From: Peter Maydell <peter.maydell@linaro.org>
To: Akihiko Odaki <akihiko.odaki@daynix.com>
Cc: Thomas Huth <thuth@redhat.com>,
Laurent Vivier <lvivier@redhat.com>,
Paolo Bonzini <pbonzini@redhat.com>,
qemu-arm@nongnu.org, qemu-devel@nongnu.org, kvm@vger.kernel.org
Subject: Re: [PATCH v3 2/5] target/arm/kvm: Fix PMU feature bit early
Date: Thu, 18 Jul 2024 13:07:35 +0100 [thread overview]
Message-ID: <CAFEAcA8tFtdpCQobU9ytzxvf3_y3DiA1TwNq8fWgFUtCUYT4hQ@mail.gmail.com> (raw)
In-Reply-To: <20240716-pmu-v3-2-8c7c1858a227@daynix.com>
On Tue, 16 Jul 2024 at 13:50, Akihiko Odaki <akihiko.odaki@daynix.com> wrote:
>
> kvm_arm_get_host_cpu_features() used to add the PMU feature
> unconditionally, and kvm_arch_init_vcpu() removed it when it is actually
> not available. Conditionally add the PMU feature in
> kvm_arm_get_host_cpu_features() to save code.
>
> Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
> ---
> target/arm/kvm.c | 7 +------
> 1 file changed, 1 insertion(+), 6 deletions(-)
>
> diff --git a/target/arm/kvm.c b/target/arm/kvm.c
> index 70f79eda33cd..849e2e21b304 100644
> --- a/target/arm/kvm.c
> +++ b/target/arm/kvm.c
> @@ -280,6 +280,7 @@ static bool kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf)
> if (kvm_arm_pmu_supported()) {
> init.features[0] |= 1 << KVM_ARM_VCPU_PMU_V3;
> pmu_supported = true;
> + features |= 1ULL << ARM_FEATURE_PMU;
> }
>
> if (!kvm_arm_create_scratch_host_vcpu(cpus_to_try, fdarray, &init)) {
> @@ -448,7 +449,6 @@ static bool kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf)
> features |= 1ULL << ARM_FEATURE_V8;
> features |= 1ULL << ARM_FEATURE_NEON;
> features |= 1ULL << ARM_FEATURE_AARCH64;
> - features |= 1ULL << ARM_FEATURE_PMU;
> features |= 1ULL << ARM_FEATURE_GENERIC_TIMER;
>
> ahcf->features = features;
> @@ -1888,13 +1888,8 @@ int kvm_arch_init_vcpu(CPUState *cs)
> if (!arm_feature(env, ARM_FEATURE_AARCH64)) {
> cpu->kvm_init_features[0] |= 1 << KVM_ARM_VCPU_EL1_32BIT;
> }
> - if (!kvm_check_extension(cs->kvm_state, KVM_CAP_ARM_PMU_V3)) {
> - cpu->has_pmu = false;
> - }
> if (cpu->has_pmu) {
> cpu->kvm_init_features[0] |= 1 << KVM_ARM_VCPU_PMU_V3;
> - } else {
> - env->features &= ~(1ULL << ARM_FEATURE_PMU);
> }
> if (cpu_isar_feature(aa64_sve, cpu)) {
> assert(kvm_arm_sve_supported());
Not every KVM CPU is necessarily the "host" CPU type.
The "cortex-a57" and "cortex-a53" CPU types will work if you
happen to be on a host of that CPU type, and they don't go
through kvm_arm_get_host_cpu_features().
(Also, at some point in the future we're probably going to
want to support "tell the guest it has CPU type X via the
ID registers even when the host is CPU type Y". It seems
plausible that in that case also we'll end up wanting this
there too. But I don't put much weight on this because there's
probably a bunch of things we'll need to fix up if and when
we eventually try to implement this.)
thanks
-- PMM
next prev parent reply other threads:[~2024-07-18 12:08 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-16 12:50 [PATCH v3 0/5] target/arm/kvm: Report PMU unavailability Akihiko Odaki
2024-07-16 12:50 ` [PATCH v3 1/5] tests/arm-cpu-features: Do not assume PMU availability Akihiko Odaki
2024-07-16 12:50 ` [PATCH v3 2/5] target/arm/kvm: Fix PMU feature bit early Akihiko Odaki
2024-07-18 12:07 ` Peter Maydell [this message]
2024-07-19 7:21 ` Akihiko Odaki
2024-07-19 12:21 ` Cornelia Huck
2024-07-19 16:29 ` Akihiko Odaki
2024-07-16 12:50 ` [PATCH v3 3/5] target/arm: Always add pmu property for Armv8 Akihiko Odaki
2024-07-18 12:08 ` Peter Maydell
2024-07-16 12:50 ` [PATCH v3 4/5] hvf: arm: Do not advance PC when raising an exception Akihiko Odaki
2024-07-16 12:50 ` [PATCH v3 5/5] hvf: arm: Properly disable PMU Akihiko Odaki
2024-07-18 12:13 ` Peter Maydell
2024-07-18 12:14 ` [PATCH v3 0/5] target/arm/kvm: Report PMU unavailability Peter Maydell
2024-07-18 12:47 ` 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=CAFEAcA8tFtdpCQobU9ytzxvf3_y3DiA1TwNq8fWgFUtCUYT4hQ@mail.gmail.com \
--to=peter.maydell@linaro.org \
--cc=akihiko.odaki@daynix.com \
--cc=kvm@vger.kernel.org \
--cc=lvivier@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-arm@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=thuth@redhat.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 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).