[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAJZ5v0gD+QjuW6iiJS7M0H_BAdXH7tnDGa+hp5BnvSNF+Cb0aQ@mail.gmail.com>
Date: Wed, 30 Oct 2024 21:14:34 +0100
From: "Rafael J. Wysocki" <rafael@...nel.org>
To: Dave Hansen <dave.hansen@...el.com>
Cc: "Rafael J. Wysocki" <rafael@...nel.org>, Artem Bityutskiy <artem.bityutskiy@...ux.intel.com>,
Patryk Wlazlyn <patryk.wlazlyn@...ux.intel.com>, x86@...nel.org,
linux-kernel@...r.kernel.org, linux-pm@...r.kernel.org,
rafael.j.wysocki@...el.com, len.brown@...el.com, dave.hansen@...ux.intel.com
Subject: Re: [PATCH v2 2/3] x86/smp: Allow forcing the mwait hint for play
dead loop
On Wed, Oct 30, 2024 at 9:11 PM Dave Hansen <dave.hansen@...el.com> wrote:
>
> On 10/30/24 12:53, Rafael J. Wysocki wrote:
> > clearly referred to as a kexec() hack, which cannot be done in
> > cpuidle_play_dead() because the cpuidle driver doesn't know how to get
> > to md->control.
>
> What if we have an mwait_play_dead() _helper_? It takes the hint as an
> argument and retains all the kexec hacks. All the cpuidle driver has to
> do is call the helper with the hint that the cpuidle driver determines.
The same idea has occurred to me in the meantime, so yes, I think that
it would work.
Powered by blists - more mailing lists