[v2] Linux RISC-V AIA Support

[PATCH v2 5/9] irqchip: Add RISC-V incoming MSI controller driver

Posted by Anup Patel 2 years, 8 months ago

The RISC-V advanced interrupt architecture (AIA) specification defines
a new MSI controller for managing MSIs on a RISC-V platform. This new
MSI controller is referred to as incoming message signaled interrupt
controller (IMSIC) which manages MSI on per-HART (or per-CPU) basis.
(For more details refer https://github.com/riscv/riscv-aia)

This patch adds an irqchip driver for RISC-V IMSIC found on RISC-V
platforms.

Signed-off-by: Anup Patel <apatel@ventanamicro.com>
---
 drivers/irqchip/Kconfig             |   14 +-
 drivers/irqchip/Makefile            |    1 +
 drivers/irqchip/irq-riscv-imsic.c   | 1174 +++++++++++++++++++++++++++
 include/linux/irqchip/riscv-imsic.h |   92 +++
 4 files changed, 1280 insertions(+), 1 deletion(-)
 create mode 100644 drivers/irqchip/irq-riscv-imsic.c
 create mode 100644 include/linux/irqchip/riscv-imsic.h

diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig
index 9e65345ca3f6..a1315189a595 100644
--- a/drivers/irqchip/Kconfig
+++ b/drivers/irqchip/Kconfig
@@ -29,7 +29,6 @@ config ARM_GIC_V2M
 
 config GIC_NON_BANKED
 	bool
-
 config ARM_GIC_V3
 	bool
 	select IRQ_DOMAIN_HIERARCHY
@@ -548,6 +547,19 @@ config SIFIVE_PLIC
 	select IRQ_DOMAIN_HIERARCHY
 	select GENERIC_IRQ_EFFECTIVE_AFF_MASK if SMP
 
