From: Mohamed Mediouni <mohamed@unpredictable.fr>
To: qemu-devel@nongnu.org
Cc: "Alexander Graf" <agraf@csgraf.de>,
"Richard Henderson" <richard.henderson@linaro.org>,
"Cameron Esfahani" <dirty@apple.com>,
"Mads Ynddal" <mads@ynddal.dk>,
qemu-arm@nongnu.org, "Philippe Mathieu-Daudé" <philmd@linaro.org>,
"Ani Sinha" <anisinha@redhat.com>,
"Phil Dennis-Jordan" <phil@philjordan.eu>,
"Eduardo Habkost" <eduardo@habkost.net>,
"Sunil Muthuswamy" <sunilmut@microsoft.com>,
"Michael S. Tsirkin" <mst@redhat.com>,
"Igor Mammedov" <imammedo@redhat.com>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Yanan Wang" <wangyanan55@huawei.com>,
"\"Daniel P. Berrangé\"" <berrange@redhat.com>,
"Shannon Zhao" <shannon.zhaosl@gmail.com>,
kvm@vger.kernel.org, "Peter Maydell" <peter.maydell@linaro.org>,
"Marcel Apfelbaum" <marcel.apfelbaum@gmail.com>,
"Marc-André Lureau" <marcandre.lureau@redhat.com>,
"Pedro Barbuda" <pbarbuda@microsoft.com>,
"Zhao Liu" <zhao1.liu@intel.com>,
"Roman Bolshakov" <rbolshakov@ddn.com>
Subject: Re: [PATCH v7 15/24] hw/arm: virt: cleanly fail on attempt to use the platform vGIC together with ITS
Date: Thu, 16 Oct 2025 20:19:48 +0200 [thread overview]
Message-ID: <25E6C302-43A7-4D52-BFBA-0712085DF63F@unpredictable.fr> (raw)
In-Reply-To: <20251016165520.62532-16-mohamed@unpredictable.fr>
Note for this specific one (already posted on IRC, but worth keeping in mind for a merge), mail server is misbehaving… (so v8 has been posted partially. And this is the change there)
It has a bug, with the changed commit being at https://github.com/mediouni-m/qemu/commit/863eb69916be4a43ddde62276bb643caf46ef236
The delta between the two:
diff --git a/hw/arm/virt-acpi-build.c b/hw/arm/virt-acpi-build.c
index c716130206..8e730731ca 100644
--- a/hw/arm/virt-acpi-build.c
+++ b/hw/arm/virt-acpi-build.c
@@ -961,7 +961,8 @@ build_madt(GArray *table_data, BIOSLinker *linker, VirtMachineState *vms)
}
}
- if (virt_is_its_enabled(vms) && !vms->no_gicv3_with_gicv2m) {
+ if (!(vms->gic_version != VIRT_GIC_VERSION_2 && virt_is_its_enabled(vms))
+ && !vms->no_gicv3_with_gicv2m) {
const uint16_t spi_base = vms->irqmap[VIRT_GIC_V2M] + ARM_SPI_BASE;
/* 5.2.12.16 GIC MSI Frame Structure */
diff --git a/hw/arm/virt.c b/hw/arm/virt.c
index 1904765db3..b5bed029b9 100644
--- a/hw/arm/virt.c
+++ b/hw/arm/virt.c
@@ -2717,10 +2717,6 @@ static void virt_set_highmem_mmio_size(Object *obj, Visitor *v,
bool virt_is_its_enabled(VirtMachineState *vms)
{
- if (vms->gic_version == VIRT_GIC_VERSION_2) {
- return false;
- }
-
if (vms->its == ON_OFF_AUTO_OFF) {
return false;
}
> On 16. Oct 2025, at 18:55, Mohamed Mediouni <mohamed@unpredictable.fr> wrote:
>
> Switch its to a tristate.
>
> Windows Hypervisor Platform's vGIC doesn't support ITS.
> Deal with this by reporting to the user and exiting.
>
> Regular configuration: GICv3 + ITS
> New default configuration with WHPX: GICv3 with GICv2m
> And its=off explicitly for the newest machine version: GICv3 + GICv2m
>
> Signed-off-by: Mohamed Mediouni <mohamed@unpredictable.fr>
> ---
> hw/arm/virt-acpi-build.c | 14 +++++------
> hw/arm/virt.c | 50 ++++++++++++++++++++++++++++++++--------
> include/hw/arm/virt.h | 4 +++-
> 3 files changed, 50 insertions(+), 18 deletions(-)
>
> diff --git a/hw/arm/virt-acpi-build.c b/hw/arm/virt-acpi-build.c
> index 0a6ec74aa0..c716130206 100644
> --- a/hw/arm/virt-acpi-build.c
> +++ b/hw/arm/virt-acpi-build.c
> @@ -472,7 +472,7 @@ build_iort(GArray *table_data, BIOSLinker *linker, VirtMachineState *vms)
> nb_nodes = num_smmus + 1; /* RC and SMMUv3 */
> rc_mapping_count = rc_smmu_idmaps_len;
>
> - if (vms->its) {
> + if (virt_is_its_enabled(vms)) {
> /*
> * Knowing the ID ranges from the RC to the SMMU, it's possible to
> * determine the ID ranges from RC that go directly to ITS.
