From: Haibo Xu <haibo.xu@linaro.org>
To: Andrew Jones <drjones@redhat.com>
Cc: Peter Maydell <peter.maydell@linaro.org>,
qemu-arm@nongnu.org,
Richard Henderson <richard.henderson@linaro.org>,
qemu-devel@nongnu.org, philmd@redhat.com
Subject: Re: [PATCH v2 09/12] hw/arm/virt: spe: Add SPE fdt binding for virt machine
Date: Wed, 9 Sep 2020 15:51:14 +0800 [thread overview]
Message-ID: <CAJc+Z1GCmf5fEjifa84O6jmfd6WCih1jT07AXb7vWFC7=sXfrA@mail.gmail.com> (raw)
In-Reply-To: <20200908111625.2wzzn7jh5wt4nnmu@kamzik.brq.redhat.com>
On Tue, 8 Sep 2020 at 19:16, Andrew Jones <drjones@redhat.com> wrote:
>
>
> This patch does much more than the summary "hw/arm/virt: spe: Add SPE fdt
> binding for virt machine" says it does. Please revise the summary.
>
Will revise it in v3.
Thanks,
Haibo
> On Tue, Sep 08, 2020 at 08:13:27AM +0000, Haibo Xu wrote:
> > Add a virtual SPE device for virt machine while using
> > PPI 5 for SPE overflow interrupt number which has already
> > selected in kvmtool.
> >
> > Signed-off-by: Haibo Xu <haibo.xu@linaro.org>
> > ---
> > hw/arm/virt-acpi-build.c | 3 +++
> > hw/arm/virt.c | 43 ++++++++++++++++++++++++++++++++++++-
> > include/hw/acpi/acpi-defs.h | 3 +++
> > include/hw/arm/virt.h | 1 +
> > target/arm/cpu.c | 2 ++
> > target/arm/cpu.h | 2 ++
> > 6 files changed, 53 insertions(+), 1 deletion(-)
> >
> > diff --git a/hw/arm/virt-acpi-build.c b/hw/arm/virt-acpi-build.c
> > index 9efd7a3881..3fd80fda53 100644
> > --- a/hw/arm/virt-acpi-build.c
> > +++ b/hw/arm/virt-acpi-build.c
> > @@ -665,6 +665,9 @@ build_madt(GArray *table_data, BIOSLinker *linker, VirtMachineState *vms)
> > if (arm_feature(&armcpu->env, ARM_FEATURE_PMU)) {
> > gicc->performance_interrupt = cpu_to_le32(PPI(VIRTUAL_PMU_IRQ));
> > }
> > + if (cpu_isar_feature(aa64_spe, armcpu)) {
> > + gicc->spe_interrupt = cpu_to_le32(PPI(VIRTUAL_SPE_IRQ));
> > + }
> > if (vms->virt) {
> > gicc->vgic_interrupt = cpu_to_le32(PPI(ARCH_GIC_MAINT_IRQ));
> > }
> > diff --git a/hw/arm/virt.c b/hw/arm/virt.c
> > index 6bacfb668d..bdb1ce925c 100644
> > --- a/hw/arm/virt.c
> > +++ b/hw/arm/virt.c
> > @@ -545,6 +545,32 @@ static void fdt_add_pmu_nodes(const VirtMachineState *vms)
> > }
> > }
> >
> > +static void fdt_add_spe_nodes(const VirtMachineState *vms)
> > +{
> > + ARMCPU *armcpu = ARM_CPU(first_cpu);
> > + uint32_t irqflags = GIC_FDT_IRQ_FLAGS_LEVEL_HI;
> > +
> > + if (!cpu_isar_feature(aa64_spe, armcpu)) {
> > + assert(!object_property_get_bool(OBJECT(armcpu), "spe", NULL));
> > + return;
> > + }
> > +
> > + if (vms->gic_version == VIRT_GIC_VERSION_2) {
> > + irqflags = deposit32(irqflags, GIC_FDT_IRQ_PPI_CPU_START,
> > + GIC_FDT_IRQ_PPI_CPU_WIDTH,
> > + (1 << vms->smp_cpus) - 1);
> > + }
> > +
> > + qemu_fdt_add_subnode(vms->fdt, "/spe");
> > + if (arm_feature(&armcpu->env, ARM_FEATURE_V8)) {
> > + const char compat[] = "arm,statistical-profiling-extension-v1";
> > + qemu_fdt_setprop(vms->fdt, "/spe", "compatible",
> > + compat, sizeof(compat));
> > + qemu_fdt_setprop_cells(vms->fdt, "/spe", "interrupts",
> > + GIC_FDT_IRQ_TYPE_PPI, VIRTUAL_SPE_IRQ, irqflags);
> > + }
> > +}
> > +
> > static inline DeviceState *create_acpi_ged(VirtMachineState *vms)
