From: Dave Martin <Dave.Martin@arm.com>
To: Marc Zyngier <marc.zyngier@arm.com>
Cc: Christoffer Dall <cdall@kernel.org>,
Ard Biesheuvel <ard.biesheuvel@linaro.org>,
Catalin Marinas <catalin.marinas@arm.com>,
Will Deacon <will.deacon@arm.com>,
Okamoto Takayuki <tokamoto@jp.fujitsu.com>,
Zhang Lei <zhang.lei@jp.fujitsu.com>,
Julien Grall <julien.grall@arm.com>,
kvmarm@lists.cs.columbia.edu,
linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v7 09/27] KVM: arm64: Add a vcpu flag to control SVE visibility for the guest
Date: Thu, 4 Apr 2019 08:53:19 +0100 [thread overview]
Message-ID: <20190404075317.GW3567@e103592.cambridge.arm.com> (raw)
In-Reply-To: <86k1ga4fy3.wl-marc.zyngier@arm.com>
On Thu, Apr 04, 2019 at 04:17:24AM +0100, Marc Zyngier wrote:
> On Wed, 03 Apr 2019 20:14:13 +0100,
> Andrew Jones <drjones@redhat.com> wrote:
> >
> > On Fri, Mar 29, 2019 at 01:00:34PM +0000, Dave Martin wrote:
> > > Since SVE will be enabled or disabled on a per-vcpu basis, a flag
> > > is needed in order to track which vcpus have it enabled.
> > >
> > > This patch adds a suitable flag and a helper for checking it.
> > >
> > > Signed-off-by: Dave Martin <Dave.Martin@arm.com>
> > > Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
> > > Tested-by: zhang.lei <zhang.lei@jp.fujitsu.com>
> > > ---
> > > arch/arm64/include/asm/kvm_host.h | 4 ++++
> > > 1 file changed, 4 insertions(+)
> > >
> > > diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
> > > index 6d10100..ad4f7f0 100644
> > > --- a/arch/arm64/include/asm/kvm_host.h
> > > +++ b/arch/arm64/include/asm/kvm_host.h
> > > @@ -328,6 +328,10 @@ struct kvm_vcpu_arch {
> > > #define KVM_ARM64_FP_HOST (1 << 2) /* host FP regs loaded */
> > > #define KVM_ARM64_HOST_SVE_IN_USE (1 << 3) /* backup for host TIF_SVE */
> > > #define KVM_ARM64_HOST_SVE_ENABLED (1 << 4) /* SVE enabled for EL0 */
> > > +#define KVM_ARM64_GUEST_HAS_SVE (1 << 5) /* SVE exposed to guest */
> > > +
> > > +#define vcpu_has_sve(vcpu) (system_supports_sve() && \
> > > + ((vcpu)->arch.flags & KVM_ARM64_GUEST_HAS_SVE))
> >
> > We shouldn't need the system_supports_sve() here. vcpu->arch.flags can
> > only have the KVM_ARM64_GUEST_HAS_SVE flag set when system_supports_sve()
> > is true, and it can't change later.
>
> This is a performance optimisation. system_supports_sve() results in a
> static key being emitted, and that avoids a number of loads on the
> fast path for non-SVE systems (which is likely to be the absolute
> majority for the foreseeable future).
Exactly so. The same applies to many of the other call sites for
system_supports_sve().
Cheers
---Dave
WARNING: multiple messages have this Message-ID (diff)
From: Dave Martin <Dave.Martin@arm.com>
To: Marc Zyngier <marc.zyngier@arm.com>
Cc: Andrew Jones <drjones@redhat.com>,
Christoffer Dall <cdall@kernel.org>,
Ard Biesheuvel <ard.biesheuvel@linaro.org>,
Catalin Marinas <catalin.marinas@arm.com>,
Will Deacon <will.deacon@arm.com>,
Okamoto Takayuki <tokamoto@jp.fujitsu.com>,
Zhang Lei <zhang.lei@jp.fujitsu.com>,
Julien Grall <julien.grall@arm.com>,
kvmarm@lists.cs.columbia.edu,
linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v7 09/27] KVM: arm64: Add a vcpu flag to control SVE visibility for the guest
Date: Thu, 4 Apr 2019 08:53:19 +0100 [thread overview]
Message-ID: <20190404075317.GW3567@e103592.cambridge.arm.com> (raw)
In-Reply-To: <86k1ga4fy3.wl-marc.zyngier@arm.com>
On Thu, Apr 04, 2019 at 04:17:24AM +0100, Marc Zyngier wrote:
> On Wed, 03 Apr 2019 20:14:13 +0100,
> Andrew Jones <drjones@redhat.com> wrote:
> >
> > On Fri, Mar 29, 2019 at 01:00:34PM +0000, Dave Martin wrote:
> > > Since SVE will be enabled or disabled on a per-vcpu basis, a flag
> > > is needed in order to track which vcpus have it enabled.
