From: Igor Mammedov <imammedo@redhat.com>
To: salil.mehta@opnsrc.net
Cc: qemu-devel@nongnu.org, qemu-arm@nongnu.org, mst@redhat.com,
salil.mehta@huawei.com, maz@kernel.org, jean-philippe@linaro.org,
jonathan.cameron@huawei.com, lpieralisi@kernel.org,
peter.maydell@linaro.org, richard.henderson@linaro.org,
armbru@redhat.com, andrew.jones@linux.dev, david@redhat.com,
philmd@linaro.org, eric.auger@redhat.com, will@kernel.org,
ardb@kernel.org, oliver.upton@linux.dev, pbonzini@redhat.com,
gshan@redhat.com, rafael@kernel.org, borntraeger@linux.ibm.com,
alex.bennee@linaro.org, gustavo.romero@linaro.org,
npiggin@gmail.com, harshpb@linux.ibm.com, linux@armlinux.org.uk,
darren@os.amperecomputing.com, ilkka@os.amperecomputing.com,
vishnu@os.amperecomputing.com,
gankulkarni@os.amperecomputing.com, karl.heubaum@oracle.com,
miguel.luis@oracle.com, zhukeqian1@huawei.com,
wangxiongfeng2@huawei.com, wangyanan55@huawei.com,
wangzhou1@hisilicon.com, linuxarm@huawei.com,
jiakernel2@gmail.com, maobibo@loongson.cn,
lixianglai@loongson.cn, shahuang@redhat.com, zhao1.liu@intel.com
Subject: Re: [PATCH RFC V6 10/24] arm/virt: Init PMU at host for all present vCPUs
Date: Fri, 3 Oct 2025 17:02:45 +0200 [thread overview]
Message-ID: <20251003170245.75ad2405@fedora> (raw)
In-Reply-To: <20251001010127.3092631-11-salil.mehta@opnsrc.net>
On Wed, 1 Oct 2025 01:01:13 +0000
salil.mehta@opnsrc.net wrote:
> From: Salil Mehta <salil.mehta@huawei.com>
>
...
> diff --git a/include/hw/core/cpu.h b/include/hw/core/cpu.h
> index 5eaf41a566..2ee202a8a5 100644
> --- a/include/hw/core/cpu.h
> +++ b/include/hw/core/cpu.h
> @@ -602,6 +602,63 @@ extern CPUTailQ cpus_queue;
> #define CPU_FOREACH_SAFE(cpu, next_cpu) \
> QTAILQ_FOREACH_SAFE_RCU(cpu, &cpus_queue, node, next_cpu)
>
> +
> +/**
> + * CPU_FOREACH_POSSIBLE(cpu_, archid_list_)
> + *
> + * Iterate over all entries in a CPUArchIdList, assigning each entry’s
> + * CPUState* to @cpu_. This hides the loop index and reads like a normal
> + * C for-loop.
> + *
> + * A CPUArchIdList represents the set of *possible* CPUs for a machine.
> + * Each entry contains:
> + * - @cpu: CPUState pointer, or NULL if not realized yet
> + * - @arch_id: architecture-specific identifier (e.g. MPIDR)
> + * - @vcpus_count: number of vCPUs represented (usually 1)
> + *
> + * The list models *possible* CPUs: it includes (a) currently plugged vCPUs
> + * made available through hotplug, (b) present (and perhaps visible to OSPM)
> + * but kept ACPI-disabled vCPUs, and (c) reserved slots for CPUs that may be
> + * created in the future. This supports co-existence of hotpluggable and
> + * admin-disabled vCPUs if architectures permit.
> + *
> + * Example:
> + *
> + * CPUArchIdList *alist = machine_possible_cpus(ms);
> + * CPUState *cpu;
> + *
> + * CPU_FOREACH_POSSIBLE(cpu, alist) {
> + * if (!cpu) {
> + * continue; // reserved slot for hotplug case
> + * }
> + *
> + * < Do Something >
> + * }
> + *
> + * Expanded equivalent:
> + *
> + * for (int __cpu_idx = 0; alist && __cpu_idx < alist->len; __cpu_idx++) {
> + * if ((cpu = alist->cpus[__cpu_idx].cpu, 1)) {
> + * if (!cpu) {
> + * continue;
> + * }
> + *
> + * < Do Something >
> + * }
> + * }
> + *
> + * Notes:
> + * - Callers must check @cpu for NULL when filtering unplugged CPUs.
