qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Philippe Mathieu-Daudé" <philmd@linaro.org>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>,
	QEMU Developers <qemu-devel@nongnu.org>
Subject: Re: Should -overcommit cpu-pm=on|off be converted to some machine sugar property?
Date: Tue, 20 Dec 2022 15:16:17 +0100	[thread overview]
Message-ID: <cc14240c-6fc7-7412-8023-a2d4a047efbc@linaro.org> (raw)
In-Reply-To: <CABgObfaWX=x-RinK0F+XbYi1oVHF53kiahb7FLQjBuoR4eQ_UQ@mail.gmail.com>

On 20/12/22 09:16, Paolo Bonzini wrote:
> It's implemented only for x86, but it is not architecture-dependent. For 
> example on ARM you would implement it by mapping guest WFx to the host 
> instructions.

OK, thanks for clarifying!

>     Hi Paolo,
> 
>     I just noticed 'enable_cpu_pm' which seems architecture specific
>     option that ended into common code:
> 
>     $ git grep -E 'cpu[_-]pm'
>     accel/tcg/user-exec-stub.c:5:bool enable_cpu_pm = false;
>     include/sysemu/sysemu.h:49:extern bool enable_cpu_pm;
>     Binary file pc-bios/skiboot.lid matches
>     qemu-options.hx:4196:    "-overcommit
>     [mem-lock=on|off][cpu-pm=on|off]\n"
>     qemu-options.hx:4199:    "                cpu-pm=on|off controls cpu
>     power management (default: off)\n",
>     qemu-options.hx:4204:``-overcommit cpu-pm=on|off``
>     qemu-options.hx:4214:    guest) can be enabled via ``cpu-pm=on``
>     (disabled by default). This
>     softmmu/globals.c:38:bool enable_cpu_pm;
>     softmmu/vl.c:350:            .name = "cpu-pm",
>     softmmu/vl.c:3440:                enable_cpu_pm =
>     qemu_opt_get_bool(opts, "cpu-pm", false);
>     target/i386/host-cpu.c:44:static void host_cpu_enable_cpu_pm(X86CPU
>     *cpu)
>     target/i386/host-cpu.c:88:    if (cpu->max_features && enable_cpu_pm) {
>     target/i386/host-cpu.c:89:        host_cpu_enable_cpu_pm(cpu);
>     target/i386/kvm/kvm-cpu.c:43:        if (enable_cpu_pm &&
>     kvm_has_waitpkg()) {
>     target/i386/kvm/kvm.c:390:        if (enable_cpu_pm) {
>     target/i386/kvm/kvm.c:2576:    if (enable_cpu_pm) {
> 
>     Do we want to maintain this as a generic property or
>     can it be deprecated / converted to a machine-specific one?
> 
>     Thanks,
> 
>     Phil.
> 



      reply	other threads:[~2022-12-20 14:21 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-12-19 16:11 Should -overcommit cpu-pm=on|off be converted to some machine sugar property? Philippe Mathieu-Daudé
2022-12-19 16:23 ` Michael S. Tsirkin
2022-12-20  8:16 ` Paolo Bonzini
2022-12-20 14:16   ` Philippe Mathieu-Daudé [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=cc14240c-6fc7-7412-8023-a2d4a047efbc@linaro.org \
    --to=philmd@linaro.org \
    --cc=mst@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).