> > >
> > > This patch adds a suitable flag and a helper for checking it.
> > >
> > > Signed-off-by: Dave Martin <Dave.Martin@arm.com>
> > > Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
> > > Tested-by: zhang.lei <zhang.lei@jp.fujitsu.com>
> > > ---
> > > arch/arm64/include/asm/kvm_host.h | 4 ++++
> > > 1 file changed, 4 insertions(+)
> > >
> > > diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
> > > index 6d10100..ad4f7f0 100644
> > > --- a/arch/arm64/include/asm/kvm_host.h
> > > +++ b/arch/arm64/include/asm/kvm_host.h
> > > @@ -328,6 +328,10 @@ struct kvm_vcpu_arch {
> > > #define KVM_ARM64_FP_HOST (1 << 2) /* host FP regs loaded */
> > > #define KVM_ARM64_HOST_SVE_IN_USE (1 << 3) /* backup for host TIF_SVE */
> > > #define KVM_ARM64_HOST_SVE_ENABLED (1 << 4) /* SVE enabled for EL0 */
> > > +#define KVM_ARM64_GUEST_HAS_SVE (1 << 5) /* SVE exposed to guest */
> > > +
> > > +#define vcpu_has_sve(vcpu) (system_supports_sve() && \
> > > + ((vcpu)->arch.flags & KVM_ARM64_GUEST_HAS_SVE))
> >
> > We shouldn't need the system_supports_sve() here. vcpu->arch.flags can
> > only have the KVM_ARM64_GUEST_HAS_SVE flag set when system_supports_sve()
> > is true, and it can't change later.
>
> This is a performance optimisation. system_supports_sve() results in a
> static key being emitted, and that avoids a number of loads on the
> fast path for non-SVE systems (which is likely to be the absolute
> majority for the foreseeable future).
Exactly so. The same applies to many of the other call sites for
system_supports_sve().
Cheers
---Dave
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
next prev parent reply other threads:[~2019-04-04 7:53 UTC|newest]
Thread overview: 224+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-03-29 13:00 [PATCH v7 00/27] KVM: arm64: SVE guest support Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 01/27] KVM: Documentation: Document arm64 core registers in detail Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-24 9:25 ` Alex Bennée
2019-04-24 9:25 ` Alex Bennée
2019-03-29 13:00 ` [PATCH v7 02/27] arm64: fpsimd: Always set TIF_FOREIGN_FPSTATE on task state flush Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 03/27] KVM: arm64: Delete orphaned declaration for __fpsimd_enabled() Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 04/27] KVM: arm64: Refactor kvm_arm_num_regs() for easier maintenance Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 05/27] KVM: arm64: Add missing #includes to kvm_host.h Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 06/27] arm64/sve: Clarify role of the VQ map maintenance functions Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-04 21:21 ` Andrew Jones
2019-04-04 21:21 ` Andrew Jones
2019-03-29 13:00 ` [PATCH v7 07/27] arm64/sve: Check SVE virtualisability Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-04 21:21 ` Andrew Jones
2019-04-04 21:21 ` Andrew Jones
2019-04-05 9:35 ` Dave Martin
2019-04-05 9:35 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 08/27] arm64/sve: Enable SVE state tracking for non-task contexts Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 09/27] KVM: arm64: Add a vcpu flag to control SVE visibility for the guest Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-03 19:14 ` Andrew Jones
2019-04-03 19:14 ` Andrew Jones
2019-04-04 3:17 ` Marc Zyngier
2019-04-04 3:17 ` Marc Zyngier
2019-04-04 7:53 ` Dave Martin [this message]
2019-04-04 7:53 ` Dave Martin
2019-04-04 21:15 ` Andrew Jones
2019-04-04 21:15 ` Andrew Jones
2019-03-29 13:00 ` [PATCH v7 10/27] KVM: arm64: Propagate vcpu into read_id_reg() Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-04 21:15 ` Andrew Jones
2019-04-04 21:15 ` Andrew Jones
2019-03-29 13:00 ` [PATCH v7 11/27] KVM: arm64: Support runtime sysreg visibility filtering Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-03 19:17 ` Andrew Jones