> > {
> > DeviceState *dev;
> > @@ -717,6 +743,10 @@ static void create_gic(VirtMachineState *vms)
> > qdev_get_gpio_in(vms->gic, ppibase
> > + VIRTUAL_PMU_IRQ));
> >
> > + qdev_connect_gpio_out_named(cpudev, "spe-interrupt", 0,
> > + qdev_get_gpio_in(vms->gic, ppibase
> > + + VIRTUAL_SPE_IRQ));
> > +
> > sysbus_connect_irq(gicbusdev, i, qdev_get_gpio_in(cpudev, ARM_CPU_IRQ));
> > sysbus_connect_irq(gicbusdev, i + smp_cpus,
> > qdev_get_gpio_in(cpudev, ARM_CPU_FIQ));
> > @@ -1664,11 +1694,12 @@ static void finalize_gic_version(VirtMachineState *vms)
> >
> > static void virt_cpu_post_init(VirtMachineState *vms)
> > {
> > - bool aarch64, pmu;
> > + bool aarch64, pmu, spe;
> > CPUState *cpu;
> >
> > aarch64 = object_property_get_bool(OBJECT(first_cpu), "aarch64", NULL);
> > pmu = object_property_get_bool(OBJECT(first_cpu), "pmu", NULL);
> > + spe = object_property_get_bool(OBJECT(first_cpu), "spe", NULL);
> >
> > if (kvm_enabled()) {
> > CPU_FOREACH(cpu) {
> > @@ -1679,6 +1710,14 @@ static void virt_cpu_post_init(VirtMachineState *vms)
> > }
> > kvm_arm_pmu_init(cpu);
> > }
> > +
> > + if (spe) {
> > + assert(ARM_CPU(cpu)->has_spe == ON_OFF_AUTO_ON);
> > + if (kvm_irqchip_in_kernel()) {
> > + kvm_arm_spe_set_irq(cpu, PPI(VIRTUAL_SPE_IRQ));
> > + }
> > + kvm_arm_spe_init(cpu);
>
> A later patch introduces userspace irqchip support. Should we avoid
> allowing it until then to avoid breaking bisection?
>
Yes, it's possible to break the bisection. To avoid it I think we can
move the above codes
block to a separate patch after adding the userspace irqchip support,
Or, just put the userspace
irqchip support patch before this patch. What's your opinion?
Thanks,
Haibo
> > + }
> > }
> > } else {
> > if (aarch64 && vms->highmem) {
> > @@ -1927,6 +1966,8 @@ static void machvirt_init(MachineState *machine)
> >
> > fdt_add_pmu_nodes(vms);
> >
> > + fdt_add_spe_nodes(vms);
> > +
> > create_uart(vms, VIRT_UART, sysmem, serial_hd(0));
> >
> > if (vms->secure) {
> > diff --git a/include/hw/acpi/acpi-defs.h b/include/hw/acpi/acpi-defs.h
> > index 38a42f409a..21e58f27c5 100644
> > --- a/include/hw/acpi/acpi-defs.h
> > +++ b/include/hw/acpi/acpi-defs.h
> > @@ -302,6 +302,9 @@ struct AcpiMadtGenericCpuInterface {
> > uint32_t vgic_interrupt;
> > uint64_t gicr_base_address;
> > uint64_t arm_mpidr;
> > + uint8_t efficiency_class;
> > + uint8_t reserved2[1];
> > + uint16_t spe_interrupt; /* ACPI 6.3 */
> > } QEMU_PACKED;
> >
> > typedef struct AcpiMadtGenericCpuInterface AcpiMadtGenericCpuInterface;
> > diff --git a/include/hw/arm/virt.h b/include/hw/arm/virt.h
> > index 72c269aaa5..6013b6d535 100644
> > --- a/include/hw/arm/virt.h
> > +++ b/include/hw/arm/virt.h
> > @@ -49,6 +49,7 @@
> > #define ARCH_TIMER_NS_EL1_IRQ 14
> > #define ARCH_TIMER_NS_EL2_IRQ 10
> >
> > +#define VIRTUAL_SPE_IRQ 5
> > #define VIRTUAL_PMU_IRQ 7
> >
> > #define PPI(irq) ((irq) + 16)
> > diff --git a/target/arm/cpu.c b/target/arm/cpu.c
> > index f211958eaa..786cc6134c 100644
> > --- a/target/arm/cpu.c
> > +++ b/target/arm/cpu.c
> > @@ -1041,6 +1041,8 @@ static void arm_cpu_initfn(Object *obj)
> > "gicv3-maintenance-interrupt", 1);