> @@ -483,7 +483,7 @@ build_iort(GArray *table_data, BIOSLinker *linker, VirtMachineState *vms)
> rc_mapping_count += rc_its_idmaps->len;
> }
> } else {
> - if (vms->its) {
> + if (virt_is_its_enabled(vms)) {
> nb_nodes = 2; /* RC and ITS */
> rc_mapping_count = 1; /* Direct map to ITS */
> } else {
> @@ -498,7 +498,7 @@ build_iort(GArray *table_data, BIOSLinker *linker, VirtMachineState *vms)
> build_append_int_noprefix(table_data, IORT_NODE_OFFSET, 4);
> build_append_int_noprefix(table_data, 0, 4); /* Reserved */
>
> - if (vms->its) {
> + if (virt_is_its_enabled(vms)) {
> /* Table 12 ITS Group Format */
> build_append_int_noprefix(table_data, 0 /* ITS Group */, 1); /* Type */
> node_size = 20 /* fixed header size */ + 4 /* 1 GIC ITS Identifier */;
> @@ -517,7 +517,7 @@ build_iort(GArray *table_data, BIOSLinker *linker, VirtMachineState *vms)
> int smmu_mapping_count, offset_to_id_array;
> int irq = sdev->irq;
>
> - if (vms->its) {
> + if (virt_is_its_enabled(vms)) {
> smmu_mapping_count = 1; /* ITS Group node */
> offset_to_id_array = SMMU_V3_ENTRY_SIZE; /* Just after the header */
> } else {
> @@ -610,7 +610,7 @@ build_iort(GArray *table_data, BIOSLinker *linker, VirtMachineState *vms)
> }
> }
>
> - if (vms->its) {
> + if (virt_is_its_enabled(vms)) {
> /*
> * Map bypassed (don't go through the SMMU) RIDs (input) to
> * ITS Group node directly: RC -> ITS.
> @@ -945,7 +945,7 @@ build_madt(GArray *table_data, BIOSLinker *linker, VirtMachineState *vms)
> memmap[VIRT_HIGH_GIC_REDIST2].size);
> }
>
> - if (vms->its) {
> + if (virt_is_its_enabled(vms)) {
> /*
> * ACPI spec, Revision 6.0 Errata A
> * (original 6.0 definition has invalid Length)
> @@ -961,7 +961,7 @@ build_madt(GArray *table_data, BIOSLinker *linker, VirtMachineState *vms)
> }
> }
>
> - if (!(vms->gic_version != VIRT_GIC_VERSION_2 && vms->its) && !vms->no_gicv3_with_gicv2m) {
> + if (virt_is_its_enabled(vms) && !vms->no_gicv3_with_gicv2m) {
> const uint16_t spi_base = vms->irqmap[VIRT_GIC_V2M] + ARM_SPI_BASE;
>
> /* 5.2.12.16 GIC MSI Frame Structure */
> diff --git a/hw/arm/virt.c b/hw/arm/virt.c
> index 9121eb37eb..1bebbc265d 100644
> --- a/hw/arm/virt.c
> +++ b/hw/arm/virt.c
> @@ -735,7 +735,7 @@ static void create_its(VirtMachineState *vms)
> {
> DeviceState *dev;
>
> - assert(vms->its);
> + assert(virt_is_its_enabled(vms));
> if (!kvm_irqchip_in_kernel() && !vms->tcg_its) {
> /*
> * Do nothing if ITS is neither supported by the host nor emulated by
> @@ -744,6 +744,15 @@ static void create_its(VirtMachineState *vms)
> return;
> }
>
> + if (whpx_enabled() && vms->tcg_its) {
> + /*
> + * Signal to the user when ITS is neither supported by the host
> + * nor emulated by the machine.