2019-04-03 19:17 ` Andrew Jones
2019-04-24 9:39 ` Alex Bennée
2019-04-24 9:39 ` Alex Bennée
2019-04-24 13:47 ` Dave Martin
2019-04-24 13:47 ` Dave Martin
2019-04-24 13:47 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 12/27] KVM: arm64/sve: System register context switch and access support Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-03 19:39 ` Andrew Jones
2019-04-03 19:39 ` Andrew Jones
2019-04-04 8:06 ` Dave Martin
2019-04-04 8:06 ` Dave Martin
2019-04-04 8:32 ` Andrew Jones
2019-04-04 8:32 ` Andrew Jones
2019-04-04 8:47 ` Dave Martin
2019-04-04 8:47 ` Dave Martin
2019-04-04 8:59 ` Andrew Jones
2019-04-04 8:59 ` Andrew Jones
2019-04-24 15:21 ` Alex Bennée
2019-04-24 15:21 ` Alex Bennée
2019-04-25 13:28 ` Dave Martin
2019-04-25 13:28 ` Dave Martin
2019-04-25 13:28 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 13/27] KVM: arm64/sve: Context switch the SVE registers Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-03 20:01 ` Andrew Jones
2019-04-03 20:01 ` Andrew Jones
2019-04-04 8:10 ` Dave Martin
2019-04-04 8:10 ` Dave Martin
2019-04-04 8:35 ` Andrew Jones
2019-04-04 8:35 ` Andrew Jones
2019-04-04 8:36 ` Dave Martin
2019-04-04 8:36 ` Dave Martin
2019-04-24 14:51 ` Alex Bennée
2019-04-24 14:51 ` Alex Bennée
2019-04-25 13:35 ` Dave Martin
2019-04-25 13:35 ` Dave Martin
2019-04-25 13:35 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 14/27] KVM: Allow 2048-bit register access via ioctl interface Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-04 21:11 ` Andrew Jones
2019-04-04 21:11 ` Andrew Jones
2019-03-29 13:00 ` [PATCH v7 15/27] KVM: arm64: Add missing #include of <linux/string.h> in guest.c Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 16/27] KVM: arm64: Factor out core register ID enumeration Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-02 2:41 ` Marc Zyngier
2019-04-02 2:41 ` Marc Zyngier
2019-04-02 8:59 ` Dave Martin
2019-04-02 8:59 ` Dave Martin
2019-04-02 9:32 ` Marc Zyngier
2019-04-02 9:32 ` Marc Zyngier
2019-04-02 9:54 ` Dave P Martin
2019-04-02 9:54 ` Dave P Martin
2019-03-29 13:00 ` [PATCH v7 17/27] KVM: arm64: Reject ioctl access to FPSIMD V-regs on SVE vcpus Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-03 20:15 ` Andrew Jones
2019-04-03 20:15 ` Andrew Jones
2019-04-24 13:45 ` Alex Bennée
2019-04-24 13:45 ` Alex Bennée
2019-03-29 13:00 ` [PATCH v7 18/27] KVM: arm64/sve: Add SVE support to register access ioctl interface Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-04 13:57 ` Andrew Jones
2019-04-04 13:57 ` Andrew Jones
2019-04-04 14:50 ` Dave Martin
2019-04-04 14:50 ` Dave Martin
2019-04-04 16:25 ` Andrew Jones
2019-04-04 16:25 ` Andrew Jones
2019-04-04 16:56 ` Dave Martin
2019-04-04 16:56 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 19/27] KVM: arm64: Enumerate SVE register indices for KVM_GET_REG_LIST Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-04 14:08 ` Andrew Jones
2019-04-04 14:08 ` Andrew Jones
2019-04-05 9:35 ` Dave Martin
2019-04-05 9:35 ` Dave Martin
2019-04-05 9:45 ` Andrew Jones
2019-04-05 9:45 ` Andrew Jones
2019-04-05 11:11 ` Dave Martin
2019-04-05 11:11 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 20/27] arm64/sve: In-kernel vector length availability query interface Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-04 14:20 ` Andrew Jones
2019-04-04 14:20 ` Andrew Jones
2019-04-05 9:35 ` Dave Martin
2019-04-05 9:35 ` Dave Martin
2019-04-05 9:54 ` Andrew Jones
2019-04-05 9:54 ` Andrew Jones
2019-04-05 11:13 ` Dave Martin
2019-04-05 11:13 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 21/27] KVM: arm/arm64: Add hook for arch-specific KVM initialisation Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-04 14:25 ` Andrew Jones
2019-04-04 14:25 ` Andrew Jones
