Message-ID: <87bktlyszw.wl-maz@kernel.org>
Date:   Tue, 19 Jul 2022 17:20:03 +0100
From:   Marc Zyngier <maz@...nel.org>
To:     Anup Patel <apatel@...tanamicro.com>
Cc:     Palmer Dabbelt <palmer@...belt.com>,
        Paul Walmsley <paul.walmsley@...ive.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Daniel Lezcano <daniel.lezcano@...aro.org>,
        Atish Patra <atishp@...shpatra.org>,
        Alistair Francis <Alistair.Francis@....com>,
        Anup Patel <anup@...infault.org>,
        linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v6 3/7] genirq: Add mechanism to multiplex a single HW IPI

On Mon, 18 Apr 2022 11:53:01 +0100,
Anup Patel <apatel@...tanamicro.com> wrote:
> 
> All RISC-V platforms have a single HW IPI provided by the INTC local
> interrupt controller. The HW method to trigger INTC IPI can be through
> external irqchip (e.g. RISC-V AIA), through platform specific device
> (e.g. SiFive CLINT timer), or through firmware (e.g. SBI IPI call).
> 
> To support multiple IPIs on RISC-V, we add a generic IPI multiplexing
> mechanism which help us create multiple virtual IPIs using a single
> HW IPI. This generic IPI multiplexing is shared among various RISC-V
> irqchip drivers.
> 
> Signed-off-by: Anup Patel <apatel@...tanamicro.com>
> ---
>  include/linux/irq.h  |  11 +++
>  kernel/irq/Kconfig   |   4 +
>  kernel/irq/Makefile  |   1 +
>  kernel/irq/ipi-mux.c | 197 +++++++++++++++++++++++++++++++++++++++++++
>  4 files changed, 213 insertions(+)
>  create mode 100644 kernel/irq/ipi-mux.c
> 
> diff --git a/include/linux/irq.h b/include/linux/irq.h
> index f92788ccdba2..5bb4e2db63d7 100644
> --- a/include/linux/irq.h
> +++ b/include/linux/irq.h
> @@ -1247,6 +1247,17 @@ int __ipi_send_mask(struct irq_desc *desc, const struct cpumask *dest);
>  int ipi_send_single(unsigned int virq, unsigned int cpu);
>  int ipi_send_mask(unsigned int virq, const struct cpumask *dest);
>  
> +#define IPI_MUX_NR_IRQS		BITS_PER_LONG
> +
> +struct ipi_mux_ops {
> +	void (*ipi_mux_clear)(unsigned int parent_virq);
> +	void (*ipi_mux_send)(unsigned int parent_virq,
> +			     const struct cpumask *mask);

You really cannot just dump this like this. This requires
documentation so that architecture maintainers can move over to this.
I appreciate that this area is pretty poorly documented, but we need
to start somewhere.