+config RISCV_IMSIC
+	bool
+	depends on RISCV
+	select IRQ_DOMAIN_HIERARCHY
+	select GENERIC_MSI_IRQ_DOMAIN
+
+config RISCV_IMSIC_PCI
+	bool
+	depends on RISCV_IMSIC
+	depends on PCI
+	depends on PCI_MSI
+	default RISCV_IMSIC
+
 config EXYNOS_IRQ_COMBINER
 	bool "Samsung Exynos IRQ combiner support" if COMPILE_TEST
 	depends on (ARCH_EXYNOS && ARM) || COMPILE_TEST
diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile
index 87b49a10962c..22c723cc6ec8 100644
--- a/drivers/irqchip/Makefile
+++ b/drivers/irqchip/Makefile
@@ -96,6 +96,7 @@ obj-$(CONFIG_QCOM_MPM)			+= irq-qcom-mpm.o
 obj-$(CONFIG_CSKY_MPINTC)		+= irq-csky-mpintc.o
 obj-$(CONFIG_CSKY_APB_INTC)		+= irq-csky-apb-intc.o
 obj-$(CONFIG_RISCV_INTC)		+= irq-riscv-intc.o
+obj-$(CONFIG_RISCV_IMSIC)		+= irq-riscv-imsic.o
 obj-$(CONFIG_SIFIVE_PLIC)		+= irq-sifive-plic.o
 obj-$(CONFIG_IMX_IRQSTEER)		+= irq-imx-irqsteer.o
 obj-$(CONFIG_IMX_INTMUX)		+= irq-imx-intmux.o
diff --git a/drivers/irqchip/irq-riscv-imsic.c b/drivers/irqchip/irq-riscv-imsic.c
new file mode 100644
index 000000000000..4c16b66738d6
--- /dev/null
+++ b/drivers/irqchip/irq-riscv-imsic.c
@@ -0,0 +1,1174 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2021 Western Digital Corporation or its affiliates.
+ * Copyright (C) 2022 Ventana Micro Systems Inc.
+ */
+
+#define pr_fmt(fmt) "riscv-imsic: " fmt
+#include <linux/bitmap.h>
+#include <linux/cpu.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/iommu.h>
+#include <linux/irq.h>
+#include <linux/irqchip.h>
+#include <linux/irqchip/chained_irq.h>
+#include <linux/irqchip/riscv-imsic.h>
+#include <linux/irqdomain.h>
+#include <linux/module.h>
+#include <linux/msi.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+#include <linux/pci.h>
+#include <linux/platform_device.h>
+#include <linux/spinlock.h>
+#include <linux/smp.h>
+#include <asm/hwcap.h>
+
+#define IMSIC_DISABLE_EIDELIVERY	0
+#define IMSIC_ENABLE_EIDELIVERY		1
+#define IMSIC_DISABLE_EITHRESHOLD	1
+#define IMSIC_ENABLE_EITHRESHOLD	0
+
+#define imsic_csr_write(__c, __v)	\
+do {					\
+	csr_write(CSR_ISELECT, __c);	\
+	csr_write(CSR_IREG, __v);	\
+} while (0)
+
+#define imsic_csr_read(__c)		\
+({					\
+	unsigned long __v;		\
+	csr_write(CSR_ISELECT, __c);	\
+	__v = csr_read(CSR_IREG);	\
+	__v;				\
+})
+
+#define imsic_csr_set(__c, __v)		\
+do {					\
+	csr_write(CSR_ISELECT, __c);	\
+	csr_set(CSR_IREG, __v);		\
+} while (0)
+
+#define imsic_csr_clear(__c, __v)	\
+do {					\
+	csr_write(CSR_ISELECT, __c);	\
+	csr_clear(CSR_IREG, __v);	\
+} while (0)
+
+struct imsic_mmio {
+	phys_addr_t pa;
+	void __iomem *va;
+	unsigned long size;
+};
+
+struct imsic_priv {
+	/* Global configuration common for all HARTs */
+	struct imsic_global_config global;
+
+	/* MMIO regions */
+	u32 num_mmios;
+	struct imsic_mmio *mmios;
+
+	/* Global state of interrupt identities */
+	raw_spinlock_t ids_lock;
+	unsigned long *ids_used_bimap;
+	unsigned long *ids_enabled_bimap;
+	unsigned int *ids_target_cpu;
+
+	/* Mask for connected CPUs */
+	struct cpumask lmask;
+
+	/* IPI interrupt identity */
+	u32 ipi_id;
+	u32 ipi_lsync_id;
+
+	/* IRQ domains */
+	struct irq_domain *base_domain;
+	struct irq_domain *pci_domain;
+	struct irq_domain *plat_domain;
+};
+
+struct imsic_handler {
+	/* Local configuration for given HART */
+	struct imsic_local_config local;
+
+	/* Pointer to private context */
+	struct imsic_priv *priv;
+};
+
+static bool imsic_init_done;
+
+static int imsic_parent_irq;
+static DEFINE_PER_CPU(struct imsic_handler, imsic_handlers);
+
+const struct imsic_global_config *imsic_get_global_config(void)
+{
+	struct imsic_handler *handler = this_cpu_ptr(&imsic_handlers);
+
+	if (!handler || !handler->priv)
+		return NULL;
+
+	return &handler->priv->global;
+}
+EXPORT_SYMBOL_GPL(imsic_get_global_config);
+
+const struct imsic_local_config *imsic_get_local_config(unsigned int cpu)
+{
+	struct imsic_handler *handler = per_cpu_ptr(&imsic_handlers, cpu);
+
+	if (!handler || !handler->priv)
+		return NULL;
+
+	return &handler->local;
+}
+EXPORT_SYMBOL_GPL(imsic_get_local_config);
+
+static int imsic_cpu_page_phys(unsigned int cpu,
+			       unsigned int guest_index,
+			       phys_addr_t *out_msi_pa)
+{
+	struct imsic_handler *handler = per_cpu_ptr(&imsic_handlers, cpu);
+	struct imsic_global_config *global;
+	struct imsic_local_config *local;
+
+	if (!handler || !handler->priv)
+		return -ENODEV;
+	local = &handler->local;
+	global = &handler->priv->global;
+
+	if (BIT(global->guest_index_bits) <= guest_index)
+		return -EINVAL;
+
+	if (out_msi_pa)
+		*out_msi_pa = local->msi_pa +
+			      (guest_index * IMSIC_MMIO_PAGE_SZ);
+
+	return 0;
+}
+
+static int imsic_get_cpu(struct imsic_priv *priv,
+			 const struct cpumask *mask_val, bool force,
+			 unsigned int *out_target_cpu)
+{
+	struct cpumask amask;
+	unsigned int cpu;
+
+	cpumask_and(&amask, &priv->lmask, mask_val);
+
+	if (force)
+		cpu = cpumask_first(&amask);
+	else
+		cpu = cpumask_any_and(&amask, cpu_online_mask);
+
+	if (cpu >= nr_cpu_ids)
+		return -EINVAL;
+
+	if (out_target_cpu)
+		*out_target_cpu = cpu;
+
+	return 0;
+}
+
+static int imsic_get_cpu_msi_msg(unsigned int cpu, unsigned int id,
+				 struct msi_msg *msg)
+{
+	phys_addr_t msi_addr;
+	int err;
+
+	err = imsic_cpu_page_phys(cpu, 0, &msi_addr);
+	if (err)
+		return err;
+
+	msg->address_hi = upper_32_bits(msi_addr);
+	msg->address_lo = lower_32_bits(msi_addr);
+	msg->data = id;
+
+	return err;
+}
+
+static void imsic_id_set_target(struct imsic_priv *priv,
+				 unsigned int id, unsigned int target_cpu)
+{
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&priv->ids_lock, flags);
+	priv->ids_target_cpu[id] = target_cpu;
+	raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
+}
+
+static unsigned int imsic_id_get_target(struct imsic_priv *priv,
+					unsigned int id)
+{
+	unsigned int ret;
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&priv->ids_lock, flags);
+	ret = priv->ids_target_cpu[id];
+	raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
+
+	return ret;
+}
+
+static void __imsic_eix_update(unsigned long base_id,
+			       unsigned long num_id, bool pend, bool val)
+{
+	unsigned long i, isel, ireg, flags;
+	unsigned long id = base_id, last_id = base_id + num_id;
+
+	while (id < last_id) {
+		isel = id / BITS_PER_LONG;
+		isel *= BITS_PER_LONG / IMSIC_EIPx_BITS;
+		isel += (pend) ? IMSIC_EIP0 : IMSIC_EIE0;
+
+		ireg = 0;
+		for (i = id & (__riscv_xlen - 1);
+		     (id < last_id) && (i < __riscv_xlen); i++) {
+			ireg |= BIT(i);
+			id++;
+		}
+
+		/*
+		 * The IMSIC EIEx and EIPx registers are indirectly
+		 * accessed via using ISELECT and IREG CSRs so we
+		 * save/restore local IRQ to ensure that we don't
+		 * get preempted while accessing IMSIC registers.
+		 */
+		local_irq_save(flags);
+		if (val)
+			imsic_csr_set(isel, ireg);
+		else
+			imsic_csr_clear(isel, ireg);
+		local_irq_restore(flags);
+	}
+}
+
+#define __imsic_id_enable(__id)		\
+	__imsic_eix_update((__id), 1, false, true)
+#define __imsic_id_disable(__id)	\
+	__imsic_eix_update((__id), 1, false, false)
+
+#ifdef CONFIG_SMP
+static void __imsic_id_smp_sync(struct imsic_priv *priv)
+{
+	struct imsic_handler *handler;
+	struct cpumask amask;
+	int cpu;
+
+	cpumask_and(&amask, &priv->lmask, cpu_online_mask);
+	for_each_cpu(cpu, &amask) {
+		if (cpu == smp_processor_id())
+			continue;
+
+		handler = per_cpu_ptr(&imsic_handlers, cpu);
+		if (!handler || !handler->priv || !handler->local.msi_va) {
+			pr_warn("CPU%d: handler not initialized\n", cpu);
+			continue;
+		}
+
+		writel(handler->priv->ipi_lsync_id, handler->local.msi_va);
+	}
+}
+#else
+#define __imsic_id_smp_sync(__priv)
+#endif
+
+static void imsic_id_enable(struct imsic_priv *priv, unsigned int id)
+{
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&priv->ids_lock, flags);
+	bitmap_set(priv->ids_enabled_bimap, id, 1);
+	__imsic_id_enable(id);
+	raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
+
+	__imsic_id_smp_sync(priv);
+}
+
+static void imsic_id_disable(struct imsic_priv *priv, unsigned int id)
+{
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&priv->ids_lock, flags);
+	bitmap_clear(priv->ids_enabled_bimap, id, 1);
+	__imsic_id_disable(id);
+	raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
+
+	__imsic_id_smp_sync(priv);
+}
+
+static void imsic_ids_local_sync(struct imsic_priv *priv)
+{
+	int i;
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&priv->ids_lock, flags);
+	for (i = 1; i <= priv->global.nr_ids; i++) {
+		if (priv->ipi_id == i || priv->ipi_lsync_id == i)
+			continue;
+
+		if (test_bit(i, priv->ids_enabled_bimap))
+			__imsic_id_enable(i);
+		else
+			__imsic_id_disable(i);
+	}
+	raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
+}
+
+static void imsic_ids_local_delivery(struct imsic_priv *priv, bool enable)
+{
+	if (enable) {
+		imsic_csr_write(IMSIC_EITHRESHOLD, IMSIC_ENABLE_EITHRESHOLD);
+		imsic_csr_write(IMSIC_EIDELIVERY, IMSIC_ENABLE_EIDELIVERY);
+	} else {
+		imsic_csr_write(IMSIC_EIDELIVERY, IMSIC_DISABLE_EIDELIVERY);
+		imsic_csr_write(IMSIC_EITHRESHOLD, IMSIC_DISABLE_EITHRESHOLD);
+	}
+}
+
+static int imsic_ids_alloc(struct imsic_priv *priv,
+			   unsigned int max_id, unsigned int order)
+{
+	int ret;
+	unsigned long flags;
+
+	if ((priv->global.nr_ids < max_id) ||
+	    (max_id < BIT(order)))
+		return -EINVAL;
+
+	raw_spin_lock_irqsave(&priv->ids_lock, flags);
+	ret = bitmap_find_free_region(priv->ids_used_bimap,
+				      max_id + 1, order);
+	raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
+
+	return ret;
+}
+
+static void imsic_ids_free(struct imsic_priv *priv, unsigned int base_id,
+			   unsigned int order)
+{
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&priv->ids_lock, flags);
+	bitmap_release_region(priv->ids_used_bimap, base_id, order);
+	raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
+}
+
+static int __init imsic_ids_init(struct imsic_priv *priv)
+{
+	int i;
+	struct imsic_global_config *global = &priv->global;
+
+	raw_spin_lock_init(&priv->ids_lock);
+
+	/* Allocate used bitmap */
+	priv->ids_used_bimap = kcalloc(BITS_TO_LONGS(global->nr_ids + 1),
+					sizeof(unsigned long), GFP_KERNEL);
+	if (!priv->ids_used_bimap)
+		return -ENOMEM;
+
+	/* Allocate enabled bitmap */
+	priv->ids_enabled_bimap = kcalloc(BITS_TO_LONGS(global->nr_ids + 1),
+					   sizeof(unsigned long), GFP_KERNEL);
+	if (!priv->ids_enabled_bimap) {
+		kfree(priv->ids_used_bimap);
+		return -ENOMEM;
+	}
+
+	/* Allocate target CPU array */
+	priv->ids_target_cpu = kcalloc(global->nr_ids + 1,
+				       sizeof(unsigned int), GFP_KERNEL);
+	if (!priv->ids_target_cpu) {
+		kfree(priv->ids_enabled_bimap);
+		kfree(priv->ids_used_bimap);
+		return -ENOMEM;
+	}
+	for (i = 0; i <= global->nr_ids; i++)
+		priv->ids_target_cpu[i] = UINT_MAX;
+
+	/* Reserve ID#0 because it is special and never implemented */
+	bitmap_set(priv->ids_used_bimap, 0, 1);
+
+	return 0;
+}
+
+static void __init imsic_ids_cleanup(struct imsic_priv *priv)
+{
+	kfree(priv->ids_target_cpu);
+	kfree(priv->ids_enabled_bimap);
+	kfree(priv->ids_used_bimap);
+}
+
+#ifdef CONFIG_SMP
+static void imsic_ipi_send(unsigned int cpu)
+{
+	struct imsic_handler *handler = per_cpu_ptr(&imsic_handlers, cpu);
+
+	if (!handler || !handler->priv || !handler->local.msi_va) {
+		pr_warn("CPU%d: handler not initialized\n", cpu);
+		return;
+	}
+
+	writel(handler->priv->ipi_id, handler->local.msi_va);
+}
+
+static void imsic_ipi_enable(struct imsic_priv *priv)
+{
+	__imsic_id_enable(priv->ipi_id);
+	__imsic_id_enable(priv->ipi_lsync_id);
+}
+
+static int __init imsic_ipi_domain_init(struct imsic_priv *priv)
+{
+	int virq;
+
+	/* Allocate interrupt identity for IPIs */
+	virq = imsic_ids_alloc(priv, priv->global.nr_ids, get_count_order(1));
+	if (virq < 0)
+		return virq;
+	priv->ipi_id = virq;
+
+	/* Create IMSIC IPI multiplexing */
+	virq = ipi_mux_create(BITS_PER_BYTE, imsic_ipi_send);
+	if (virq <= 0) {
+		imsic_ids_free(priv, priv->ipi_id, get_count_order(1));
+		return (virq < 0) ? virq : -ENOMEM;
+	}
+
+	/* Set vIRQ range */
+	riscv_ipi_set_virq_range(virq, BITS_PER_BYTE, true);
+
+	/* Allocate interrupt identity for local enable/disable sync */
+	virq = imsic_ids_alloc(priv, priv->global.nr_ids, get_count_order(1));
+	if (virq < 0) {
+		imsic_ids_free(priv, priv->ipi_id, get_count_order(1));
+		return virq;
+	}
+	priv->ipi_lsync_id = virq;
+
+	return 0;
+}
+
+static void __init imsic_ipi_domain_cleanup(struct imsic_priv *priv)
+{
+	imsic_ids_free(priv, priv->ipi_lsync_id, get_count_order(1));
+	if (priv->ipi_id)
+		imsic_ids_free(priv, priv->ipi_id, get_count_order(1));
+}
+#else
+static void imsic_ipi_enable(struct imsic_priv *priv)
+{
+}
+
+static int __init imsic_ipi_domain_init(struct imsic_priv *priv)
+{
+	/* Clear the IPI ids because we are not using IPIs */
+	priv->ipi_id = 0;
+	priv->ipi_lsync_id = 0;
+	return 0;
+}
+
+static void __init imsic_ipi_domain_cleanup(struct imsic_priv *priv)
+{
+}
+#endif
+
+static void imsic_irq_mask(struct irq_data *d)
+{
+	imsic_id_disable(irq_data_get_irq_chip_data(d), d->hwirq);
+}
+
+static void imsic_irq_unmask(struct irq_data *d)
+{
+	imsic_id_enable(irq_data_get_irq_chip_data(d), d->hwirq);
+}
+
+static void imsic_irq_compose_msi_msg(struct irq_data *d,
+				      struct msi_msg *msg)
+{
+	struct imsic_priv *priv = irq_data_get_irq_chip_data(d);
+	unsigned int cpu;
+	int err;
+
+	cpu = imsic_id_get_target(priv, d->hwirq);
+	WARN_ON(cpu == UINT_MAX);
+
+	err = imsic_get_cpu_msi_msg(cpu, d->hwirq, msg);
+	WARN_ON(err);
+
+	iommu_dma_compose_msi_msg(irq_data_get_msi_desc(d), msg);
+}
+
+#ifdef CONFIG_SMP
+static int imsic_irq_set_affinity(struct irq_data *d,
+				  const struct cpumask *mask_val,
+				  bool force)
+{
+	struct imsic_priv *priv = irq_data_get_irq_chip_data(d);
+	unsigned int target_cpu;
+	int rc;
+
+	rc = imsic_get_cpu(priv, mask_val, force, &target_cpu);
+	if (rc)
+		return rc;
+
+	imsic_id_set_target(priv, d->hwirq, target_cpu);
+	irq_data_update_effective_affinity(d, cpumask_of(target_cpu));
+
+	return IRQ_SET_MASK_OK;
+}
+#endif
+
+static struct irq_chip imsic_irq_base_chip = {
+	.name			= "RISC-V IMSIC-BASE",
+	.irq_mask		= imsic_irq_mask,
+	.irq_unmask		= imsic_irq_unmask,
+#ifdef CONFIG_SMP
+	.irq_set_affinity	= imsic_irq_set_affinity,
+#endif
+	.irq_compose_msi_msg	= imsic_irq_compose_msi_msg,
+	.flags			= IRQCHIP_SKIP_SET_WAKE |
+				  IRQCHIP_MASK_ON_SUSPEND,
+};
+
+static int imsic_irq_domain_alloc(struct irq_domain *domain,
+				  unsigned int virq,
+				  unsigned int nr_irqs,
+				  void *args)
+{
+	struct imsic_priv *priv = domain->host_data;
+	msi_alloc_info_t *info = args;
+	phys_addr_t msi_addr;
+	int i, hwirq, err = 0;
+	unsigned int cpu;
+
+	err = imsic_get_cpu(priv, &priv->lmask, false, &cpu);
+	if (err)
+		return err;
+
+	err = imsic_cpu_page_phys(cpu, 0, &msi_addr);
+	if (err)
+		return err;
+
+	hwirq = imsic_ids_alloc(priv, priv->global.nr_ids,
+				get_count_order(nr_irqs));
+	if (hwirq < 0)
+		return hwirq;
+
+	err = iommu_dma_prepare_msi(info->desc, msi_addr);
+	if (err)
+		goto fail;
+
+	for (i = 0; i < nr_irqs; i++) {
+		imsic_id_set_target(priv, hwirq + i, cpu);
+		irq_domain_set_info(domain, virq + i, hwirq + i,
+				    &imsic_irq_base_chip, priv,
+				    handle_simple_irq, NULL, NULL);
+		irq_set_noprobe(virq + i);
+		irq_set_affinity(virq + i, &priv->lmask);
+	}
+
+	return 0;
+
+fail:
+	imsic_ids_free(priv, hwirq, get_count_order(nr_irqs));
+	return err;
+}
+
+static void imsic_irq_domain_free(struct irq_domain *domain,
+				  unsigned int virq,
+				  unsigned int nr_irqs)
+{
+	struct irq_data *d = irq_domain_get_irq_data(domain, virq);
+	struct imsic_priv *priv = domain->host_data;
+
+	imsic_ids_free(priv, d->hwirq, get_count_order(nr_irqs));
+	irq_domain_free_irqs_parent(domain, virq, nr_irqs);
+}
+
+static const struct irq_domain_ops imsic_base_domain_ops = {
+	.alloc		= imsic_irq_domain_alloc,
+	.free		= imsic_irq_domain_free,
+};
+
+#ifdef CONFIG_RISCV_IMSIC_PCI
+
+static void imsic_pci_mask_irq(struct irq_data *d)
+{
+	pci_msi_mask_irq(d);
+	irq_chip_mask_parent(d);
+}
+
+static void imsic_pci_unmask_irq(struct irq_data *d)
+{
+	pci_msi_unmask_irq(d);
+	irq_chip_unmask_parent(d);
+}
+
+static struct irq_chip imsic_pci_irq_chip = {
+	.name			= "RISC-V IMSIC-PCI",
+	.irq_mask		= imsic_pci_mask_irq,
+	.irq_unmask		= imsic_pci_unmask_irq,
+	.irq_eoi		= irq_chip_eoi_parent,
+};
+
+static struct msi_domain_ops imsic_pci_domain_ops = {
+};
+
+static struct msi_domain_info imsic_pci_domain_info = {
+	.flags	= (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS |
+		   MSI_FLAG_PCI_MSIX | MSI_FLAG_MULTI_PCI_MSI),
+	.ops	= &imsic_pci_domain_ops,
+	.chip	= &imsic_pci_irq_chip,
+};
+
+#endif
+
+static struct irq_chip imsic_plat_irq_chip = {
+	.name			= "RISC-V IMSIC-PLAT",
+};
+
+static struct msi_domain_ops imsic_plat_domain_ops = {
+};
+
+static struct msi_domain_info imsic_plat_domain_info = {
+	.flags	= (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS),
+	.ops	= &imsic_plat_domain_ops,
+	.chip	= &imsic_plat_irq_chip,
+};
+
+static int __init imsic_irq_domains_init(struct imsic_priv *priv,
+					 struct fwnode_handle *fwnode)
+{
+	/* Create Base IRQ domain */
+	priv->base_domain = irq_domain_create_tree(fwnode,
+						&imsic_base_domain_ops, priv);
+	if (!priv->base_domain) {
+		pr_err("Failed to create IMSIC base domain\n");
+		return -ENOMEM;
