All of lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Maydell <peter.maydell@linaro.org>
To: Alexander Graf <agraf@csgraf.de>
Cc: "Eduardo Habkost" <ehabkost@redhat.com>,
	"Philippe Mathieu-Daudé" <philmd@redhat.com>,
	"Richard Henderson" <richard.henderson@linaro.org>,
	"QEMU Developers" <qemu-devel@nongnu.org>,
	"Cameron Esfahani" <dirty@apple.com>,
	"Roman Bolshakov" <r.bolshakov@yadro.com>,
	qemu-arm <qemu-arm@nongnu.org>, "Frank Yang" <lfy@google.com>,
	"Paolo Bonzini" <pbonzini@redhat.com>,
	"Peter Collingbourne" <pcc@google.com>
Subject: Re: [PATCH v8 14/19] arm/hvf: Add a WFI handler
Date: Tue, 15 Jun 2021 11:38:07 +0100	[thread overview]
Message-ID: <CAFEAcA9ATvn2yp=_VPitFRnWzZacGy7aodQGNTPMnwsUoFr0xA@mail.gmail.com> (raw)
In-Reply-To: <20210519202253.76782-15-agraf@csgraf.de>

On Wed, 19 May 2021 at 21:23, Alexander Graf <agraf@csgraf.de> wrote:
>
> From: Peter Collingbourne <pcc@google.com>
>
> Sleep on WFI until the VTIMER is due but allow ourselves to be woken
> up on IPI.
>
> In this implementation IPI is blocked on the CPU thread at startup and
> pselect() is used to atomically unblock the signal and begin sleeping.
> The signal is sent unconditionally so there's no need to worry about
> races between actually sleeping and the "we think we're sleeping"
> state. It may lead to an extra wakeup but that's better than missing
> it entirely.
>
> Signed-off-by: Peter Collingbourne <pcc@google.com>
> [agraf: Remove unused 'set' variable, always advance PC on WFX trap]
> Signed-off-by: Alexander Graf <agraf@csgraf.de>
> Acked-by: Roman Bolshakov <r.bolshakov@yadro.com>
>
> ---

> +static void hvf_wfi(CPUState *cpu)
> +{
> +    ARMCPU *arm_cpu = ARM_CPU(cpu);
> +    hv_return_t r;
> +    uint64_t ctl;
> +
> +    if (cpu->interrupt_request & (CPU_INTERRUPT_HARD | CPU_INTERRUPT_FIQ)) {
> +        /* Interrupt pending, no need to wait */
> +        return;
> +    }
> +
> +    r = hv_vcpu_get_sys_reg(cpu->hvf->fd, HV_SYS_REG_CNTV_CTL_EL0,
> +                            &ctl);
> +    assert_hvf_ok(r);
> +
> +    if (!(ctl & 1) || (ctl & 2)) {
> +        /* Timer disabled or masked, just wait for an IPI. */
> +        hvf_wait_for_ipi(cpu, NULL);
> +        return;
> +    }
> +
> +    uint64_t cval;
> +    r = hv_vcpu_get_sys_reg(cpu->hvf->fd, HV_SYS_REG_CNTV_CVAL_EL0,
> +                            &cval);
> +    assert_hvf_ok(r);
> +
> +    int64_t ticks_to_sleep = cval - mach_absolute_time();

This looks odd. The CNTV_CVAL is the compare value against
the CNTVCT (virtual count), which should start at 0 when the
VM starts, pause when the VM is paused, and so on. But here
we are comparing it against what looks like a host absolute
timecount...

> +    if (ticks_to_sleep < 0) {
> +        return;
> +    }
> +
> +    uint64_t seconds = ticks_to_sleep / arm_cpu->gt_cntfrq_hz;
> +    uint64_t nanos =
> +        (ticks_to_sleep - arm_cpu->gt_cntfrq_hz * seconds) *
> +        1000000000 / arm_cpu->gt_cntfrq_hz;

Should this be calling gt_cntfrq_period_ns() ?
(If not, please use the NANOSECONDS_PER_SECOND constant instead of
a raw 1000000000.)

> +
> +    /*
> +     * Don't sleep for less than the time a context switch would take,
> +     * so that we can satisfy fast timer requests on the same CPU.
> +     * Measurements on M1 show the sweet spot to be ~2ms.
> +     */
> +    if (!seconds && nanos < 2000000) {

"2 * SCALE_MS" is a bit easier to read I think.

> +        return;
> +    }
> +
> +    struct timespec ts = { seconds, nanos };
> +    hvf_wait_for_ipi(cpu, &ts);
> +}

thanks
-- PMM


  parent reply	other threads:[~2021-06-15 10:39 UTC|newest]