> + */
> + info_report("ITS not supported on WHPX.");
> + exit(1);
> + }
> +
> dev = qdev_new(its_class_name());
>
> object_property_set_link(OBJECT(dev), "parent-gicv3", OBJECT(vms->gic),
> @@ -955,7 +964,7 @@ static void create_gic(VirtMachineState *vms, MemoryRegion *mem)
>
> fdt_add_gic_node(vms);
>
> - if (vms->gic_version != VIRT_GIC_VERSION_2 && vms->its) {
> + if (vms->gic_version != VIRT_GIC_VERSION_2 && virt_is_its_enabled(vms)) {
> create_its(vms);
> } else if (vms->gic_version != VIRT_GIC_VERSION_2 && !vms->no_gicv3_with_gicv2m) {
> create_v2m(vms);
> @@ -2705,18 +2714,38 @@ static void virt_set_highmem_mmio_size(Object *obj, Visitor *v,
> extended_memmap[VIRT_HIGH_PCIE_MMIO].size = size;
> }
>
> -static bool virt_get_its(Object *obj, Error **errp)
> +bool virt_is_its_enabled(VirtMachineState *vms)
> +{
> + if (vms->gic_version == VIRT_GIC_VERSION_2) {
> + return false;
> + }
> +
> + if (vms->its == ON_OFF_AUTO_OFF) {
> + return false;
> + }
> + if (vms->its == ON_OFF_AUTO_AUTO) {
> + if (whpx_enabled()) {
> + return false;
> + }
> + }
> + return true;
> +}
> +
> +static void virt_get_its(Object *obj, Visitor *v, const char *name,
> + void *opaque, Error **errp)
> {
> VirtMachineState *vms = VIRT_MACHINE(obj);
> + OnOffAuto its = vms->its;
>
> - return vms->its;
> + visit_type_OnOffAuto(v, name, &its, errp);
> }
>
> -static void virt_set_its(Object *obj, bool value, Error **errp)
> +static void virt_set_its(Object *obj, Visitor *v, const char *name,
> + void *opaque, Error **errp)
> {
> VirtMachineState *vms = VIRT_MACHINE(obj);
>
> - vms->its = value;
> + visit_type_OnOffAuto(v, name, &vms->its, errp);
> }
>
> static bool virt_get_dtb_randomness(Object *obj, Error **errp)
> @@ -3426,8 +3455,9 @@ static void virt_machine_class_init(ObjectClass *oc, const void *data)
> "guest CPU which implements the ARM "
> "Memory Tagging Extension");
>
> - object_class_property_add_bool(oc, "its", virt_get_its,
> - virt_set_its);
> + object_class_property_add(oc, "its", "OnOffAuto",
> + virt_get_its, virt_set_its,
> + NULL, NULL);
> object_class_property_set_description(oc, "its",
> "Set on/off to enable/disable "
> "ITS instantiation");
> @@ -3487,8 +3517,8 @@ static void virt_instance_init(Object *obj)
> vms->highmem_mmio = true;
> vms->highmem_redists = true;
>
> - /* Default allows ITS instantiation */
> - vms->its = true;
> + /* Default allows ITS instantiation if available */
> + vms->its = ON_OFF_AUTO_AUTO;
> /* Allow ITS emulation if the machine version supports it */
> vms->tcg_its = !vmc->no_tcg_its;
> vms->no_gicv3_with_gicv2m = false;
> diff --git a/include/hw/arm/virt.h b/include/hw/arm/virt.h
> index d31348dd61..997dd51678 100644
> --- a/include/hw/arm/virt.h
> +++ b/include/hw/arm/virt.h
> @@ -149,7 +149,7 @@ struct VirtMachineState {
> bool highmem_ecam;
> bool highmem_mmio;
> bool highmem_redists;
> - bool its;
> + OnOffAuto its;
> bool tcg_its;
> bool virt;
> bool ras;
> @@ -218,4 +218,6 @@ static inline int virt_gicv3_redist_region_count(VirtMachineState *vms)
> vms->highmem_redists) ? 2 : 1;
> }
>
> +bool virt_is_its_enabled(VirtMachineState *vms);
> +
> #endif /* QEMU_ARM_VIRT_H */
> --
> 2.50.1 (Apple Git-155)
>
next prev parent reply other threads:[~2025-10-16 18:22 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-16 16:54 [PATCH v7 00/24] WHPX support for Arm Mohamed Mediouni
2025-10-16 16:54 ` [PATCH v7 01/24] target/arm/kvm: add constants for new PSCI versions Mohamed Mediouni
2025-10-16 17:07 ` Pierrick Bouvier
2025-10-16 16:54 ` [PATCH v7 02/24] accel/system: Introduce hwaccel_enabled() helper Mohamed Mediouni
2025-10-16 16:54 ` [PATCH v7 03/24] qtest: hw/arm: virt: skip ACPI test for ITS off Mohamed Mediouni