+	}
+	irq_domain_update_bus_token(priv->base_domain, DOMAIN_BUS_NEXUS);
+
+#ifdef CONFIG_RISCV_IMSIC_PCI
+	/* Create PCI MSI domain */
+	priv->pci_domain = pci_msi_create_irq_domain(fwnode,
+						&imsic_pci_domain_info,
+						priv->base_domain);
+	if (!priv->pci_domain) {
+		pr_err("Failed to create IMSIC PCI domain\n");
+		irq_domain_remove(priv->base_domain);
+		return -ENOMEM;
+	}
+#endif
+
+	/* Create Platform MSI domain */
+	priv->plat_domain = platform_msi_create_irq_domain(fwnode,
+						&imsic_plat_domain_info,
+						priv->base_domain);
+	if (!priv->plat_domain) {
+		pr_err("Failed to create IMSIC platform domain\n");
+		if (priv->pci_domain)
+			irq_domain_remove(priv->pci_domain);
+		irq_domain_remove(priv->base_domain);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+/*
+ * To handle an interrupt, we read the TOPEI CSR and write zero in one
+ * instruction. If TOPEI CSR is non-zero then we translate TOPEI.ID to
+ * Linux interrupt number and let Linux IRQ subsystem handle it.
+ */
+static void imsic_handle_irq(struct irq_desc *desc)
+{
+	struct imsic_handler *handler = this_cpu_ptr(&imsic_handlers);
+	struct irq_chip *chip = irq_desc_get_chip(desc);
+	struct imsic_priv *priv = handler->priv;
+	irq_hw_number_t hwirq;
+	int err;
+
+	WARN_ON_ONCE(!handler->priv);
+
+	chained_irq_enter(chip, desc);
+
+	while ((hwirq = csr_swap(CSR_TOPEI, 0))) {
+		hwirq = hwirq >> TOPEI_ID_SHIFT;
+
+		if (hwirq == priv->ipi_id) {
+#ifdef CONFIG_SMP
+			ipi_mux_process();
+#endif
+			continue;
+		} else if (hwirq == priv->ipi_lsync_id) {
+			imsic_ids_local_sync(priv);
+			continue;
+		}
+
+		err = generic_handle_domain_irq(priv->base_domain, hwirq);
+		if (unlikely(err))
+			pr_warn_ratelimited(
+				"hwirq %lu mapping not found\n", hwirq);
+	}
+
+	chained_irq_exit(chip, desc);
+}
+
+static int imsic_starting_cpu(unsigned int cpu)
+{
+	struct imsic_handler *handler = this_cpu_ptr(&imsic_handlers);
+	struct imsic_priv *priv = handler->priv;
+
+	/* Enable per-CPU parent interrupt */
+	if (imsic_parent_irq)
+		enable_percpu_irq(imsic_parent_irq,
+				  irq_get_trigger_type(imsic_parent_irq));
+	else
+		pr_warn("cpu%d: parent irq not available\n", cpu);
+
+	/* Enable IPIs */
+	imsic_ipi_enable(priv);
+
+	/*
+	 * Interrupts identities might have been enabled/disabled while
+	 * this CPU was not running so sync-up local enable/disable state.
+	 */
+	imsic_ids_local_sync(priv);
+
+	/* Locally enable interrupt delivery */
+	imsic_ids_local_delivery(priv, true);
+
+	return 0;
+}
+
+struct imsic_fwnode_ops {
+	u32 (*nr_parent_irq)(struct fwnode_handle *fwnode,
+			     void *fwopaque);
+	int (*parent_hartid)(struct fwnode_handle *fwnode,
+			     void *fwopaque, u32 index,
+			     unsigned long *out_hartid);
+	u32 (*nr_mmio)(struct fwnode_handle *fwnode, void *fwopaque);
+	int (*mmio_to_resource)(struct fwnode_handle *fwnode,
+				void *fwopaque, u32 index,
+				struct resource *res);
+	void __iomem *(*mmio_map)(struct fwnode_handle *fwnode,
+				  void *fwopaque, u32 index);
+	int (*read_u32)(struct fwnode_handle *fwnode,
+			void *fwopaque, const char *prop, u32 *out_val);
+	bool (*read_bool)(struct fwnode_handle *fwnode,
+			  void *fwopaque, const char *prop);
+};
+
+static int __init imsic_init(struct imsic_fwnode_ops *fwops,
+			     struct fwnode_handle *fwnode,
+			     void *fwopaque)
+{
+	struct resource res;
+	phys_addr_t base_addr;
+	int rc, nr_parent_irqs;
+	struct imsic_mmio *mmio;
+	struct imsic_priv *priv;
+	struct irq_domain *domain;
+	struct imsic_handler *handler;
+	struct imsic_global_config *global;
+	u32 i, tmp, nr_handlers = 0;
+
+	if (imsic_init_done) {
+		pr_err("%pfwP: already initialized hence ignoring\n",
+			fwnode);
+		return -ENODEV;
+	}
+
+	if (!riscv_isa_extension_available(NULL, SxAIA)) {
+		pr_err("%pfwP: AIA support not available\n", fwnode);
+		return -ENODEV;
+	}
+
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+	global = &priv->global;
+
+	/* Find number of parent interrupts */
+	nr_parent_irqs = fwops->nr_parent_irq(fwnode, fwopaque);
+	if (!nr_parent_irqs) {
+		pr_err("%pfwP: no parent irqs available\n", fwnode);
+		return -EINVAL;
+	}
+
+	/* Find number of guest index bits in MSI address */
+	rc = fwops->read_u32(fwnode, fwopaque, "riscv,guest-index-bits",
+			     &global->guest_index_bits);
+	if (rc)
+		global->guest_index_bits = 0;
+	tmp = BITS_PER_LONG - IMSIC_MMIO_PAGE_SHIFT;
+	if (tmp < global->guest_index_bits) {
+		pr_err("%pfwP: guest index bits too big\n", fwnode);
+		return -EINVAL;
+	}
+
+	/* Find number of HART index bits */
+	rc = fwops->read_u32(fwnode, fwopaque, "riscv,hart-index-bits",
+			     &global->hart_index_bits);
+	if (rc) {
+		/* Assume default value */
+		global->hart_index_bits = __fls(nr_parent_irqs);
+		if (BIT(global->hart_index_bits) < nr_parent_irqs)
+			global->hart_index_bits++;
+	}
+	tmp = BITS_PER_LONG - IMSIC_MMIO_PAGE_SHIFT -
+	      global->guest_index_bits;
+	if (tmp < global->hart_index_bits) {
+		pr_err("%pfwP: HART index bits too big\n", fwnode);
+		return -EINVAL;
+	}
+
+	/* Find number of group index bits */
+	rc = fwops->read_u32(fwnode, fwopaque, "riscv,group-index-bits",
+			     &global->group_index_bits);
+	if (rc)
+		global->group_index_bits = 0;
+	tmp = BITS_PER_LONG - IMSIC_MMIO_PAGE_SHIFT -
+	      global->guest_index_bits - global->hart_index_bits;
+	if (tmp < global->group_index_bits) {
+		pr_err("%pfwP: group index bits too big\n", fwnode);
+		return -EINVAL;
+	}
+
+	/*
+	 * Find first bit position of group index.
+	 * If not specified assumed the default APLIC-IMSIC configuration.
+	 */
+	rc = fwops->read_u32(fwnode, fwopaque, "riscv,group-index-shift",
+			     &global->group_index_shift);
+	if (rc)
+		global->group_index_shift = IMSIC_MMIO_PAGE_SHIFT * 2;
+	tmp = global->group_index_bits + global->group_index_shift - 1;
+	if (tmp >= BITS_PER_LONG) {
+		pr_err("%pfwP: group index shift too big\n", fwnode);
+		return -EINVAL;
+	}
+
+	/* Find number of interrupt identities */
+	rc = fwops->read_u32(fwnode, fwopaque, "riscv,num-ids",
+			     &global->nr_ids);
+	if (rc) {
+		pr_err("%pfwP: number of interrupt identities not found\n",
+			fwnode);
+		return rc;
+	}
+	if ((global->nr_ids < IMSIC_MIN_ID) ||
+	    (global->nr_ids >= IMSIC_MAX_ID) ||
+	    ((global->nr_ids & IMSIC_MIN_ID) != IMSIC_MIN_ID)) {
+		pr_err("%pfwP: invalid number of interrupt identities\n",
+			fwnode);
+		return -EINVAL;
+	}
+
+	/* Find number of guest interrupt identities */
+	if (fwops->read_u32(fwnode, fwopaque, "riscv,num-guest-ids",
+			    &global->nr_guest_ids))
+		global->nr_guest_ids = global->nr_ids;
+	if ((global->nr_guest_ids < IMSIC_MIN_ID) ||
+	    (global->nr_guest_ids >= IMSIC_MAX_ID) ||
+	    ((global->nr_guest_ids & IMSIC_MIN_ID) != IMSIC_MIN_ID)) {
+		pr_err("%pfwP: invalid number of guest interrupt identities\n",
+			fwnode);
+		return -EINVAL;
+	}
+
+	/* Compute base address */
+	rc = fwops->mmio_to_resource(fwnode, fwopaque, 0, &res);
+	if (rc) {
+		pr_err("%pfwP: first MMIO resource not found\n", fwnode);
+		return -EINVAL;
+	}
+	global->base_addr = res.start;
+	global->base_addr &= ~(BIT(global->guest_index_bits +
+				   global->hart_index_bits +
+				   IMSIC_MMIO_PAGE_SHIFT) - 1);
+	global->base_addr &= ~((BIT(global->group_index_bits) - 1) <<
+			       global->group_index_shift);
+
+	/* Find number of MMIO register sets */
+	priv->num_mmios = fwops->nr_mmio(fwnode, fwopaque);
+
+	/* Allocate MMIO register sets */
+	priv->mmios = kcalloc(priv->num_mmios, sizeof(*mmio), GFP_KERNEL);
+	if (!priv->mmios) {
+		rc = -ENOMEM;
+		goto out_free_priv;
+	}
+
+	/* Parse and map MMIO register sets */
+	for (i = 0; i < priv->num_mmios; i++) {
+		mmio = &priv->mmios[i];
+		rc = fwops->mmio_to_resource(fwnode, fwopaque, i, &res);
+		if (rc) {
+			pr_err("%pfwP: unable to parse MMIO regset %d\n",
+				fwnode, i);
+			goto out_iounmap;
+		}
+		mmio->pa = res.start;
+		mmio->size = res.end - res.start + 1;
+
+		base_addr = mmio->pa;
+		base_addr &= ~(BIT(global->guest_index_bits +
+				   global->hart_index_bits +
+				   IMSIC_MMIO_PAGE_SHIFT) - 1);
+		base_addr &= ~((BIT(global->group_index_bits) - 1) <<
+			       global->group_index_shift);
+		if (base_addr != global->base_addr) {
+			rc = -EINVAL;
+			pr_err("%pfwP: address mismatch for regset %d\n",
+				fwnode, i);
+			goto out_iounmap;
+		}
+
+		mmio->va = fwops->mmio_map(fwnode, fwopaque, i);
+		if (!mmio->va) {
+			rc = -EIO;
+			pr_err("%pfwP: unable to map MMIO regset %d\n",
+				fwnode, i);
+			goto out_iounmap;
+		}
+	}
+
+	/* Initialize interrupt identity management */
+	rc = imsic_ids_init(priv);
+	if (rc) {
+		pr_err("%pfwP: failed to initialize interrupt management\n",
+		       fwnode);
+		goto out_iounmap;
+	}
+
+	/* Configure handlers for target CPUs */
+	for (i = 0; i < nr_parent_irqs; i++) {
+		unsigned long reloff, hartid;
+		int j, cpu;
+
+		rc = fwops->parent_hartid(fwnode, fwopaque, i, &hartid);
+		if (rc) {
+			pr_warn("%pfwP: hart ID for parent irq%d not found\n",
+				fwnode, i);
+			continue;
+		}
+
+		cpu = riscv_hartid_to_cpuid(hartid);
+		if (cpu < 0) {
+			pr_warn("%pfwP: invalid cpuid for parent irq%d\n",
+				fwnode, i);
+			continue;
+		}
+
+		/* Find MMIO location of MSI page */
+		mmio = NULL;
+		reloff = i * BIT(global->guest_index_bits) *
+			 IMSIC_MMIO_PAGE_SZ;
+		for (j = 0; priv->num_mmios; j++) {
+			if (reloff < priv->mmios[j].size) {
+				mmio = &priv->mmios[j];
+				break;
+			}
+
+			/*
+			 * MMIO region size may not be aligned to
+			 * BIT(global->guest_index_bits) * IMSIC_MMIO_PAGE_SZ
+			 * if holes are present.
+			 */
+			reloff -= ALIGN(priv->mmios[j].size,
+			BIT(global->guest_index_bits) * IMSIC_MMIO_PAGE_SZ);
+		}
+		if (!mmio) {
+			pr_warn("%pfwP: MMIO not found for parent irq%d\n",
+				fwnode, i);
+			continue;
+		}
+
+		handler = per_cpu_ptr(&imsic_handlers, cpu);
+		if (handler->priv) {
+			pr_warn("%pfwP: CPU%d handler already configured.\n",
+				fwnode, cpu);
+			goto done;
+		}
+
+		cpumask_set_cpu(cpu, &priv->lmask);
+		handler->local.msi_pa = mmio->pa + reloff;
+		handler->local.msi_va = mmio->va + reloff;
+		handler->priv = priv;
+
+done:
+		nr_handlers++;
+	}
+
+	/* If no CPU handlers found then can't take interrupts */
+	if (!nr_handlers) {
+		pr_err("%pfwP: No CPU handlers found\n", fwnode);
+		rc = -ENODEV;
+		goto out_ids_cleanup;
+	}
+
+	/* Find parent domain and register chained handler */
+	domain = irq_find_matching_fwnode(riscv_get_intc_hwnode(),
+					  DOMAIN_BUS_ANY);
+	if (!domain) {
+		pr_err("%pfwP: Failed to find INTC domain\n", fwnode);
+		rc = -ENOENT;
+		goto out_ids_cleanup;
+	}
+	imsic_parent_irq = irq_create_mapping(domain, RV_IRQ_EXT);
+	if (!imsic_parent_irq) {
+		pr_err("%pfwP: Failed to create INTC mapping\n", fwnode);
+		rc = -ENOENT;
+		goto out_ids_cleanup;
+	}
+	irq_set_chained_handler(imsic_parent_irq, imsic_handle_irq);
+
+	/* Initialize IPI domain */
+	rc = imsic_ipi_domain_init(priv);
+	if (rc) {
+		pr_err("%pfwP: Failed to initialize IPI domain\n", fwnode);
+		goto out_ids_cleanup;
+	}
+
+	/* Initialize IRQ and MSI domains */
+	rc = imsic_irq_domains_init(priv, fwnode);
+	if (rc) {
+		pr_err("%pfwP: Failed to initialize IRQ and MSI domains\n",
+		       fwnode);
+		goto out_ipi_domain_cleanup;
+	}
+
+	/*
+	 * Setup cpuhp state
+	 *
+	 * Don't disable per-CPU IMSIC file when CPU goes offline
+	 * because this affects IPI and the masking/unmasking of
+	 * virtual IPIs is done via generic IPI-Mux
+	 */
+	cpuhp_setup_state(CPUHP_AP_ONLINE_DYN,
+			  "irqchip/riscv/imsic:starting",
+			  imsic_starting_cpu, NULL);
+
+	/*
+	 * Only one IMSIC instance allowed in a platform for clean
+	 * implementation of SMP IRQ affinity and per-CPU IPIs.
+	 *
+	 * This means on a multi-socket (or multi-die) platform we
+	 * will have multiple MMIO regions for one IMSIC instance.
+	 */
+	imsic_init_done = true;
+
+	pr_info("%pfwP:  hart-index-bits: %d,  guest-index-bits: %d\n",
+		fwnode, global->hart_index_bits, global->guest_index_bits);
+	pr_info("%pfwP: group-index-bits: %d, group-index-shift: %d\n",
+		fwnode, global->group_index_bits, global->group_index_shift);
+	pr_info("%pfwP: mapped %d interrupts for %d CPUs at %pa\n",
+		fwnode, global->nr_ids, nr_handlers, &global->base_addr);
+	if (priv->ipi_lsync_id)
+		pr_info("%pfwP: enable/disable sync using interrupt %d\n",
+			fwnode, priv->ipi_lsync_id);
+	if (priv->ipi_id)
+		pr_info("%pfwP: providing IPIs using interrupt %d\n",
+			fwnode, priv->ipi_id);
+
+	return 0;
+
+out_ipi_domain_cleanup:
+	imsic_ipi_domain_cleanup(priv);
+out_ids_cleanup:
+	imsic_ids_cleanup(priv);
+out_iounmap:
+	for (i = 0; i < priv->num_mmios; i++) {
+		if (priv->mmios[i].va)
+			iounmap(priv->mmios[i].va);
+	}
+	kfree(priv->mmios);
+out_free_priv:
+	kfree(priv);
+	return rc;
+}
+
+static u32 __init imsic_dt_nr_parent_irq(struct fwnode_handle *fwnode,
+					 void *fwopaque)
+{
+	return of_irq_count(to_of_node(fwnode));
+}
+
+static int __init imsic_dt_parent_hartid(struct fwnode_handle *fwnode,
+					 void *fwopaque, u32 index,
+					 unsigned long *out_hartid)
+{
+	struct of_phandle_args parent;
+	int rc;
+
+	rc = of_irq_parse_one(to_of_node(fwnode), index, &parent);
+	if (rc)
+		return rc;
+
+	/*
+	 * Skip interrupts other than external interrupts for
+	 * current privilege level.
+	 */
+	if (parent.args[0] != RV_IRQ_EXT)
+		return -EINVAL;
+
+	return riscv_of_parent_hartid(parent.np, out_hartid);
+}
+
+static u32 __init imsic_dt_nr_mmio(struct fwnode_handle *fwnode,
+				   void *fwopaque)
+{
+	u32 ret = 0;
+	struct resource res;
+
+	while (!of_address_to_resource(to_of_node(fwnode), ret, &res))
+		ret++;
+
+	return ret;
+}
+
+static int __init imsic_mmio_to_resource(struct fwnode_handle *fwnode,
+					 void *fwopaque, u32 index,
+					 struct resource *res)
+{
+	return of_address_to_resource(to_of_node(fwnode), index, res);
+}
+
+static void __iomem __init *imsic_dt_mmio_map(struct fwnode_handle *fwnode,
+					      void *fwopaque, u32 index)
+{
+	return of_iomap(to_of_node(fwnode), index);
+}
+
+static int __init imsic_dt_read_u32(struct fwnode_handle *fwnode,
+				    void *fwopaque, const char *prop,
+				    u32 *out_val)
+{
+	return of_property_read_u32(to_of_node(fwnode), prop, out_val);
+}
+
+static bool __init imsic_dt_read_bool(struct fwnode_handle *fwnode,
+				      void *fwopaque, const char *prop)
+{
+	return of_property_read_bool(to_of_node(fwnode), prop);
+}
+
+static int __init imsic_dt_init(struct device_node *node,
+				struct device_node *parent)
+{
+	struct imsic_fwnode_ops ops = {
+		.nr_parent_irq = imsic_dt_nr_parent_irq,
+		.parent_hartid = imsic_dt_parent_hartid,
+		.nr_mmio = imsic_dt_nr_mmio,
+		.mmio_to_resource = imsic_mmio_to_resource,
+		.mmio_map = imsic_dt_mmio_map,
+		.read_u32 = imsic_dt_read_u32,
+		.read_bool = imsic_dt_read_bool,
+	};
+
+	return imsic_init(&ops, &node->fwnode, NULL);
+}
+IRQCHIP_DECLARE(riscv_imsic, "riscv,imsics", imsic_dt_init);
diff --git a/include/linux/irqchip/riscv-imsic.h b/include/linux/irqchip/riscv-imsic.h
new file mode 100644
index 000000000000..5d1387adc0ba
--- /dev/null
+++ b/include/linux/irqchip/riscv-imsic.h
@@ -0,0 +1,92 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2021 Western Digital Corporation or its affiliates.
+ * Copyright (C) 2022 Ventana Micro Systems Inc.
+ */
+#ifndef __LINUX_IRQCHIP_RISCV_IMSIC_H
+#define __LINUX_IRQCHIP_RISCV_IMSIC_H
+
+#include <linux/types.h>
+#include <asm/csr.h>
+
+#define IMSIC_MMIO_PAGE_SHIFT		12
+#define IMSIC_MMIO_PAGE_SZ		(1UL << IMSIC_MMIO_PAGE_SHIFT)
+#define IMSIC_MMIO_PAGE_LE		0x00
+#define IMSIC_MMIO_PAGE_BE		0x04
+
+#define IMSIC_MIN_ID			63
+#define IMSIC_MAX_ID			2048
+
+#define IMSIC_EIDELIVERY		0x70
+
+#define IMSIC_EITHRESHOLD		0x72
+
+#define IMSIC_EIP0			0x80
+#define IMSIC_EIP63			0xbf
+#define IMSIC_EIPx_BITS			32
+
+#define IMSIC_EIE0			0xc0
+#define IMSIC_EIE63			0xff
+#define IMSIC_EIEx_BITS			32
+
+#define IMSIC_FIRST			IMSIC_EIDELIVERY
+#define IMSIC_LAST			IMSIC_EIE63
+
+#define IMSIC_MMIO_SETIPNUM_LE		0x00
+#define IMSIC_MMIO_SETIPNUM_BE		0x04
+
+struct imsic_global_config {
+	/*
+	 * MSI Target Address Scheme
+	 *
+	 * XLEN-1                                                12     0
+	 * |                                                     |     |
+	 * -------------------------------------------------------------
+	 * |xxxxxx|Group Index|xxxxxxxxxxx|HART Index|Guest Index|  0  |
+	 * -------------------------------------------------------------
+	 */
+
+	/* Bits representing Guest index, HART index, and Group index */
+	u32 guest_index_bits;
+	u32 hart_index_bits;
+	u32 group_index_bits;
+	u32 group_index_shift;
+
+	/* Global base address matching all target MSI addresses */
+	phys_addr_t base_addr;
+
+	/* Number of interrupt identities */
+	u32 nr_ids;
+
+	/* Number of guest interrupt identities */
+	u32 nr_guest_ids;
+};
+
+struct imsic_local_config {
+	phys_addr_t msi_pa;
+	void __iomem *msi_va;
+};
+
+#ifdef CONFIG_RISCV_IMSIC
+
+extern const struct imsic_global_config *imsic_get_global_config(void);
+
+extern const struct imsic_local_config *imsic_get_local_config(
+							unsigned int cpu);
+
+#else
+
+static inline const struct imsic_global_config *imsic_get_global_config(void)
+{
+	return NULL;
+}
+
+static inline const struct imsic_local_config *imsic_get_local_config(
+							unsigned int cpu)
+{
+	return NULL;
+}
+
+#endif
+
+#endif
-- 
2.34.1