Thread overview: 66+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-19 20:22 [PATCH v8 00/19] hvf: Implement Apple Silicon Support Alexander Graf
2021-05-19 20:22 ` [PATCH v8 01/19] hvf: Move assert_hvf_ok() into common directory Alexander Graf
2021-05-27 10:00   ` Sergio Lopez
2021-05-19 20:22 ` [PATCH v8 02/19] hvf: Move vcpu thread functions " Alexander Graf
2021-05-27 10:01   ` Sergio Lopez
2021-05-19 20:22 ` [PATCH v8 03/19] hvf: Move cpu " Alexander Graf
2021-05-27 10:02   ` Sergio Lopez
2021-05-19 20:22 ` [PATCH v8 04/19] hvf: Move hvf internal definitions into common header Alexander Graf
2021-05-27 10:04   ` Sergio Lopez
2021-05-19 20:22 ` [PATCH v8 05/19] hvf: Make hvf_set_phys_mem() static Alexander Graf
2021-05-27 10:06   ` Sergio Lopez
2021-05-19 20:22 ` [PATCH v8 06/19] hvf: Remove use of hv_uvaddr_t and hv_gpaddr_t Alexander Graf
2021-05-27 10:07   ` Sergio Lopez
2021-05-19 20:22 ` [PATCH v8 07/19] hvf: Split out common code on vcpu init and destroy Alexander Graf
2021-05-27 10:09   ` Sergio Lopez
2021-05-19 20:22 ` [PATCH v8 08/19] hvf: Use cpu_synchronize_state() Alexander Graf
2021-05-27 10:15   ` Sergio Lopez
2021-05-19 20:22 ` [PATCH v8 09/19] hvf: Make synchronize functions static Alexander Graf
2021-05-27 10:15   ` Sergio Lopez
2021-05-19 20:22 ` [PATCH v8 10/19] hvf: Remove hvf-accel-ops.h Alexander Graf
2021-05-27 10:16   ` Sergio Lopez
2021-05-19 20:22 ` [PATCH v8 11/19] hvf: Introduce hvf vcpu struct Alexander Graf
2021-05-27 10:18   ` Sergio Lopez
2021-05-19 20:22 ` [PATCH v8 12/19] hvf: Simplify post reset/init/loadvm hooks Alexander Graf
2021-05-27 10:20   ` Sergio Lopez
2021-05-19 20:22 ` [PATCH v8 13/19] hvf: Add Apple Silicon support Alexander Graf
2021-05-27 10:55   ` Sergio Lopez
2021-06-15 10:21   ` Peter Maydell
2021-06-27 20:40     ` Alexander Graf
2021-05-19 20:22 ` [PATCH v8 14/19] arm/hvf: Add a WFI handler Alexander Graf
2021-05-27 10:53   ` Sergio Lopez
2021-06-15 10:38   ` Peter Maydell [this message]
2021-05-19 20:22 ` [PATCH v8 15/19] hvf: arm: Implement -cpu host Alexander Graf
2021-05-27 11:18   ` Sergio Lopez
2021-06-15 10:56   ` Peter Maydell
2021-09-12 20:23     ` Alexander Graf
2021-09-13  8:28       ` Peter Maydell
2021-09-13 10:46         ` Alexander Graf
2021-09-13 10:52           ` Peter Maydell
2021-09-13 11:46             ` Alexander Graf
2021-09-13 11:48               ` Peter Maydell
2021-09-13 11:55                 ` Alexander Graf
2021-05-19 20:22 ` [PATCH v8 16/19] hvf: arm: Implement PSCI handling Alexander Graf
2021-05-27 11:20   ` Sergio Lopez
2021-06-15 12:54   ` Peter Maydell
2021-09-12 20:36     ` Alexander Graf
2021-09-12 21:20       ` Richard Henderson
2021-09-12 21:37         ` Alexander Graf
2021-09-12 21:40           ` Richard Henderson
2021-09-13 10:06             ` Alexander Graf
2021-09-13 10:30               ` Philippe Mathieu-Daudé
2021-05-19 20:22 ` [PATCH v8 17/19] arm: Add Hypervisor.framework build target Alexander Graf
2021-05-27 11:20   ` Sergio Lopez
2021-06-15 10:59   ` Peter Maydell
2021-05-19 20:22 ` [PATCH v8 18/19] arm: Enable Windows 10 trusted SMCCC boot call Alexander Graf
2021-05-27 11:21   ` Sergio Lopez
2021-06-15 11:02   ` Peter Maydell
2021-06-27 21:12     ` Alexander Graf
2021-05-19 20:22 ` [PATCH v8 19/19] hvf: arm: Handle Windows 10 SMC call Alexander Graf
2021-05-27 11:22   ` Sergio Lopez
2021-06-15  9:31   ` Peter Maydell
2021-06-27 21:07     ` Alexander Graf
2021-05-19 20:45 ` [PATCH v8 00/19] hvf: Implement Apple Silicon Support no-reply
2021-06-03 13:43 ` Peter Maydell
2021-06-03 13:53   ` Alexander Graf
2021-06-15 12:54   ` Peter Maydell

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='CAFEAcA9ATvn2yp=_VPitFRnWzZacGy7aodQGNTPMnwsUoFr0xA@mail.gmail.com' \
    --to=peter.maydell@linaro.org \
    --cc=agraf@csgraf.de \
    --cc=dirty@apple.com \
    --cc=ehabkost@redhat.com \
    --cc=lfy@google.com \
    --cc=pbonzini@redhat.com \
    --cc=pcc@google.com \
    --cc=philmd@redhat.com \
    --cc=qemu-arm@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=r.bolshakov@yadro.com \
    --cc=richard.henderson@linaro.org \
    --subject='Re: [PATCH v8 14/19] arm/hvf: Add a WFI handler' \
    /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

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.