2025-10-16 17:08 ` Pierrick Bouvier
2025-10-16 16:55 ` [PATCH v7 04/24] hw/arm: virt: add GICv2m for the case when ITS is not available Mohamed Mediouni
2025-10-16 17:09 ` Pierrick Bouvier
2025-10-16 16:55 ` [PATCH v7 05/24] tests: data: update AArch64 ACPI tables Mohamed Mediouni
2025-10-16 17:10 ` Pierrick Bouvier
2025-10-16 16:55 ` [PATCH v7 06/24] whpx: Move around files before introducing AArch64 support Mohamed Mediouni
2025-10-16 16:55 ` [PATCH v7 07/24] whpx: reshuffle common code Mohamed Mediouni
2025-10-16 16:55 ` [PATCH v7 08/24] whpx: ifdef out winhvemulation on non-x86_64 Mohamed Mediouni
2025-10-16 16:55 ` [PATCH v7 09/24] whpx: common: add WHPX_INTERCEPT_DEBUG_TRAPS define Mohamed Mediouni
2025-10-16 16:55 ` [PATCH v7 10/24] hw, target, accel: whpx: change apic_in_platform to kernel_irqchip Mohamed Mediouni
2025-10-16 16:55 ` [PATCH v7 11/24] whpx: interrupt controller support Mohamed Mediouni
2025-10-16 16:55 ` [PATCH v7 12/24] whpx: add arm64 support Mohamed Mediouni
2025-10-16 16:55 ` [PATCH v7 13/24] whpx: copy over memory management logic from hvf Mohamed Mediouni
2025-10-22 18:55 ` Philippe Mathieu-Daudé
2025-10-16 16:55 ` [PATCH v7 14/24] target/arm: cpu: mark WHPX as supporting PSCI 1.3 Mohamed Mediouni
2025-10-16 16:55 ` [PATCH v7 15/24] hw/arm: virt: cleanly fail on attempt to use the platform vGIC together with ITS Mohamed Mediouni
2025-10-16 17:10 ` Pierrick Bouvier
2025-10-16 18:19 ` Mohamed Mediouni [this message]
2025-10-16 16:55 ` [PATCH v7 16/24] docs: arm: update virt machine model description Mohamed Mediouni
2025-10-16 17:11 ` Pierrick Bouvier
2025-10-16 16:55 ` [PATCH v7 17/24] whpx: arm64: clamp down IPA size Mohamed Mediouni
2025-10-16 16:55 ` [PATCH v7 18/24] hw/arm, accel/hvf, whpx: unify get_physical_address_range between WHPX and HVF Mohamed Mediouni
2025-10-16 16:55 ` [PATCH v7 19/24] whpx: arm64: implement -cpu host Mohamed Mediouni
2025-10-16 17:12 ` Pierrick Bouvier
2025-10-16 16:55 ` [PATCH v7 20/24] target/arm: whpx: instantiate GIC early Mohamed Mediouni
2025-10-16 16:55 ` [PATCH v7 21/24] whpx: arm64: gicv3: add migration blocker Mohamed Mediouni
2025-10-16 17:14 ` Pierrick Bouvier
2025-10-16 16:55 ` [PATCH v7 22/24] whpx: enable arm64 builds Mohamed Mediouni
2025-10-16 17:14 ` Pierrick Bouvier
2025-10-16 16:55 ` [PATCH v7 23/24] MAINTAINERS: update maintainers for WHPX Mohamed Mediouni
2025-10-16 17:15 ` Pierrick Bouvier
2025-10-16 16:55 ` [PATCH v7 24/24] whpx: apic: use non-deprecated APIs to control interrupt controller state Mohamed Mediouni
2025-10-16 17:15 ` Pierrick Bouvier
2025-10-20 10:27 ` Bernhard Beschow
2025-10-23 6:33 ` Philippe Mathieu-Daudé
2025-10-23 8:47 ` Pierrick Bouvier
2025-10-23 9:23 ` Bernhard Beschow
2025-10-23 17:02 ` Bernhard Beschow
2025-10-27 7:55 ` Pierrick Bouvier
2025-10-17 9:59 ` [PATCH v7 00/24] WHPX support for Arm Peter Maydell
2025-10-17 10:56 ` Mohamed Mediouni
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=25E6C302-43A7-4D52-BFBA-0712085DF63F@unpredictable.fr \
--to=mohamed@unpredictable.fr \
--cc=agraf@csgraf.de \
--cc=anisinha@redhat.com \
--cc=berrange@redhat.com \
--cc=dirty@apple.com \
--cc=eduardo@habkost.net \
--cc=imammedo@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=mads@ynddal.dk \
--cc=marcandre.lureau@redhat.com \
--cc=marcel.apfelbaum@gmail.com \
--cc=mst@redhat.com \
--cc=pbarbuda@microsoft.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=richard.henderson@linaro.org \
--cc=shannon.zhaosl@gmail.com \
--cc=sunilmut@microsoft.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 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).