Re: [PATCH v2 5/9] irqchip: Add RISC-V incoming MSI controller driver

Posted by Marc Zyngier 2 years, 8 months ago

On Tue, 03 Jan 2023 14:14:05 +0000,
Anup Patel <apatel@ventanamicro.com> wrote:
> 
> The RISC-V advanced interrupt architecture (AIA) specification defines
> a new MSI controller for managing MSIs on a RISC-V platform. This new
> MSI controller is referred to as incoming message signaled interrupt
> controller (IMSIC) which manages MSI on per-HART (or per-CPU) basis.
> (For more details refer https://github.com/riscv/riscv-aia)

And how about IPIs, which this driver seems to be concerned about?

> 
> This patch adds an irqchip driver for RISC-V IMSIC found on RISC-V
> platforms.
> 
> Signed-off-by: Anup Patel <apatel@ventanamicro.com>
> ---
>  drivers/irqchip/Kconfig             |   14 +-
>  drivers/irqchip/Makefile            |    1 +
>  drivers/irqchip/irq-riscv-imsic.c   | 1174 +++++++++++++++++++++++++++
>  include/linux/irqchip/riscv-imsic.h |   92 +++
>  4 files changed, 1280 insertions(+), 1 deletion(-)
>  create mode 100644 drivers/irqchip/irq-riscv-imsic.c
>  create mode 100644 include/linux/irqchip/riscv-imsic.h
> 
> diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig
> index 9e65345ca3f6..a1315189a595 100644
> --- a/drivers/irqchip/Kconfig
> +++ b/drivers/irqchip/Kconfig
> @@ -29,7 +29,6 @@ config ARM_GIC_V2M
>  
>  config GIC_NON_BANKED
>  	bool
> -
>  config ARM_GIC_V3
>  	bool
>  	select IRQ_DOMAIN_HIERARCHY
> @@ -548,6 +547,19 @@ config SIFIVE_PLIC
>  	select IRQ_DOMAIN_HIERARCHY
>  	select GENERIC_IRQ_EFFECTIVE_AFF_MASK if SMP
>  
> +config RISCV_IMSIC
> +	bool
> +	depends on RISCV
> +	select IRQ_DOMAIN_HIERARCHY
> +	select GENERIC_MSI_IRQ_DOMAIN
> +
> +config RISCV_IMSIC_PCI
> +	bool
> +	depends on RISCV_IMSIC
> +	depends on PCI
> +	depends on PCI_MSI
> +	default RISCV_IMSIC

This should definitely tell you that this driver needs splitting.