> + * - Mirrors the style of CPU_FOREACH(), but iterates all *possible* CPUs
> + * (plugged, ACPI-disabled, and reserved slots) rather than only present
> + * and enabled vCPUs.
> + */
> +#define CPU_FOREACH_POSSIBLE(cpu_, archid_list_) \
> + for (int __cpu_idx = 0; \
> + (archid_list_) && __cpu_idx < (archid_list_)->len; \
> + __cpu_idx++) \
> + if (((cpu_) = (archid_list_)->cpus[__cpu_idx].cpu, 1))
> +
> extern __thread CPUState *current_cpu;
make it a separate patch and refactor existing loops to use it.
>
> /**
next prev parent reply other threads:[~2025-10-03 15:04 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-01 1:01 [PATCH RFC V6 00/24] Support of Virtual CPU Hotplug-like Feature for ARMv8+ Arch salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 01/24] hw/core: Introduce administrative power-state property and its accessors salil.mehta
2025-10-09 10:48 ` Miguel Luis
2025-10-01 1:01 ` [PATCH RFC V6 02/24] hw/core, qemu-options.hx: Introduce 'disabledcpus' SMP parameter salil.mehta
2025-10-09 11:28 ` Miguel Luis
2025-10-09 13:17 ` Igor Mammedov
2025-10-09 11:51 ` Markus Armbruster
2025-10-01 1:01 ` [PATCH RFC V6 03/24] hw/arm/virt: Clamp 'maxcpus' as-per machine's vCPU deferred online-capability salil.mehta
2025-10-09 12:32 ` Miguel Luis
2025-10-09 13:11 ` Igor Mammedov
2025-10-01 1:01 ` [PATCH RFC V6 04/24] arm/virt, target/arm: Add new ARMCPU {socket, cluster, core, thread}-id property salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 05/24] arm/virt, kvm: Pre-create KVM vCPUs for 'disabled' QOM vCPUs at machine init salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 06/24] arm/virt, gicv3: Pre-size GIC with possible " salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 07/24] arm/gicv3: Refactor CPU interface init for shared TCG/KVM use salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 08/24] arm/virt, gicv3: Guard CPU interface access for admin disabled vCPUs salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 09/24] hw/intc/arm_gicv3_common: Migrate & check 'GICv3CPUState' accessibility mismatch salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 10/24] arm/virt: Init PMU at host for all present vCPUs salil.mehta
2025-10-03 15:02 ` Igor Mammedov [this message]
2025-10-01 1:01 ` [PATCH RFC V6 11/24] hw/arm/acpi: MADT change to size the guest with possible vCPUs salil.mehta
2025-10-03 15:09 ` Igor Mammedov
[not found] ` <0175e40f70424dd9a29389b8a4f16c42@huawei.com>
2025-10-07 12:20 ` Igor Mammedov
2025-10-10 3:15 ` Salil Mehta
2025-10-01 1:01 ` [PATCH RFC V6 12/24] hw/core: Introduce generic device power-state handler interface salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 13/24] qdev: make admin power state changes trigger platform transitions via ACPI salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 14/24] arm/acpi: Introduce dedicated CPU OSPM interface for ARM-like platforms salil.mehta
2025-10-03 14:58 ` Igor Mammedov
[not found] ` <7da6a9c470684754810414f0abd23a62@huawei.com>
2025-10-07 12:06 ` Igor Mammedov
2025-10-10 3:00 ` Salil Mehta
