[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <01480f4e-9ca6-50bd-6a95-b2e61834d2e4@amd.com>
Date: Mon, 19 Dec 2022 17:29:34 -0600
From: "Limonciello, Mario" <mario.limonciello@....com>
To: Perry Yuan <perry.yuan@....com>, rafael.j.wysocki@...el.com,
ray.huang@....com, viresh.kumar@...aro.org
Cc: Deepak.Sharma@....com, Nathan.Fontenot@....com,
Alexander.Deucher@....com, Shimmer.Huang@....com,
Xiaojian.Du@....com, Li.Meng@....com, wyes.karny@....com,
linux-pm@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v8 11/13] Documentation: introduce amd pstate active mode
kernel command line options
On 12/19/2022 00:40, Perry Yuan wrote:
> AMD Pstate driver support another firmware based autonomous mode
> with "amd_pstate=active" added to the kernel command line.
> In autonomous mode SMU firmware decides frequencies at 1 ms timescale
This might be the case right now, but I don't know that it will always
be this timescale. Suggest to drop drop "at 1ms timescale".
> based on workload utilization, usage in other IPs, infrastructure
> limits such as power, thermals and so on.
>
> Signed-off-by: Perry Yuan <perry.yuan@....com>
With nit fixed:
Reviewed-by: Mario Limonciello <mario.limonciello@....com>
> ---
> Documentation/admin-guide/kernel-parameters.txt | 7 +++++++
> 1 file changed, 7 insertions(+)
>
> diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
> index 42af9ca0127e..73a02816f6f8 100644
> --- a/Documentation/admin-guide/kernel-parameters.txt
> +++ b/Documentation/admin-guide/kernel-parameters.txt
> @@ -6970,3 +6970,10 @@
> management firmware translates the requests into actual
> hardware states (core frequency, data fabric and memory
> clocks etc.)
> + active
> + Use amd_pstate_epp driver instance as the scaling driver,
> + driver provides a hint to the hardware if software wants
> + to bias toward performance (0x0) or energy efficiency (0xff)
> + to the CPPC firmware. then CPPC power algorithm will
> + calculate the runtime workload and adjust the realtime cores
> + frequency.
Powered by blists - more mailing lists