> +
>  config EXYNOS_IRQ_COMBINER
>  	bool "Samsung Exynos IRQ combiner support" if COMPILE_TEST
>  	depends on (ARCH_EXYNOS && ARM) || COMPILE_TEST
> diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile
> index 87b49a10962c..22c723cc6ec8 100644
> --- a/drivers/irqchip/Makefile
> +++ b/drivers/irqchip/Makefile
> @@ -96,6 +96,7 @@ obj-$(CONFIG_QCOM_MPM)			+= irq-qcom-mpm.o
>  obj-$(CONFIG_CSKY_MPINTC)		+= irq-csky-mpintc.o
>  obj-$(CONFIG_CSKY_APB_INTC)		+= irq-csky-apb-intc.o
>  obj-$(CONFIG_RISCV_INTC)		+= irq-riscv-intc.o
> +obj-$(CONFIG_RISCV_IMSIC)		+= irq-riscv-imsic.o
>  obj-$(CONFIG_SIFIVE_PLIC)		+= irq-sifive-plic.o
>  obj-$(CONFIG_IMX_IRQSTEER)		+= irq-imx-irqsteer.o
>  obj-$(CONFIG_IMX_INTMUX)		+= irq-imx-intmux.o
> diff --git a/drivers/irqchip/irq-riscv-imsic.c b/drivers/irqchip/irq-riscv-imsic.c
> new file mode 100644
> index 000000000000..4c16b66738d6
> --- /dev/null
> +++ b/drivers/irqchip/irq-riscv-imsic.c
> @@ -0,0 +1,1174 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Copyright (C) 2021 Western Digital Corporation or its affiliates.
> + * Copyright (C) 2022 Ventana Micro Systems Inc.
> + */
> +
> +#define pr_fmt(fmt) "riscv-imsic: " fmt
> +#include <linux/bitmap.h>
> +#include <linux/cpu.h>
> +#include <linux/interrupt.h>
> +#include <linux/io.h>
> +#include <linux/iommu.h>
> +#include <linux/irq.h>
> +#include <linux/irqchip.h>
> +#include <linux/irqchip/chained_irq.h>
> +#include <linux/irqchip/riscv-imsic.h>
> +#include <linux/irqdomain.h>
> +#include <linux/module.h>
> +#include <linux/msi.h>
> +#include <linux/of.h>
> +#include <linux/of_address.h>
> +#include <linux/of_irq.h>
> +#include <linux/pci.h>
> +#include <linux/platform_device.h>
> +#include <linux/spinlock.h>
> +#include <linux/smp.h>
> +#include <asm/hwcap.h>
> +
> +#define IMSIC_DISABLE_EIDELIVERY	0
> +#define IMSIC_ENABLE_EIDELIVERY		1
> +#define IMSIC_DISABLE_EITHRESHOLD	1
> +#define IMSIC_ENABLE_EITHRESHOLD	0
> +
> +#define imsic_csr_write(__c, __v)	\
> +do {					\
> +	csr_write(CSR_ISELECT, __c);	\
> +	csr_write(CSR_IREG, __v);	\
> +} while (0)
> +
> +#define imsic_csr_read(__c)		\
> +({					\
> +	unsigned long __v;		\
> +	csr_write(CSR_ISELECT, __c);	\
> +	__v = csr_read(CSR_IREG);	\
> +	__v;				\
> +})
> +
> +#define imsic_csr_set(__c, __v)		\
> +do {					\
> +	csr_write(CSR_ISELECT, __c);	\
> +	csr_set(CSR_IREG, __v);		\
> +} while (0)
> +
> +#define imsic_csr_clear(__c, __v)	\
> +do {					\
> +	csr_write(CSR_ISELECT, __c);	\
> +	csr_clear(CSR_IREG, __v);	\
> +} while (0)
> +
> +struct imsic_mmio {
> +	phys_addr_t pa;
> +	void __iomem *va;
> +	unsigned long size;
> +};
> +
> +struct imsic_priv {
> +	/* Global configuration common for all HARTs */
> +	struct imsic_global_config global;
> +
> +	/* MMIO regions */
> +	u32 num_mmios;
> +	struct imsic_mmio *mmios;
> +
> +	/* Global state of interrupt identities */
> +	raw_spinlock_t ids_lock;
> +	unsigned long *ids_used_bimap;
> +	unsigned long *ids_enabled_bimap;
> +	unsigned int *ids_target_cpu;
> +
> +	/* Mask for connected CPUs */
> +	struct cpumask lmask;
> +
> +	/* IPI interrupt identity */
> +	u32 ipi_id;
> +	u32 ipi_lsync_id;
> +
> +	/* IRQ domains */
> +	struct irq_domain *base_domain;
> +	struct irq_domain *pci_domain;
> +	struct irq_domain *plat_domain;
> +};
> +
> +struct imsic_handler {
> +	/* Local configuration for given HART */
> +	struct imsic_local_config local;
> +
> +	/* Pointer to private context */
> +	struct imsic_priv *priv;
> +};
> +
> +static bool imsic_init_done;
> +
> +static int imsic_parent_irq;
> +static DEFINE_PER_CPU(struct imsic_handler, imsic_handlers);
> +
> +const struct imsic_global_config *imsic_get_global_config(void)
> +{
> +	struct imsic_handler *handler = this_cpu_ptr(&imsic_handlers);
> +
> +	if (!handler || !handler->priv)
> +		return NULL;
> +
> +	return &handler->priv->global;
> +}
> +EXPORT_SYMBOL_GPL(imsic_get_global_config);
> +
> +const struct imsic_local_config *imsic_get_local_config(unsigned int cpu)
> +{
> +	struct imsic_handler *handler = per_cpu_ptr(&imsic_handlers, cpu);
> +
> +	if (!handler || !handler->priv)
> +		return NULL;

How can this happen?

> +
> +	return &handler->local;
> +}
> +EXPORT_SYMBOL_GPL(imsic_get_local_config);

Why are these symbols exported? They have no user, so they shouldn't
even exist here. I also seriously doubt there is a valid use case for
exposing this information to the rest of the kernel.

> +
> +static int imsic_cpu_page_phys(unsigned int cpu,
> +			       unsigned int guest_index,
> +			       phys_addr_t *out_msi_pa)
> +{
> +	struct imsic_handler *handler = per_cpu_ptr(&imsic_handlers, cpu);
> +	struct imsic_global_config *global;
> +	struct imsic_local_config *local;
> +
> +	if (!handler || !handler->priv)
> +		return -ENODEV;
> +	local = &handler->local;
> +	global = &handler->priv->global;
> +
> +	if (BIT(global->guest_index_bits) <= guest_index)
> +		return -EINVAL;
> +
> +	if (out_msi_pa)
> +		*out_msi_pa = local->msi_pa +
> +			      (guest_index * IMSIC_MMIO_PAGE_SZ);
> +
> +	return 0;
> +}
> +
> +static int imsic_get_cpu(struct imsic_priv *priv,
> +			 const struct cpumask *mask_val, bool force,
> +			 unsigned int *out_target_cpu)
> +{
> +	struct cpumask amask;
> +	unsigned int cpu;
> +
> +	cpumask_and(&amask, &priv->lmask, mask_val);
> +
> +	if (force)
> +		cpu = cpumask_first(&amask);
> +	else
> +		cpu = cpumask_any_and(&amask, cpu_online_mask);
> +
> +	if (cpu >= nr_cpu_ids)
> +		return -EINVAL;
> +
> +	if (out_target_cpu)
> +		*out_target_cpu = cpu;
> +
> +	return 0;
> +}
> +
> +static int imsic_get_cpu_msi_msg(unsigned int cpu, unsigned int id,
> +				 struct msi_msg *msg)
> +{
> +	phys_addr_t msi_addr;
> +	int err;
> +
> +	err = imsic_cpu_page_phys(cpu, 0, &msi_addr);
> +	if (err)
> +		return err;
> +
> +	msg->address_hi = upper_32_bits(msi_addr);
> +	msg->address_lo = lower_32_bits(msi_addr);
> +	msg->data = id;
> +
> +	return err;
> +}
> +
> +static void imsic_id_set_target(struct imsic_priv *priv,
> +				 unsigned int id, unsigned int target_cpu)
> +{
> +	unsigned long flags;
> +
> +	raw_spin_lock_irqsave(&priv->ids_lock, flags);
> +	priv->ids_target_cpu[id] = target_cpu;
> +	raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
> +}
> +
> +static unsigned int imsic_id_get_target(struct imsic_priv *priv,
> +					unsigned int id)
> +{
> +	unsigned int ret;
> +	unsigned long flags;
> +
> +	raw_spin_lock_irqsave(&priv->ids_lock, flags);
> +	ret = priv->ids_target_cpu[id];
> +	raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
> +
> +	return ret;
> +}
> +
> +static void __imsic_eix_update(unsigned long base_id,
> +			       unsigned long num_id, bool pend, bool val)
> +{
> +	unsigned long i, isel, ireg, flags;
> +	unsigned long id = base_id, last_id = base_id + num_id;
> +
> +	while (id < last_id) {
> +		isel = id / BITS_PER_LONG;
> +		isel *= BITS_PER_LONG / IMSIC_EIPx_BITS;
> +		isel += (pend) ? IMSIC_EIP0 : IMSIC_EIE0;
> +
> +		ireg = 0;
> +		for (i = id & (__riscv_xlen - 1);
> +		     (id < last_id) && (i < __riscv_xlen); i++) {
> +			ireg |= BIT(i);
> +			id++;
> +		}
> +
> +		/*
> +		 * The IMSIC EIEx and EIPx registers are indirectly
> +		 * accessed via using ISELECT and IREG CSRs so we
> +		 * save/restore local IRQ to ensure that we don't
> +		 * get preempted while accessing IMSIC registers.
> +		 */
> +		local_irq_save(flags);
> +		if (val)
> +			imsic_csr_set(isel, ireg);
> +		else
> +			imsic_csr_clear(isel, ireg);
> +		local_irq_restore(flags);

What is the actual requirement? no preemption? or no interrupts? This
isn't the same thing. Also, a bunch of the users already disable
interrupts. Consistency wouldn't hurt here.

> +	}
> +}
> +
> +#define __imsic_id_enable(__id)		\
> +	__imsic_eix_update((__id), 1, false, true)
> +#define __imsic_id_disable(__id)	\
> +	__imsic_eix_update((__id), 1, false, false)
> +
> +#ifdef CONFIG_SMP
> +static void __imsic_id_smp_sync(struct imsic_priv *priv)
> +{
> +	struct imsic_handler *handler;
> +	struct cpumask amask;
> +	int cpu;
> +
> +	cpumask_and(&amask, &priv->lmask, cpu_online_mask);

Can't this race against a CPU going down?

> +	for_each_cpu(cpu, &amask) {
> +		if (cpu == smp_processor_id())
> +			continue;
> +
> +		handler = per_cpu_ptr(&imsic_handlers, cpu);
> +		if (!handler || !handler->priv || !handler->local.msi_va) {
> +			pr_warn("CPU%d: handler not initialized\n", cpu);

How many times are you going to do that? On each failing synchronisation?

> +			continue;
> +		}
> +
> +		writel(handler->priv->ipi_lsync_id, handler->local.msi_va);

As I understand it, this is a "behind the scenes" IPI. Why isn't that
a *real* IPI?

> +	}
> +}
> +#else
> +#define __imsic_id_smp_sync(__priv)
> +#endif
> +
> +static void imsic_id_enable(struct imsic_priv *priv, unsigned int id)
> +{
> +	unsigned long flags;
> +
> +	raw_spin_lock_irqsave(&priv->ids_lock, flags);
> +	bitmap_set(priv->ids_enabled_bimap, id, 1);
> +	__imsic_id_enable(id);
> +	raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
> +
> +	__imsic_id_smp_sync(priv);
> +}
> +
> +static void imsic_id_disable(struct imsic_priv *priv, unsigned int id)
> +{
> +	unsigned long flags;
> +
> +	raw_spin_lock_irqsave(&priv->ids_lock, flags);
> +	bitmap_clear(priv->ids_enabled_bimap, id, 1);
> +	__imsic_id_disable(id);
> +	raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
> +
> +	__imsic_id_smp_sync(priv);
> +}
> +
> +static void imsic_ids_local_sync(struct imsic_priv *priv)
> +{
> +	int i;
> +	unsigned long flags;
> +
> +	raw_spin_lock_irqsave(&priv->ids_lock, flags);
> +	for (i = 1; i <= priv->global.nr_ids; i++) {
> +		if (priv->ipi_id == i || priv->ipi_lsync_id == i)
> +			continue;
> +
> +		if (test_bit(i, priv->ids_enabled_bimap))
> +			__imsic_id_enable(i);
> +		else
> +			__imsic_id_disable(i);
> +	}
> +	raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
> +}
> +
> +static void imsic_ids_local_delivery(struct imsic_priv *priv, bool enable)
> +{
> +	if (enable) {
> +		imsic_csr_write(IMSIC_EITHRESHOLD, IMSIC_ENABLE_EITHRESHOLD);
> +		imsic_csr_write(IMSIC_EIDELIVERY, IMSIC_ENABLE_EIDELIVERY);
> +	} else {
> +		imsic_csr_write(IMSIC_EIDELIVERY, IMSIC_DISABLE_EIDELIVERY);
> +		imsic_csr_write(IMSIC_EITHRESHOLD, IMSIC_DISABLE_EITHRESHOLD);
> +	}
> +}
> +
> +static int imsic_ids_alloc(struct imsic_priv *priv,
> +			   unsigned int max_id, unsigned int order)
> +{
> +	int ret;
> +	unsigned long flags;
> +
> +	if ((priv->global.nr_ids < max_id) ||
> +	    (max_id < BIT(order)))
> +		return -EINVAL;

Why do we need this check? Shouldn't that be guaranteed by
construction?