2025-10-01 1:01 ` [PATCH RFC V6 15/24] acpi/ged: Notify OSPM of CPU administrative state changes via GED salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 16/24] arm/virt/acpi: Update ACPI DSDT Tbl to include 'Online-Capable' CPUs AML salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 17/24] hw/arm/virt, acpi/ged: Add PowerStateHandler hooks for runtime CPU state changes salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 18/24] target/arm/kvm, tcg: Handle SMCCC hypercall exits in VMM during PSCI_CPU_{ON, OFF} salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 19/24] target/arm/cpu: Add the Accessor hook to fetch ARM CPU arch-id salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 20/24] target/arm/kvm: Write vCPU's state back to KVM on cold-reset salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 21/24] hw/intc/arm-gicv3-kvm: Pause all vCPUs & cache ICC_CTLR_EL1 for userspace PSCI CPU_ON salil.mehta
2025-10-01 1:01 ` [PATCH RFC V6 22/24] monitor, qdev: Introduce 'device_set' to change admin state of existing devices salil.mehta
2025-10-09 8:55 ` [PATCH RFC V6 22/24] monitor,qdev: " Markus Armbruster
2025-10-09 12:51 ` Igor Mammedov
2025-10-09 14:03 ` Daniel P. Berrangé
2025-10-09 14:55 ` Markus Armbruster
2025-10-09 15:19 ` Peter Maydell
2025-10-10 4:59 ` Markus Armbruster
2025-10-01 1:01 ` [PATCH RFC V6 23/24] monitor, qapi: add 'info cpus-powerstate' and QMP query (Admin + Oper states) salil.mehta
2025-10-09 11:53 ` [PATCH RFC V6 23/24] monitor,qapi: " Markus Armbruster
2025-10-01 1:01 ` [PATCH RFC V6 24/24] tcg: Defer TB flush for 'lazy realized' vCPUs on first region alloc salil.mehta
2025-10-01 21:34 ` Richard Henderson
2025-10-02 12:27 ` Salil Mehta via
2025-10-02 15:41 ` Richard Henderson
2025-10-07 10:14 ` Salil Mehta via
2025-10-06 14:00 ` [PATCH RFC V6 00/24] Support of Virtual CPU Hotplug-like Feature for ARMv8+ Arch Igor Mammedov
2025-10-13 0:34 ` Gavin Shan
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=20251003170245.75ad2405@fedora \
--to=imammedo@redhat.com \
--cc=alex.bennee@linaro.org \
--cc=andrew.jones@linux.dev \
--cc=ardb@kernel.org \
--cc=armbru@redhat.com \
--cc=borntraeger@linux.ibm.com \
--cc=darren@os.amperecomputing.com \
--cc=david@redhat.com \
--cc=eric.auger@redhat.com \
--cc=gankulkarni@os.amperecomputing.com \
--cc=gshan@redhat.com \
--cc=gustavo.romero@linaro.org \
--cc=harshpb@linux.ibm.com \
--cc=ilkka@os.amperecomputing.com \
--cc=jean-philippe@linaro.org \
--cc=jiakernel2@gmail.com \
--cc=jonathan.cameron@huawei.com \
--cc=karl.heubaum@oracle.com \
--cc=linux@armlinux.org.uk \
--cc=linuxarm@huawei.com \
--cc=lixianglai@loongson.cn \
--cc=lpieralisi@kernel.org \
--cc=maobibo@loongson.cn \
--cc=maz@kernel.org \
--cc=miguel.luis@oracle.com \
--cc=mst@redhat.com \
--cc=npiggin@gmail.com \
--cc=oliver.upton@linux.dev \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=philmd@linaro.org \
--cc=qemu-arm@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=rafael@kernel.org \
--cc=richard.henderson@linaro.org \
--cc=salil.mehta@huawei.com \
--cc=salil.mehta@opnsrc.net \
--cc=shahuang@redhat.com \
--cc=vishnu@os.amperecomputing.com \
--cc=wangxiongfeng2@huawei.com \
--cc=wangyanan55@huawei.com \
--cc=wangzhou1@hisilicon.com \
--cc=will@kernel.org \
--cc=zhao1.liu@intel.com \
--cc=zhukeqian1@huawei.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).