2019-04-04 14:53 ` Dave Martin
2019-04-04 14:53 ` Dave Martin
2019-04-04 16:33 ` Andrew Jones
2019-04-04 16:33 ` Andrew Jones
2019-04-05 9:36 ` Dave Martin
2019-04-05 9:36 ` Dave Martin
2019-04-05 10:40 ` Andrew Jones
2019-04-05 10:40 ` Andrew Jones
2019-04-05 11:14 ` Dave Martin
2019-04-05 11:14 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 22/27] KVM: arm/arm64: Add KVM_ARM_VCPU_FINALIZE ioctl Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-04 15:07 ` Andrew Jones
2019-04-04 15:07 ` Andrew Jones
2019-04-04 16:47 ` Dave Martin
2019-04-04 16:47 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 23/27] KVM: arm64/sve: Add pseudo-register for the guest's vector lengths Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-04 20:18 ` Andrew Jones
2019-04-04 20:18 ` Andrew Jones
2019-04-05 9:36 ` Dave Martin
2019-04-05 9:36 ` Dave Martin
2019-04-05 10:14 ` Andrew Jones
2019-04-05 10:14 ` Andrew Jones
2019-04-05 12:54 ` Dave Martin
2019-04-05 12:54 ` Dave Martin
2019-04-05 15:33 ` Andrew Jones
2019-04-05 15:33 ` Andrew Jones
2019-04-10 12:42 ` Dave Martin
2019-04-10 12:42 ` Dave Martin
2019-04-10 12:42 ` Dave Martin
2019-04-04 20:31 ` Andrew Jones
2019-04-04 20:31 ` Andrew Jones
2019-04-05 9:36 ` Dave Martin
2019-04-05 9:36 ` Dave Martin
2019-04-05 10:22 ` Andrew Jones
2019-04-05 10:22 ` Andrew Jones
2019-04-05 14:06 ` Dave Martin
2019-04-05 14:06 ` Dave Martin
2019-04-05 15:41 ` Andrew Jones
2019-04-05 15:41 ` Andrew Jones
2019-04-10 12:35 ` Dave Martin
2019-04-10 12:35 ` Dave Martin
2019-04-10 12:35 ` Dave Martin
2019-03-29 13:00 ` [PATCH v7 24/27] KVM: arm64/sve: Allow userspace to enable SVE for vcpus Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-04 20:36 ` Andrew Jones
2019-04-04 20:36 ` Andrew Jones
2019-03-29 13:00 ` [PATCH v7 25/27] KVM: arm64: Add a capability to advertise SVE support Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-04 20:39 ` Andrew Jones
2019-04-04 20:39 ` Andrew Jones
2019-03-29 13:00 ` [PATCH v7 26/27] KVM: Document errors for KVM_GET_ONE_REG and KVM_SET_ONE_REG Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-03 20:27 ` Andrew Jones
2019-04-03 20:27 ` Andrew Jones
2019-04-04 8:35 ` Dave Martin
2019-04-04 8:35 ` Dave Martin
2019-04-04 9:34 ` Andrew Jones
2019-04-04 9:34 ` Andrew Jones
2019-04-04 9:38 ` Dave P Martin
2019-04-04 9:38 ` Dave P Martin
2019-04-04 9:45 ` Andrew Jones
2019-04-04 9:45 ` Andrew Jones
2019-03-29 13:00 ` [PATCH v7 27/27] KVM: arm64/sve: Document KVM API extensions for SVE Dave Martin
2019-03-29 13:00 ` Dave Martin
2019-04-04 21:09 ` Andrew Jones
2019-04-04 21:09 ` Andrew Jones
2019-04-05 9:36 ` Dave Martin
2019-04-05 9:36 ` Dave Martin
2019-04-05 10:39 ` Andrew Jones
2019-04-05 10:39 ` Andrew Jones
2019-04-05 13:00 ` Dave Martin
2019-04-05 13:00 ` Dave Martin
2019-04-05 15:38 ` Andrew Jones
2019-04-05 15:38 ` Andrew Jones
2019-04-10 12:34 ` Dave Martin
2019-04-10 12:34 ` Dave Martin
2019-04-10 12:34 ` Dave Martin
2019-03-29 14:56 ` [PATCH v7 00/27] KVM: arm64: SVE guest support Marc Zyngier
2019-03-29 14:56 ` Marc Zyngier
2019-03-29 15:06 ` Dave Martin
2019-03-29 15:06 ` Dave Martin
2019-04-05 16:41 ` Dave Martin
2019-04-05 16:41 ` Dave Martin
2019-04-25 10:33 ` Alex Bennée
2019-04-25 10:33 ` Alex Bennée
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=20190404075317.GW3567@e103592.cambridge.arm.com \
--to=dave.martin@arm.com \
--cc=ard.biesheuvel@linaro.org \
--cc=catalin.marinas@arm.com \
--cc=cdall@kernel.org \
--cc=julien.grall@arm.com \
--cc=kvmarm@lists.cs.columbia.edu \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=marc.zyngier@arm.com \
--cc=tokamoto@jp.fujitsu.com \
--cc=will.deacon@arm.com \
--cc=zhang.lei@jp.fujitsu.com \
/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 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.