> +
> +	raw_spin_lock_irqsave(&priv->ids_lock, flags);
> +	ret = bitmap_find_free_region(priv->ids_used_bimap,
> +				      max_id + 1, order);
> +	raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
> +
> +	return ret;
> +}
> +
> +static void imsic_ids_free(struct imsic_priv *priv, unsigned int base_id,
> +			   unsigned int order)
> +{
> +	unsigned long flags;
> +
> +	raw_spin_lock_irqsave(&priv->ids_lock, flags);
> +	bitmap_release_region(priv->ids_used_bimap, base_id, order);
> +	raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
> +}
> +
> +static int __init imsic_ids_init(struct imsic_priv *priv)
> +{
> +	int i;
> +	struct imsic_global_config *global = &priv->global;
> +
> +	raw_spin_lock_init(&priv->ids_lock);
> +
> +	/* Allocate used bitmap */
> +	priv->ids_used_bimap = kcalloc(BITS_TO_LONGS(global->nr_ids + 1),
> +					sizeof(unsigned long), GFP_KERNEL);

How about bitmap_alloc?

> +	if (!priv->ids_used_bimap)
> +		return -ENOMEM;
> +
> +	/* Allocate enabled bitmap */
> +	priv->ids_enabled_bimap = kcalloc(BITS_TO_LONGS(global->nr_ids + 1),
> +					   sizeof(unsigned long), GFP_KERNEL);
> +	if (!priv->ids_enabled_bimap) {
> +		kfree(priv->ids_used_bimap);
> +		return -ENOMEM;
> +	}
> +
> +	/* Allocate target CPU array */
> +	priv->ids_target_cpu = kcalloc(global->nr_ids + 1,
> +				       sizeof(unsigned int), GFP_KERNEL);
> +	if (!priv->ids_target_cpu) {
> +		kfree(priv->ids_enabled_bimap);
> +		kfree(priv->ids_used_bimap);
> +		return -ENOMEM;
> +	}
> +	for (i = 0; i <= global->nr_ids; i++)
> +		priv->ids_target_cpu[i] = UINT_MAX;
> +
> +	/* Reserve ID#0 because it is special and never implemented */
> +	bitmap_set(priv->ids_used_bimap, 0, 1);
> +
> +	return 0;
> +}
> +
> +static void __init imsic_ids_cleanup(struct imsic_priv *priv)
> +{
> +	kfree(priv->ids_target_cpu);
> +	kfree(priv->ids_enabled_bimap);
> +	kfree(priv->ids_used_bimap);
> +}
> +
> +#ifdef CONFIG_SMP
> +static void imsic_ipi_send(unsigned int cpu)
> +{
> +	struct imsic_handler *handler = per_cpu_ptr(&imsic_handlers, cpu);
> +
> +	if (!handler || !handler->priv || !handler->local.msi_va) {
> +		pr_warn("CPU%d: handler not initialized\n", cpu);
> +		return;
> +	}
> +
> +	writel(handler->priv->ipi_id, handler->local.msi_va);
> +}
> +
> +static void imsic_ipi_enable(struct imsic_priv *priv)
> +{
> +	__imsic_id_enable(priv->ipi_id);
> +	__imsic_id_enable(priv->ipi_lsync_id);
> +}
> +
> +static int __init imsic_ipi_domain_init(struct imsic_priv *priv)
> +{
> +	int virq;
> +
> +	/* Allocate interrupt identity for IPIs */
> +	virq = imsic_ids_alloc(priv, priv->global.nr_ids, get_count_order(1));
> +	if (virq < 0)
> +		return virq;
> +	priv->ipi_id = virq;
> +
> +	/* Create IMSIC IPI multiplexing */
> +	virq = ipi_mux_create(BITS_PER_BYTE, imsic_ipi_send);

Please! This BITS_PER_BYTE makes zero sense here. Have a proper define
that says 8, and document *why* this is 8! You're not defining a type
system, you're writing a irqchip driver.

> +	if (virq <= 0) {
> +		imsic_ids_free(priv, priv->ipi_id, get_count_order(1));
> +		return (virq < 0) ? virq : -ENOMEM;
> +	}
> +
> +	/* Set vIRQ range */
> +	riscv_ipi_set_virq_range(virq, BITS_PER_BYTE, true);
> +
> +	/* Allocate interrupt identity for local enable/disable sync */
> +	virq = imsic_ids_alloc(priv, priv->global.nr_ids, get_count_order(1));
> +	if (virq < 0) {
> +		imsic_ids_free(priv, priv->ipi_id, get_count_order(1));
> +		return virq;
> +	}
> +	priv->ipi_lsync_id = virq;
> +
> +	return 0;
> +}
> +
> +static void __init imsic_ipi_domain_cleanup(struct imsic_priv *priv)
> +{
> +	imsic_ids_free(priv, priv->ipi_lsync_id, get_count_order(1));
> +	if (priv->ipi_id)
> +		imsic_ids_free(priv, priv->ipi_id, get_count_order(1));
> +}
> +#else
> +static void imsic_ipi_enable(struct imsic_priv *priv)
> +{
> +}
> +
> +static int __init imsic_ipi_domain_init(struct imsic_priv *priv)
> +{
> +	/* Clear the IPI ids because we are not using IPIs */
> +	priv->ipi_id = 0;
> +	priv->ipi_lsync_id = 0;
> +	return 0;
> +}
> +
> +static void __init imsic_ipi_domain_cleanup(struct imsic_priv *priv)
> +{
> +}
> +#endif
> +
> +static void imsic_irq_mask(struct irq_data *d)
> +{
> +	imsic_id_disable(irq_data_get_irq_chip_data(d), d->hwirq);
> +}
> +
> +static void imsic_irq_unmask(struct irq_data *d)
> +{
> +	imsic_id_enable(irq_data_get_irq_chip_data(d), d->hwirq);
> +}
> +
> +static void imsic_irq_compose_msi_msg(struct irq_data *d,
> +				      struct msi_msg *msg)
> +{
> +	struct imsic_priv *priv = irq_data_get_irq_chip_data(d);
> +	unsigned int cpu;
> +	int err;
> +
> +	cpu = imsic_id_get_target(priv, d->hwirq);
> +	WARN_ON(cpu == UINT_MAX);
> +
> +	err = imsic_get_cpu_msi_msg(cpu, d->hwirq, msg);
> +	WARN_ON(err);
> +
> +	iommu_dma_compose_msi_msg(irq_data_get_msi_desc(d), msg);
> +}
> +
> +#ifdef CONFIG_SMP
> +static int imsic_irq_set_affinity(struct irq_data *d,
> +				  const struct cpumask *mask_val,
> +				  bool force)
> +{
> +	struct imsic_priv *priv = irq_data_get_irq_chip_data(d);
> +	unsigned int target_cpu;
> +	int rc;
> +
> +	rc = imsic_get_cpu(priv, mask_val, force, &target_cpu);
> +	if (rc)
> +		return rc;
> +
> +	imsic_id_set_target(priv, d->hwirq, target_cpu);
> +	irq_data_update_effective_affinity(d, cpumask_of(target_cpu));
> +
> +	return IRQ_SET_MASK_OK;
> +}
> +#endif
> +
> +static struct irq_chip imsic_irq_base_chip = {
> +	.name			= "RISC-V IMSIC-BASE",
> +	.irq_mask		= imsic_irq_mask,
> +	.irq_unmask		= imsic_irq_unmask,
> +#ifdef CONFIG_SMP
> +	.irq_set_affinity	= imsic_irq_set_affinity,
> +#endif
> +	.irq_compose_msi_msg	= imsic_irq_compose_msi_msg,
> +	.flags			= IRQCHIP_SKIP_SET_WAKE |
> +				  IRQCHIP_MASK_ON_SUSPEND,
> +};
> +
> +static int imsic_irq_domain_alloc(struct irq_domain *domain,
> +				  unsigned int virq,
> +				  unsigned int nr_irqs,
> +				  void *args)
> +{
> +	struct imsic_priv *priv = domain->host_data;
> +	msi_alloc_info_t *info = args;
> +	phys_addr_t msi_addr;
> +	int i, hwirq, err = 0;
> +	unsigned int cpu;
> +
> +	err = imsic_get_cpu(priv, &priv->lmask, false, &cpu);
> +	if (err)
> +		return err;
> +
> +	err = imsic_cpu_page_phys(cpu, 0, &msi_addr);
> +	if (err)
> +		return err;
> +
> +	hwirq = imsic_ids_alloc(priv, priv->global.nr_ids,
> +				get_count_order(nr_irqs));
> +	if (hwirq < 0)
> +		return hwirq;
> +
> +	err = iommu_dma_prepare_msi(info->desc, msi_addr);
> +	if (err)
> +		goto fail;
> +
> +	for (i = 0; i < nr_irqs; i++) {
> +		imsic_id_set_target(priv, hwirq + i, cpu);
> +		irq_domain_set_info(domain, virq + i, hwirq + i,
> +				    &imsic_irq_base_chip, priv,
> +				    handle_simple_irq, NULL, NULL);
> +		irq_set_noprobe(virq + i);
> +		irq_set_affinity(virq + i, &priv->lmask);
> +	}
> +
> +	return 0;
> +
> +fail:
> +	imsic_ids_free(priv, hwirq, get_count_order(nr_irqs));
> +	return err;
> +}
> +
> +static void imsic_irq_domain_free(struct irq_domain *domain,
> +				  unsigned int virq,
> +				  unsigned int nr_irqs)
> +{
> +	struct irq_data *d = irq_domain_get_irq_data(domain, virq);
> +	struct imsic_priv *priv = domain->host_data;
> +
> +	imsic_ids_free(priv, d->hwirq, get_count_order(nr_irqs));
> +	irq_domain_free_irqs_parent(domain, virq, nr_irqs);
> +}
> +
> +static const struct irq_domain_ops imsic_base_domain_ops = {
> +	.alloc		= imsic_irq_domain_alloc,
> +	.free		= imsic_irq_domain_free,
> +};
> +
> +#ifdef CONFIG_RISCV_IMSIC_PCI
> +
> +static void imsic_pci_mask_irq(struct irq_data *d)
> +{
> +	pci_msi_mask_irq(d);
> +	irq_chip_mask_parent(d);
> +}
> +
> +static void imsic_pci_unmask_irq(struct irq_data *d)
> +{
> +	pci_msi_unmask_irq(d);
> +	irq_chip_unmask_parent(d);
> +}
> +
> +static struct irq_chip imsic_pci_irq_chip = {
> +	.name			= "RISC-V IMSIC-PCI",
> +	.irq_mask		= imsic_pci_mask_irq,
> +	.irq_unmask		= imsic_pci_unmask_irq,
> +	.irq_eoi		= irq_chip_eoi_parent,
> +};
> +
> +static struct msi_domain_ops imsic_pci_domain_ops = {
> +};
> +
> +static struct msi_domain_info imsic_pci_domain_info = {
> +	.flags	= (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS |
> +		   MSI_FLAG_PCI_MSIX | MSI_FLAG_MULTI_PCI_MSI),
> +	.ops	= &imsic_pci_domain_ops,
> +	.chip	= &imsic_pci_irq_chip,
> +};
> +
> +#endif
> +
> +static struct irq_chip imsic_plat_irq_chip = {
> +	.name			= "RISC-V IMSIC-PLAT",
> +};
> +
> +static struct msi_domain_ops imsic_plat_domain_ops = {
> +};
> +
> +static struct msi_domain_info imsic_plat_domain_info = {
> +	.flags	= (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS),
> +	.ops	= &imsic_plat_domain_ops,
> +	.chip	= &imsic_plat_irq_chip,
> +};
> +
> +static int __init imsic_irq_domains_init(struct imsic_priv *priv,
> +					 struct fwnode_handle *fwnode)
> +{
> +	/* Create Base IRQ domain */
> +	priv->base_domain = irq_domain_create_tree(fwnode,
> +						&imsic_base_domain_ops, priv);
> +	if (!priv->base_domain) {
> +		pr_err("Failed to create IMSIC base domain\n");
> +		return -ENOMEM;
> +	}
> +	irq_domain_update_bus_token(priv->base_domain, DOMAIN_BUS_NEXUS);
> +
> +#ifdef CONFIG_RISCV_IMSIC_PCI
> +	/* Create PCI MSI domain */
> +	priv->pci_domain = pci_msi_create_irq_domain(fwnode,
> +						&imsic_pci_domain_info,
> +						priv->base_domain);
> +	if (!priv->pci_domain) {
> +		pr_err("Failed to create IMSIC PCI domain\n");
> +		irq_domain_remove(priv->base_domain);
> +		return -ENOMEM;
> +	}
> +#endif
> +
> +	/* Create Platform MSI domain */
> +	priv->plat_domain = platform_msi_create_irq_domain(fwnode,
> +						&imsic_plat_domain_info,
> +						priv->base_domain);
> +	if (!priv->plat_domain) {
> +		pr_err("Failed to create IMSIC platform domain\n");
> +		if (priv->pci_domain)
> +			irq_domain_remove(priv->pci_domain);
> +		irq_domain_remove(priv->base_domain);
> +		return -ENOMEM;
> +	}
> +
> +	return 0;
> +}
> +
> +/*
> + * To handle an interrupt, we read the TOPEI CSR and write zero in one
> + * instruction. If TOPEI CSR is non-zero then we translate TOPEI.ID to
> + * Linux interrupt number and let Linux IRQ subsystem handle it.
> + */
> +static void imsic_handle_irq(struct irq_desc *desc)
> +{
> +	struct imsic_handler *handler = this_cpu_ptr(&imsic_handlers);
> +	struct irq_chip *chip = irq_desc_get_chip(desc);
> +	struct imsic_priv *priv = handler->priv;
> +	irq_hw_number_t hwirq;
> +	int err;
> +
> +	WARN_ON_ONCE(!handler->priv);
> +
> +	chained_irq_enter(chip, desc);
> +
> +	while ((hwirq = csr_swap(CSR_TOPEI, 0))) {
> +		hwirq = hwirq >> TOPEI_ID_SHIFT;
> +
> +		if (hwirq == priv->ipi_id) {
> +#ifdef CONFIG_SMP
> +			ipi_mux_process();
> +#endif
> +			continue;
> +		} else if (hwirq == priv->ipi_lsync_id) {
> +			imsic_ids_local_sync(priv);
> +			continue;
> +		}
> +
> +		err = generic_handle_domain_irq(priv->base_domain, hwirq);
> +		if (unlikely(err))
> +			pr_warn_ratelimited(
> +				"hwirq %lu mapping not found\n", hwirq);
> +	}
> +
> +	chained_irq_exit(chip, desc);
> +}
> +
> +static int imsic_starting_cpu(unsigned int cpu)
> +{
> +	struct imsic_handler *handler = this_cpu_ptr(&imsic_handlers);
> +	struct imsic_priv *priv = handler->priv;
> +
> +	/* Enable per-CPU parent interrupt */
> +	if (imsic_parent_irq)
> +		enable_percpu_irq(imsic_parent_irq,
> +				  irq_get_trigger_type(imsic_parent_irq));

Shouldn't that be the default already?

