From: Mohamed Mediouni <mohamed@unpredictable.fr>
To: qemu-devel@nongnu.org
Cc: "Marcel Apfelbaum" <marcel.apfelbaum@gmail.com>,
"Yanan Wang" <wangyanan55@huawei.com>,
"Zhao Liu" <zhao1.liu@intel.com>,
qemu-arm@nongnu.org, "Peter Maydell" <peter.maydell@linaro.org>,
"Roman Bolshakov" <rbolshakov@ddn.com>,
"Alexander Graf" <agraf@csgraf.de>,
"Philippe Mathieu-Daudé" <philmd@linaro.org>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Eduardo Habkost" <eduardo@habkost.net>,
"Phil Dennis-Jordan" <phil@philjordan.eu>,
"Mohamed Mediouni" <mohamed@unpredictable.fr>
Subject: [PATCH v20 08/15] hvf: arm: allow exposing minimal PMU when running with nested virt on
Date: Mon, 16 Mar 2026 14:06:35 +0100 [thread overview]
Message-ID: <20260316130642.13246-9-mohamed@unpredictable.fr> (raw)
In-Reply-To: <20260316130642.13246-1-mohamed@unpredictable.fr>
When running with the Apple vGIC, a minimum PMU is exposed by Hypervisor.framework
if a valid PMUVer register value is set. That PMU isn't exposed otherwise.
Signed-off-by: Mohamed Mediouni <mohamed@unpredictable.fr>
---
target/arm/hvf/hvf.c | 19 +++++++++++++++++++
1 file changed, 19 insertions(+)
diff --git a/target/arm/hvf/hvf.c b/target/arm/hvf/hvf.c
index c971aa8cce..c457e67e63 100644
--- a/target/arm/hvf/hvf.c
+++ b/target/arm/hvf/hvf.c
@@ -1145,6 +1145,25 @@ static bool hvf_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf)
clamp_id_aa64mmfr0_parange_to_ipa_size(&host_isar);
+ /*
+ * Windows wants at least the PMU's cycles counter to be available.
+ *
+ * With kernel-irqchip=off, we "emulate" the cycles counter
+ * in reference to time in QEMU. Having that, even with
+ * ID_AA64DFR0_EL1.PMUVer = 0 is enough to make Windows happy.
+ *
+ * As it's a very inaccurate implementation with its only purpose
+ * being making Windows boot, expose ID_AA64DFR0_EL1.PMUVer = 0
+ * when kernel-irqchip=off.
+ *
+ * When kernel-irqchip=on *and* ID_AA64DFR0_EL1.PMUVer = 1,
+ * the OS provides its own PMU emulation, which is currently
+ * a cycles counter only emulation.
+ */
+ if (hvf_irqchip_in_kernel()) {
+ FIELD_DP64_IDREG(&host_isar, ID_AA64DFR0, PMUVER, 0x1);
+ }
+
ahcf->isar = host_isar;
/*
--
2.50.1 (Apple Git-155)
next prev parent reply other threads:[~2026-03-16 13:09 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-16 13:06 [PATCH v20 00/15] HVF: Add support for platform vGIC and nested virtualisation Mohamed Mediouni
2026-03-16 13:06 ` [PATCH v20 01/15] hw/intc: Add hvf vGIC interrupt controller support Mohamed Mediouni
2026-04-24 6:38 ` Manos Pitsidianakis
2026-03-16 13:06 ` [PATCH v20 02/15] hw/intc: arm_gicv3_hvf: save/restore Apple GIC state Mohamed Mediouni
2026-04-24 6:56 ` Manos Pitsidianakis
2026-04-24 7:29 ` Philippe Mathieu-Daudé
2026-03-16 13:06 ` [PATCH v20 03/15] accel, hw/arm, include/system/hvf: infrastructure changes for HVF vGIC Mohamed Mediouni
2026-04-23 16:10 ` Philippe Mathieu-Daudé
2026-04-23 17:01 ` Mohamed Mediouni
2026-03-16 13:06 ` [PATCH v20 04/15] target/arm: hvf: instantiate GIC early Mohamed Mediouni
2026-03-16 13:06 ` [PATCH v20 05/15] hw/arm, target/arm: nested virtualisation on HVF Mohamed Mediouni
2026-04-24 7:07 ` Manos Pitsidianakis
2026-03-16 13:06 ` [PATCH v20 06/15] hvf: only call hvf_sync_vtimer() when running without the platform vGIC Mohamed Mediouni
2026-03-16 13:06 ` [PATCH v20 07/15] hvf: gate ARM_FEATURE_PMU register emulation when using the Apple vGIC Mohamed Mediouni
2026-04-24 7:15 ` Manos Pitsidianakis
2026-03-16 13:06 ` Mohamed Mediouni [this message]
2026-04-23 16:03 ` [PATCH v20 08/15] hvf: arm: allow exposing minimal PMU when running with nested virt on Philippe Mathieu-Daudé
2026-03-16 13:06 ` [PATCH v20 09/15] target/arm: hvf: add asserts for code paths not leveraged when using the vGIC Mohamed Mediouni
2026-03-16 13:06 ` [PATCH v20 10/15] hvf: sync registers used at EL2 Mohamed Mediouni
2026-03-16 13:06 ` [PATCH v20 11/15] target/arm: hvf: pass through CNTHCTL_EL2 and MDCCINT_EL1 Mohamed Mediouni
2026-03-16 13:06 ` [PATCH v20 12/15] hvf: arm: disable SME when nested virt is active Mohamed Mediouni
2026-03-16 13:06 ` [PATCH v20 13/15] hvf: arm: physical timer emulation Mohamed Mediouni
2026-04-23 16:07 ` Philippe Mathieu-Daudé
2026-03-16 13:06 ` [PATCH v20 14/15] hvf: enable nested virtualisation support Mohamed Mediouni
2026-04-24 7:11 ` Manos Pitsidianakis
2026-03-16 13:06 ` [PATCH v20 15/15] hvf: arm: enable vGIC by default for virt-11.1 and later Mohamed Mediouni
2026-04-24 7:13 ` Manos Pitsidianakis
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=20260316130642.13246-9-mohamed@unpredictable.fr \
--to=mohamed@unpredictable.fr \
--cc=agraf@csgraf.de \
--cc=eduardo@habkost.net \
--cc=marcel.apfelbaum@gmail.com \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=phil@philjordan.eu \
--cc=philmd@linaro.org \
--cc=qemu-arm@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=rbolshakov@ddn.com \
--cc=wangyanan55@huawei.com \
--cc=zhao1.liu@intel.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.