> +};
> +
> +void ipi_mux_process(void);
> +int ipi_mux_create(unsigned int parent_virq, const struct ipi_mux_ops *ops);
> +
>  #ifdef CONFIG_GENERIC_IRQ_MULTI_HANDLER
>  /*
>   * Registers a generic IRQ handling function as the top-level IRQ handler in
> diff --git a/kernel/irq/Kconfig b/kernel/irq/Kconfig
> index 10929eda9825..2388e7d40ed3 100644
> --- a/kernel/irq/Kconfig
> +++ b/kernel/irq/Kconfig
> @@ -84,6 +84,10 @@ config GENERIC_IRQ_IPI
>  	bool
>  	select IRQ_DOMAIN_HIERARCHY
>  
> +# Generic IRQ IPI Mux support
> +config GENERIC_IRQ_IPI_MUX
> +	bool
> +
>  # Generic MSI interrupt support
>  config GENERIC_MSI_IRQ
>  	bool
> diff --git a/kernel/irq/Makefile b/kernel/irq/Makefile
> index b4f53717d143..f19d3080bf11 100644
> --- a/kernel/irq/Makefile
> +++ b/kernel/irq/Makefile
> @@ -15,6 +15,7 @@ obj-$(CONFIG_GENERIC_IRQ_MIGRATION) += cpuhotplug.o
>  obj-$(CONFIG_PM_SLEEP) += pm.o
>  obj-$(CONFIG_GENERIC_MSI_IRQ) += msi.o
>  obj-$(CONFIG_GENERIC_IRQ_IPI) += ipi.o
> +obj-$(CONFIG_GENERIC_IRQ_IPI_MUX) += ipi-mux.o
>  obj-$(CONFIG_SMP) += affinity.o
>  obj-$(CONFIG_GENERIC_IRQ_DEBUGFS) += debugfs.o
>  obj-$(CONFIG_GENERIC_IRQ_MATRIX_ALLOCATOR) += matrix.o
> diff --git a/kernel/irq/ipi-mux.c b/kernel/irq/ipi-mux.c
> new file mode 100644
> index 000000000000..1a1fcfe3ac54
> --- /dev/null
> +++ b/kernel/irq/ipi-mux.c
> @@ -0,0 +1,197 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +/*
> + * Multiplex several virtual IPIs over a single HW IPI.
> + *
> + * Copyright (c) 2022 Ventana Micro Systems Inc.
> + */
> +
> +#define pr_fmt(fmt) "ipi-mux: " fmt
> +#include <linux/cpu.h>
> +#include <linux/init.h>
> +#include <linux/irq.h>
> +#include <linux/irqchip.h>
> +#include <linux/irqchip/chained_irq.h>
> +#include <linux/irqdomain.h>
> +#include <linux/smp.h>
> +
> +static unsigned int ipi_mux_parent_virq;
> +static struct irq_domain *ipi_mux_domain;
> +static const struct  ipi_mux_ops *ipi_mux_ops;
> +static DEFINE_PER_CPU(unsigned long, ipi_mux_bits);
> +
> +static void ipi_mux_dummy(struct irq_data *d)
> +{
> +}
> +
> +static void ipi_mux_send_mask(struct irq_data *d, const struct cpumask *mask)
> +{
> +	int cpu;
> +
> +	/* Barrier before doing atomic bit update to IPI bits */
> +	smp_mb__before_atomic();
> +
> +	for_each_cpu(cpu, mask)
> +		set_bit(d->hwirq, per_cpu_ptr(&ipi_mux_bits, cpu));
> +
> +	/* Barrier after doing atomic bit update to IPI bits */
> +	smp_mb__after_atomic();
> +
> +	/* Trigger the parent IPI */
> +	ipi_mux_ops->ipi_mux_send(ipi_mux_parent_virq, mask);
> +}
> +
> +static struct irq_chip ipi_mux_chip = {

const, please.

> +	.name		= "IPI Mux",
> +	.irq_mask	= ipi_mux_dummy,
> +	.irq_unmask	= ipi_mux_dummy,

Maybe we should consider a flow that does not require this dummy
callbacks.

> +	.ipi_send_mask	= ipi_mux_send_mask,
> +};
> +
> +static int ipi_mux_domain_map(struct irq_domain *d, unsigned int irq,
> +			      irq_hw_number_t hwirq)
> +{
> +	irq_set_percpu_devid(irq);
> +	irq_domain_set_info(d, irq, hwirq, &ipi_mux_chip, d->host_data,
> +			    handle_percpu_devid_irq, NULL, NULL);
> +
> +	return 0;
> +}
> +
> +static int ipi_mux_domain_alloc(struct irq_domain *d, unsigned int virq,
> +				unsigned int nr_irqs, void *arg)
> +{
> +	unsigned int type = IRQ_TYPE_NONE;

Really, this should be EDGE.

> +	struct irq_fwspec *fwspec = arg;
> +	irq_hw_number_t hwirq;
> +	int i, ret;
> +
> +	ret = irq_domain_translate_onecell(d, fwspec, &hwirq, &type);
> +	if (ret)
> +		return ret;
> +
> +	for (i = 0; i < nr_irqs; i++) {
> +		ret = ipi_mux_domain_map(d, virq + i, hwirq + i);
> +		if (ret)
> +			return ret;
> +	}
> +
> +	return 0;
> +}
> +
> +static const struct irq_domain_ops ipi_mux_domain_ops = {
> +	.translate	= irq_domain_translate_onecell,

What is the purpose of this callback? Firmware shouldn't be involved
in IPIs at all, and since you only have one cell, you can use the
default path. This is also a dependency on CONFIG_IRQ_DOMAIN_HIERARCHY.

> +	.alloc		= ipi_mux_domain_alloc,
> +	.free		= irq_domain_free_irqs_top,
> +};
> +
> +/**
> + * ipi_mux_process - Process multiplexed virtual IPIs
> + */
> +void ipi_mux_process(void)
> +{
> +	unsigned long irqs, *bits = this_cpu_ptr(&ipi_mux_bits);
> +	irq_hw_number_t hwirq;
> +	int err;
> +
> +	/* Clear the parent IPI */
> +	if (ipi_mux_ops->ipi_mux_clear)
> +		ipi_mux_ops->ipi_mux_clear(ipi_mux_parent_virq);
> +
> +	/*
> +	 * Barrier for IPI bits paired with smp_mb__xyz_atomic()

xyz???