> +	else
> +		pr_warn("cpu%d: parent irq not available\n", cpu);

And yet continue in sequence? Duh...

> +
> +	/* Enable IPIs */
> +	imsic_ipi_enable(priv);
> +
> +	/*
> +	 * Interrupts identities might have been enabled/disabled while
> +	 * this CPU was not running so sync-up local enable/disable state.
> +	 */
> +	imsic_ids_local_sync(priv);
> +
> +	/* Locally enable interrupt delivery */
> +	imsic_ids_local_delivery(priv, true);
> +
> +	return 0;
> +}
> +
> +struct imsic_fwnode_ops {
> +	u32 (*nr_parent_irq)(struct fwnode_handle *fwnode,
> +			     void *fwopaque);
> +	int (*parent_hartid)(struct fwnode_handle *fwnode,
> +			     void *fwopaque, u32 index,
> +			     unsigned long *out_hartid);
> +	u32 (*nr_mmio)(struct fwnode_handle *fwnode, void *fwopaque);
> +	int (*mmio_to_resource)(struct fwnode_handle *fwnode,
> +				void *fwopaque, u32 index,
> +				struct resource *res);
> +	void __iomem *(*mmio_map)(struct fwnode_handle *fwnode,
> +				  void *fwopaque, u32 index);
> +	int (*read_u32)(struct fwnode_handle *fwnode,
> +			void *fwopaque, const char *prop, u32 *out_val);
> +	bool (*read_bool)(struct fwnode_handle *fwnode,
> +			  void *fwopaque, const char *prop);
> +};

Why do we need this sort of (terrible) indirection?

> +
> +static int __init imsic_init(struct imsic_fwnode_ops *fwops,
> +			     struct fwnode_handle *fwnode,
> +			     void *fwopaque)
> +{
> +	struct resource res;
> +	phys_addr_t base_addr;
> +	int rc, nr_parent_irqs;
> +	struct imsic_mmio *mmio;
> +	struct imsic_priv *priv;
> +	struct irq_domain *domain;
> +	struct imsic_handler *handler;
> +	struct imsic_global_config *global;
> +	u32 i, tmp, nr_handlers = 0;
> +
> +	if (imsic_init_done) {
> +		pr_err("%pfwP: already initialized hence ignoring\n",
> +			fwnode);
> +		return -ENODEV;
> +	}
> +
> +	if (!riscv_isa_extension_available(NULL, SxAIA)) {
> +		pr_err("%pfwP: AIA support not available\n", fwnode);
> +		return -ENODEV;
> +	}
> +
> +	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
> +	if (!priv)
> +		return -ENOMEM;
> +	global = &priv->global;
> +
> +	/* Find number of parent interrupts */
> +	nr_parent_irqs = fwops->nr_parent_irq(fwnode, fwopaque);
> +	if (!nr_parent_irqs) {
> +		pr_err("%pfwP: no parent irqs available\n", fwnode);
> +		return -EINVAL;
> +	}
> +
> +	/* Find number of guest index bits in MSI address */
> +	rc = fwops->read_u32(fwnode, fwopaque, "riscv,guest-index-bits",
> +			     &global->guest_index_bits);
> +	if (rc)
> +		global->guest_index_bits = 0;
> +	tmp = BITS_PER_LONG - IMSIC_MMIO_PAGE_SHIFT;
> +	if (tmp < global->guest_index_bits) {
> +		pr_err("%pfwP: guest index bits too big\n", fwnode);
> +		return -EINVAL;
> +	}
> +
> +	/* Find number of HART index bits */
> +	rc = fwops->read_u32(fwnode, fwopaque, "riscv,hart-index-bits",
> +			     &global->hart_index_bits);
> +	if (rc) {
> +		/* Assume default value */
> +		global->hart_index_bits = __fls(nr_parent_irqs);
> +		if (BIT(global->hart_index_bits) < nr_parent_irqs)
> +			global->hart_index_bits++;
> +	}
> +	tmp = BITS_PER_LONG - IMSIC_MMIO_PAGE_SHIFT -
> +	      global->guest_index_bits;
> +	if (tmp < global->hart_index_bits) {
> +		pr_err("%pfwP: HART index bits too big\n", fwnode);
> +		return -EINVAL;
> +	}
> +
> +	/* Find number of group index bits */
> +	rc = fwops->read_u32(fwnode, fwopaque, "riscv,group-index-bits",
> +			     &global->group_index_bits);
> +	if (rc)
> +		global->group_index_bits = 0;
> +	tmp = BITS_PER_LONG - IMSIC_MMIO_PAGE_SHIFT -
> +	      global->guest_index_bits - global->hart_index_bits;
> +	if (tmp < global->group_index_bits) {
> +		pr_err("%pfwP: group index bits too big\n", fwnode);
> +		return -EINVAL;
> +	}
> +
> +	/*
> +	 * Find first bit position of group index.
> +	 * If not specified assumed the default APLIC-IMSIC configuration.
> +	 */
> +	rc = fwops->read_u32(fwnode, fwopaque, "riscv,group-index-shift",
> +			     &global->group_index_shift);
> +	if (rc)
> +		global->group_index_shift = IMSIC_MMIO_PAGE_SHIFT * 2;
> +	tmp = global->group_index_bits + global->group_index_shift - 1;
> +	if (tmp >= BITS_PER_LONG) {
> +		pr_err("%pfwP: group index shift too big\n", fwnode);
> +		return -EINVAL;
> +	}
> +
> +	/* Find number of interrupt identities */
> +	rc = fwops->read_u32(fwnode, fwopaque, "riscv,num-ids",
> +			     &global->nr_ids);
> +	if (rc) {
> +		pr_err("%pfwP: number of interrupt identities not found\n",
> +			fwnode);
> +		return rc;
> +	}
> +	if ((global->nr_ids < IMSIC_MIN_ID) ||
> +	    (global->nr_ids >= IMSIC_MAX_ID) ||
> +	    ((global->nr_ids & IMSIC_MIN_ID) != IMSIC_MIN_ID)) {
> +		pr_err("%pfwP: invalid number of interrupt identities\n",
> +			fwnode);
> +		return -EINVAL;
> +	}
> +
> +	/* Find number of guest interrupt identities */
> +	if (fwops->read_u32(fwnode, fwopaque, "riscv,num-guest-ids",
> +			    &global->nr_guest_ids))
> +		global->nr_guest_ids = global->nr_ids;
> +	if ((global->nr_guest_ids < IMSIC_MIN_ID) ||
> +	    (global->nr_guest_ids >= IMSIC_MAX_ID) ||
> +	    ((global->nr_guest_ids & IMSIC_MIN_ID) != IMSIC_MIN_ID)) {
> +		pr_err("%pfwP: invalid number of guest interrupt identities\n",
> +			fwnode);
> +		return -EINVAL;
> +	}

Please split the whole guest stuff out. It is totally unused!

I've stopped reading. This needs structure, cleanups and a bit of
taste. Not a lot of that here at the moment.

	M.

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

Re: [PATCH v2 5/9] irqchip: Add RISC-V incoming MSI controller driver

Posted by Anup Patel 2 years, 4 months ago

On Fri, Jan 13, 2023 at 3:40 PM Marc Zyngier <maz@kernel.org> wrote:
>
> On Tue, 03 Jan 2023 14:14:05 +0000,
> Anup Patel <apatel@ventanamicro.com> wrote:
> >
> > The RISC-V advanced interrupt architecture (AIA) specification defines
> > a new MSI controller for managing MSIs on a RISC-V platform. This new
> > MSI controller is referred to as incoming message signaled interrupt
> > controller (IMSIC) which manages MSI on per-HART (or per-CPU) basis.
> > (For more details refer https://github.com/riscv/riscv-aia)
>
> And how about IPIs, which this driver seems to be concerned about?

Okay, I will mention about IPIs in the commit description.

>
> >
> > This patch adds an irqchip driver for RISC-V IMSIC found on RISC-V
> > platforms.
> >
> > Signed-off-by: Anup Patel <apatel@ventanamicro.com>
> > ---
> >  drivers/irqchip/Kconfig             |   14 +-
> >  drivers/irqchip/Makefile            |    1 +
> >  drivers/irqchip/irq-riscv-imsic.c   | 1174 +++++++++++++++++++++++++++
> >  include/linux/irqchip/riscv-imsic.h |   92 +++
> >  4 files changed, 1280 insertions(+), 1 deletion(-)
> >  create mode 100644 drivers/irqchip/irq-riscv-imsic.c
> >  create mode 100644 include/linux/irqchip/riscv-imsic.h
> >
> > diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig
> > index 9e65345ca3f6..a1315189a595 100644
> > --- a/drivers/irqchip/Kconfig
> > +++ b/drivers/irqchip/Kconfig
> > @@ -29,7 +29,6 @@ config ARM_GIC_V2M
> >
> >  config GIC_NON_BANKED
> >       bool
> > -
> >  config ARM_GIC_V3
> >       bool
> >       select IRQ_DOMAIN_HIERARCHY
> > @@ -548,6 +547,19 @@ config SIFIVE_PLIC
> >       select IRQ_DOMAIN_HIERARCHY
> >       select GENERIC_IRQ_EFFECTIVE_AFF_MASK if SMP
> >
> > +config RISCV_IMSIC
> > +     bool
> > +     depends on RISCV
> > +     select IRQ_DOMAIN_HIERARCHY
> > +     select GENERIC_MSI_IRQ_DOMAIN
> > +
> > +config RISCV_IMSIC_PCI
> > +     bool
> > +     depends on RISCV_IMSIC
> > +     depends on PCI
> > +     depends on PCI_MSI
> > +     default RISCV_IMSIC
>
> This should definitely tell you that this driver needs splitting.

The code under "#ifdef CONFIG_RISCV_IMSIC_PCI" is hardly 40 lines
so I felt it was too small to deserve its own source file.