> > qdev_init_gpio_out_named(DEVICE(cpu), &cpu->pmu_interrupt,
> > "pmu-interrupt", 1);
> > + qdev_init_gpio_out_named(DEVICE(cpu), &cpu->spe_interrupt,
> > + "spe-interrupt", 1);
> > #endif
> >
> > /* DTB consumers generally don't in fact care what the 'compatible'
> > diff --git a/target/arm/cpu.h b/target/arm/cpu.h
> > index baf2bbcee8..395a1e5df8 100644
> > --- a/target/arm/cpu.h
> > +++ b/target/arm/cpu.h
> > @@ -800,6 +800,8 @@ struct ARMCPU {
> > qemu_irq gicv3_maintenance_interrupt;
> > /* GPIO output for the PMU interrupt */
> > qemu_irq pmu_interrupt;
> > + /* GPIO output for the SPE interrupt */
> > + qemu_irq spe_interrupt;
> >
> > /* MemoryRegion to use for secure physical accesses */
> > MemoryRegion *secure_memory;
> > --
> > 2.17.1
> >
> >
>
> Otherwise
>
> Reviewed-by: Andrew Jones <drjones@redhat.com>
>
next prev parent reply other threads:[~2020-09-09 7:52 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-08 8:13 [PATCH v2 00/12] target/arm: Add vSPE support to KVM guest Haibo Xu
2020-09-08 8:13 ` [PATCH v2 01/12] update Linux headers with new vSPE macros Haibo Xu
2020-09-08 8:13 ` [PATCH v2 02/12] target/arm/kvm: spe: Add helper to detect SPE when using KVM Haibo Xu
2020-09-08 10:43 ` Andrew Jones
2020-09-08 8:13 ` [PATCH v2 03/12] target/arm/cpu: spe: Add an option to turn on/off vSPE support Haibo Xu
2020-09-08 10:51 ` Andrew Jones
2020-09-08 8:13 ` [PATCH v2 04/12] target/arm: spe: Only enable SPE from 5.2 compat machines Haibo Xu
2020-09-08 10:52 ` Andrew Jones
2020-09-08 8:13 ` [PATCH v2 05/12] target/arm/kvm: spe: Unify device attr operation helper Haibo Xu
2020-09-08 10:56 ` Andrew Jones
2020-09-09 2:39 ` Haibo Xu
2020-09-08 8:13 ` [PATCH v2 06/12] target/arm/kvm: spe: Add device init and set_irq operations Haibo Xu
2020-09-08 10:58 ` Andrew Jones
2020-09-08 8:13 ` [PATCH v2 07/12] hw/arm/virt: Move post cpu realize check into its own function Haibo Xu
2020-09-08 11:03 ` Andrew Jones
2020-09-09 6:54 ` Haibo Xu
2020-09-08 8:13 ` [PATCH v2 08/12] hw/arm/virt: Move kvm pmu setup to virt_cpu_post_init Haibo Xu
2020-09-08 11:07 ` Andrew Jones
2020-09-08 8:13 ` [PATCH v2 09/12] hw/arm/virt: spe: Add SPE fdt binding for virt machine Haibo Xu
2020-09-08 11:16 ` Andrew Jones
2020-09-09 7:51 ` Haibo Xu [this message]
2020-09-09 9:34 ` Andrew Jones
2020-09-10 0:31 ` Haibo Xu
2020-09-08 8:13 ` [PATCH v2 10/12] target/arm/cpu: spe: Enable spe to work with host cpu Haibo Xu
2020-09-08 11:32 ` Andrew Jones
2020-09-09 8:35 ` Haibo Xu
2020-09-08 8:13 ` [PATCH v2 11/12] target/arm/kvm: spe: Enable userspace irqchip support Haibo Xu
2020-09-08 11:34 ` Andrew Jones
2020-09-09 3:12 ` Haibo Xu
2020-09-08 8:13 ` [PATCH v2 12/12] target/arm: spe: Add corresponding doc and test Haibo Xu
2020-09-08 11:41 ` Andrew Jones
2020-09-09 3:19 ` Haibo Xu
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='CAJc+Z1GCmf5fEjifa84O6jmfd6WCih1jT07AXb7vWFC7=sXfrA@mail.gmail.com' \
--to=haibo.xu@linaro.org \
--cc=drjones@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=philmd@redhat.com \
--cc=qemu-arm@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=richard.henderson@linaro.org \
/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).