> +	 * in ipi_mux_send_mask()
> +	 */
> +	smp_mb();
> +
> +	irqs = xchg(bits, 0);
> +	if (!irqs)
> +		return;
> +
> +	for_each_set_bit(hwirq, &irqs, IPI_MUX_NR_IRQS) {
> +		err = generic_handle_domain_irq(ipi_mux_domain,
> +						hwirq);
> +		if (unlikely(err))
> +			pr_warn_ratelimited(
> +				"can't find mapping for hwirq %lu\n",
> +				hwirq);
> +	}
> +}
> +
> +static void ipi_mux_handler(struct irq_desc *desc)
> +{
> +	struct irq_chip *chip = irq_desc_get_chip(desc);
> +
> +	chained_irq_enter(chip, desc);
> +	ipi_mux_process();
> +	chained_irq_exit(chip, desc);
> +}
> +
> +static int ipi_mux_dying_cpu(unsigned int cpu)
> +{
> +	disable_percpu_irq(ipi_mux_parent_virq);
> +	return 0;
> +}
> +
> +static int ipi_mux_starting_cpu(unsigned int cpu)
> +{
> +	enable_percpu_irq(ipi_mux_parent_virq,
> +			  irq_get_trigger_type(ipi_mux_parent_virq));
> +	return 0;
> +}
> +
> +/**
> + * ipi_mux_create - Create virtual IPIs (total IPI_MUX_NR_IRQS) multiplexed
> + * on top of a single parent IPI.
> + * @parent_virq:	virq of the parent IPI
> + * @ops:		multiplexing operations for the parent IPI
> + *
> + * If the parent IPI > 0 then ipi_mux_process() will be automatically
> + * called via chained handler.
> + *
> + * If the parent IPI <= 0 then it is responsiblity of irqchip drivers
> + * to explicitly call ipi_mux_process() for processing muxed IPIs.
> + *
> + * Returns first virq of the newly created virutal IPIs upon success
> + * or <=0 upon failure
> + */
> +int ipi_mux_create(unsigned int parent_virq, const struct ipi_mux_ops *ops)
> +{

Why should be parent_virq be unique? I also see nothing that checks
that this is a per-CPU interrupt. If anything, this needs documenting.

> +	struct irq_domain *domain;
> +	struct irq_fwspec ipi;
> +	int virq;
> +
> +	if (ipi_mux_domain || !ops || !ops->ipi_mux_send)
> +		return 0;
> +
> +	domain = irq_domain_add_linear(NULL, IPI_MUX_NR_IRQS,

Urgh. For a start, please use the irq_domain_create_* version, as this
shouldn't be DT specific. Then, don't use a NULL fwnode, as this
results in a "default domain", which nobody sane should ever use
anymore. Also, defaulting to BITS_PER_LONG is a lot of interrupts for
not much (most archs use only a handful). You may want to consider
this being a parameter, and cap it at BITS_PER_LONG.

> +				       &ipi_mux_domain_ops, NULL);
> +	if (!domain) {
> +		pr_err("unable to add IPI Mux domain\n");
> +		return 0;
> +	}
> +
> +	ipi.fwnode = domain->fwnode;

Which is NULL (see above).

> +	ipi.param_count = 1;
> +	ipi.param[0] = 0;
> +	virq = __irq_domain_alloc_irqs(domain, -1, IPI_MUX_NR_IRQS,
> +				       NUMA_NO_NODE, &ipi, false, NULL);
> +	if (virq <= 0) {
> +		pr_err("unable to alloc IRQs from IPI Mux domain\n");
> +		irq_domain_remove(domain);
> +		return virq;
> +	}
> +
> +	ipi_mux_domain = domain;
> +	ipi_mux_parent_virq = parent_virq;
> +	ipi_mux_ops = ops;
> +
> +	if (parent_virq > 0) {
> +		irq_set_chained_handler(parent_virq, ipi_mux_handler);
> +
> +		cpuhp_setup_state(CPUHP_AP_ONLINE_DYN,
> +				  "irqchip/ipi-mux:starting",
> +				  ipi_mux_starting_cpu, ipi_mux_dying_cpu);
> +	}
> +
> +	return virq;
> +}

Thanks,

	M.

-- 
Without deviation from the norm, progress is not possible.

Powered by blists - more mailing lists