab219a..7bc668d22728 100644
> --- a/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.c
> +++ b/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.c
> @@ -38,6 +38,8 @@ struct msr_hwp_cap {
>   	unsigned char guaranteed;
>   	unsigned char efficient;
>   	unsigned char lowest;
> +	unsigned char nominal_perf;
> +	unsigned char lowest_nonlinear_perf;
>   };
>   
>   struct msr_hwp_request {
> @@ -820,8 +822,13 @@ void print_hwp_cap(int cpu, struct msr_hwp_cap *cap, char *str)
>   	if (cpu != -1)
>   		printf("cpu%d: ", cpu);
>   
> -	printf("HWP_CAP: low %d eff %d guar %d high %d\n",
> -		cap->lowest, cap->efficient, cap->guaranteed, cap->highest);
> +	if (genuine_intel) {
> +		printf("HWP_CAP: low %d eff %d guar %d high %d\n",
> +			cap->lowest, cap->efficient, cap->guaranteed, cap->highest);
> +	} else if (authentic_amd) {
> +		printf("[AMD HWP_CAP]: low %d lowest_non %d nominal %d highest %d\n\n",
> +			cap->lowest, cap->lowest_nonlinear_perf, cap->nominal_perf, cap->highest);
> +	}
>   }
>   void read_hwp_cap(int cpu, struct msr_hwp_cap *cap, unsigned int msr_offset)
>   {
> @@ -840,6 +847,8 @@ void read_hwp_cap(int cpu, struct msr_hwp_cap *cap, unsigned int msr_offset)
>   			errx(-1, "failed to get msr with return %d", ret);
>   		cap->highest = msr_perf_2_ratio(AMD_CPPC_HIGHEST_PERF(msr));
>   		cap->lowest = msr_perf_2_ratio(AMD_CPPC_LOWEST_PERF(msr));
> +		cap->nominal_perf = msr_perf_2_ratio(AMD_CPPC_NOMINAL_PERF(msr));
> +		cap->lowest_nonlinear_perf = msr_perf_2_ratio(AMD_CPPC_LOWNONLIN_PERF(msr));
>   	}
>   }
>