>
> > +
> >  config EXYNOS_IRQ_COMBINER
> >       bool "Samsung Exynos IRQ combiner support" if COMPILE_TEST
> >       depends on (ARCH_EXYNOS && ARM) || COMPILE_TEST
> > diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile
> > index 87b49a10962c..22c723cc6ec8 100644
> > --- a/drivers/irqchip/Makefile
> > +++ b/drivers/irqchip/Makefile
> > @@ -96,6 +96,7 @@ obj-$(CONFIG_QCOM_MPM)                      += irq-qcom-mpm.o
> >  obj-$(CONFIG_CSKY_MPINTC)            += irq-csky-mpintc.o
> >  obj-$(CONFIG_CSKY_APB_INTC)          += irq-csky-apb-intc.o
> >  obj-$(CONFIG_RISCV_INTC)             += irq-riscv-intc.o
> > +obj-$(CONFIG_RISCV_IMSIC)            += irq-riscv-imsic.o
> >  obj-$(CONFIG_SIFIVE_PLIC)            += irq-sifive-plic.o
> >  obj-$(CONFIG_IMX_IRQSTEER)           += irq-imx-irqsteer.o
> >  obj-$(CONFIG_IMX_INTMUX)             += irq-imx-intmux.o
> > diff --git a/drivers/irqchip/irq-riscv-imsic.c b/drivers/irqchip/irq-riscv-imsic.c
> > new file mode 100644
> > index 000000000000..4c16b66738d6
> > --- /dev/null
> > +++ b/drivers/irqchip/irq-riscv-imsic.c
> > @@ -0,0 +1,1174 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * Copyright (C) 2021 Western Digital Corporation or its affiliates.
> > + * Copyright (C) 2022 Ventana Micro Systems Inc.
> > + */
> > +
> > +#define pr_fmt(fmt) "riscv-imsic: " fmt
> > +#include <linux/bitmap.h>
> > +#include <linux/cpu.h>
> > +#include <linux/interrupt.h>
> > +#include <linux/io.h>
> > +#include <linux/iommu.h>
> > +#include <linux/irq.h>
> > +#include <linux/irqchip.h>
> > +#include <linux/irqchip/chained_irq.h>
> > +#include <linux/irqchip/riscv-imsic.h>
> > +#include <linux/irqdomain.h>
> > +#include <linux/module.h>
> > +#include <linux/msi.h>
> > +#include <linux/of.h>
> > +#include <linux/of_address.h>
> > +#include <linux/of_irq.h>
> > +#include <linux/pci.h>
> > +#include <linux/platform_device.h>
> > +#include <linux/spinlock.h>
> > +#include <linux/smp.h>
> > +#include <asm/hwcap.h>
> > +
> > +#define IMSIC_DISABLE_EIDELIVERY     0
> > +#define IMSIC_ENABLE_EIDELIVERY              1
> > +#define IMSIC_DISABLE_EITHRESHOLD    1
> > +#define IMSIC_ENABLE_EITHRESHOLD     0
> > +
> > +#define imsic_csr_write(__c, __v)    \
> > +do {                                 \
> > +     csr_write(CSR_ISELECT, __c);    \
> > +     csr_write(CSR_IREG, __v);       \
> > +} while (0)
> > +
> > +#define imsic_csr_read(__c)          \
> > +({                                   \
> > +     unsigned long __v;              \
> > +     csr_write(CSR_ISELECT, __c);    \
> > +     __v = csr_read(CSR_IREG);       \
> > +     __v;                            \
> > +})
> > +
> > +#define imsic_csr_set(__c, __v)              \
> > +do {                                 \
> > +     csr_write(CSR_ISELECT, __c);    \
> > +     csr_set(CSR_IREG, __v);         \
> > +} while (0)
> > +
> > +#define imsic_csr_clear(__c, __v)    \
> > +do {                                 \
> > +     csr_write(CSR_ISELECT, __c);    \
> > +     csr_clear(CSR_IREG, __v);       \
> > +} while (0)
> > +
> > +struct imsic_mmio {
> > +     phys_addr_t pa;
> > +     void __iomem *va;
> > +     unsigned long size;
> > +};
> > +
> > +struct imsic_priv {
> > +     /* Global configuration common for all HARTs */
> > +     struct imsic_global_config global;
> > +
> > +     /* MMIO regions */
> > +     u32 num_mmios;
> > +     struct imsic_mmio *mmios;
> > +
> > +     /* Global state of interrupt identities */
> > +     raw_spinlock_t ids_lock;
> > +     unsigned long *ids_used_bimap;
> > +     unsigned long *ids_enabled_bimap;
> > +     unsigned int *ids_target_cpu;
> > +
> > +     /* Mask for connected CPUs */
> > +     struct cpumask lmask;
> > +
> > +     /* IPI interrupt identity */
> > +     u32 ipi_id;
> > +     u32 ipi_lsync_id;
> > +
> > +     /* IRQ domains */
> > +     struct irq_domain *base_domain;
> > +     struct irq_domain *pci_domain;
> > +     struct irq_domain *plat_domain;
> > +};
> > +
> > +struct imsic_handler {
> > +     /* Local configuration for given HART */
> > +     struct imsic_local_config local;
> > +
> > +     /* Pointer to private context */
> > +     struct imsic_priv *priv;
> > +};
> > +
> > +static bool imsic_init_done;
> > +
> > +static int imsic_parent_irq;
> > +static DEFINE_PER_CPU(struct imsic_handler, imsic_handlers);
> > +
> > +const struct imsic_global_config *imsic_get_global_config(void)
> > +{
> > +     struct imsic_handler *handler = this_cpu_ptr(&imsic_handlers);
> > +
> > +     if (!handler || !handler->priv)
> > +             return NULL;
> > +
> > +     return &handler->priv->global;
> > +}
> > +EXPORT_SYMBOL_GPL(imsic_get_global_config);
> > +
> > +const struct imsic_local_config *imsic_get_local_config(unsigned int cpu)
> > +{
> > +     struct imsic_handler *handler = per_cpu_ptr(&imsic_handlers, cpu);
> > +
> > +     if (!handler || !handler->priv)
> > +             return NULL;
>
> How can this happen?

These are redundant checks. I will drop.

>
> > +
> > +     return &handler->local;
> > +}
> > +EXPORT_SYMBOL_GPL(imsic_get_local_config);
>
> Why are these symbols exported? They have no user, so they shouldn't
> even exist here. I also seriously doubt there is a valid use case for
> exposing this information to the rest of the kernel.

The imsic_get_global_config() is used by APLIC driver and KVM RISC-V
module whereas imsic_get_local_config() is only used by KVM RISC-V.

The KVM RISC-V AIA irqchip patches are available in riscv_kvm_aia_v1
branch at: https://github.com/avpatel/linux.git. I have not posted KVM RISC-V
patches due various interdependencies.

>
> > +
> > +static int imsic_cpu_page_phys(unsigned int cpu,
> > +                            unsigned int guest_index,
> > +                            phys_addr_t *out_msi_pa)
> > +{
> > +     struct imsic_handler *handler = per_cpu_ptr(&imsic_handlers, cpu);
> > +     struct imsic_global_config *global;
> > +     struct imsic_local_config *local;
> > +
> > +     if (!handler || !handler->priv)
> > +             return -ENODEV;
> > +     local = &handler->local;
> > +     global = &handler->priv->global;
> > +
> > +     if (BIT(global->guest_index_bits) <= guest_index)
> > +             return -EINVAL;
> > +
> > +     if (out_msi_pa)
> > +             *out_msi_pa = local->msi_pa +
> > +                           (guest_index * IMSIC_MMIO_PAGE_SZ);
> > +
> > +     return 0;
> > +}
> > +
> > +static int imsic_get_cpu(struct imsic_priv *priv,
> > +                      const struct cpumask *mask_val, bool force,
> > +                      unsigned int *out_target_cpu)
> > +{
> > +     struct cpumask amask;
> > +     unsigned int cpu;
> > +
> > +     cpumask_and(&amask, &priv->lmask, mask_val);
> > +
> > +     if (force)
> > +             cpu = cpumask_first(&amask);
> > +     else
> > +             cpu = cpumask_any_and(&amask, cpu_online_mask);
> > +
> > +     if (cpu >= nr_cpu_ids)
> > +             return -EINVAL;
> > +
> > +     if (out_target_cpu)
> > +             *out_target_cpu = cpu;
> > +
> > +     return 0;
> > +}
> > +
> > +static int imsic_get_cpu_msi_msg(unsigned int cpu, unsigned int id,
> > +                              struct msi_msg *msg)
> > +{
> > +     phys_addr_t msi_addr;
> > +     int err;
> > +
> > +     err = imsic_cpu_page_phys(cpu, 0, &msi_addr);
> > +     if (err)
> > +             return err;
> > +
> > +     msg->address_hi = upper_32_bits(msi_addr);
> > +     msg->address_lo = lower_32_bits(msi_addr);
> > +     msg->data = id;
> > +
> > +     return err;
> > +}
> > +
> > +static void imsic_id_set_target(struct imsic_priv *priv,
> > +                              unsigned int id, unsigned int target_cpu)
> > +{
> > +     unsigned long flags;
> > +
> > +     raw_spin_lock_irqsave(&priv->ids_lock, flags);
> > +     priv->ids_target_cpu[id] = target_cpu;
> > +     raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
> > +}
> > +
> > +static unsigned int imsic_id_get_target(struct imsic_priv *priv,
> > +                                     unsigned int id)
> > +{
> > +     unsigned int ret;
> > +     unsigned long flags;
> > +
> > +     raw_spin_lock_irqsave(&priv->ids_lock, flags);
> > +     ret = priv->ids_target_cpu[id];
> > +     raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
> > +
> > +     return ret;
> > +}
> > +
> > +static void __imsic_eix_update(unsigned long base_id,
> > +                            unsigned long num_id, bool pend, bool val)
> > +{
> > +     unsigned long i, isel, ireg, flags;
> > +     unsigned long id = base_id, last_id = base_id + num_id;
> > +
> > +     while (id < last_id) {
> > +             isel = id / BITS_PER_LONG;
> > +             isel *= BITS_PER_LONG / IMSIC_EIPx_BITS;
> > +             isel += (pend) ? IMSIC_EIP0 : IMSIC_EIE0;
> > +
> > +             ireg = 0;
> > +             for (i = id & (__riscv_xlen - 1);
> > +                  (id < last_id) && (i < __riscv_xlen); i++) {
> > +                     ireg |= BIT(i);
> > +                     id++;
> > +             }
> > +
> > +             /*
> > +              * The IMSIC EIEx and EIPx registers are indirectly
> > +              * accessed via using ISELECT and IREG CSRs so we
> > +              * save/restore local IRQ to ensure that we don't
> > +              * get preempted while accessing IMSIC registers.
> > +              */
> > +             local_irq_save(flags);
> > +             if (val)
> > +                     imsic_csr_set(isel, ireg);
> > +             else
> > +                     imsic_csr_clear(isel, ireg);
> > +             local_irq_restore(flags);
>
> What is the actual requirement? no preemption? or no interrupts? This
> isn't the same thing. Also, a bunch of the users already disable
> interrupts. Consistency wouldn't hurt here.

No preemption is the only requirement. Since the callers of these
functions disable local IRQ, I think we don't need to not anything
special here. I will drop the local IRQ save/restore and update
the comments as well.

>
> > +     }
> > +}
> > +
> > +#define __imsic_id_enable(__id)              \
> > +     __imsic_eix_update((__id), 1, false, true)
> > +#define __imsic_id_disable(__id)     \
> > +     __imsic_eix_update((__id), 1, false, false)
> > +
> > +#ifdef CONFIG_SMP
> > +static void __imsic_id_smp_sync(struct imsic_priv *priv)
> > +{
> > +     struct imsic_handler *handler;
> > +     struct cpumask amask;
> > +     int cpu;
> > +
> > +     cpumask_and(&amask, &priv->lmask, cpu_online_mask);
>
> Can't this race against a CPU going down?

Yes, it can race if a CPU goes down while we are in this function
but this won't be a problem because the imsic_starting_cpu()
will unconditionally do imsic_ids_local_sync() when the CPU is
brought-up again. I will add a multiline comment block explaining
this.

>
> > +     for_each_cpu(cpu, &amask) {
> > +             if (cpu == smp_processor_id())
> > +                     continue;
> > +
> > +             handler = per_cpu_ptr(&imsic_handlers, cpu);
> > +             if (!handler || !handler->priv || !handler->local.msi_va) {
> > +                     pr_warn("CPU%d: handler not initialized\n", cpu);
>
> How many times are you going to do that? On each failing synchronisation?

My bad for adding these paranoid checks. I remove these checks
wherever possible.

>
> > +                     continue;
> > +             }
> > +
> > +             writel(handler->priv->ipi_lsync_id, handler->local.msi_va);
>
> As I understand it, this is a "behind the scenes" IPI. Why isn't that
> a *real* IPI?

Yes, that's correct. The ID enable bits are per-CPU accessible only
via CSRs hence we have a special "behind the scenes" IPI to
synchronize state of ID enable bits.

>
> > +     }
> > +}
> > +#else
> > +#define __imsic_id_smp_sync(__priv)
> > +#endif
> > +
> > +static void imsic_id_enable(struct imsic_priv *priv, unsigned int id)
> > +{
> > +     unsigned long flags;
> > +
> > +     raw_spin_lock_irqsave(&priv->ids_lock, flags);
> > +     bitmap_set(priv->ids_enabled_bimap, id, 1);
> > +     __imsic_id_enable(id);
> > +     raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
> > +
> > +     __imsic_id_smp_sync(priv);
> > +}
> > +
> > +static void imsic_id_disable(struct imsic_priv *priv, unsigned int id)
> > +{
> > +     unsigned long flags;
> > +
> > +     raw_spin_lock_irqsave(&priv->ids_lock, flags);
> > +     bitmap_clear(priv->ids_enabled_bimap, id, 1);
> > +     __imsic_id_disable(id);
> > +     raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
> > +
> > +     __imsic_id_smp_sync(priv);
> > +}
> > +
> > +static void imsic_ids_local_sync(struct imsic_priv *priv)
> > +{
> > +     int i;
> > +     unsigned long flags;
> > +
> > +     raw_spin_lock_irqsave(&priv->ids_lock, flags);
> > +     for (i = 1; i <= priv->global.nr_ids; i++) {
> > +             if (priv->ipi_id == i || priv->ipi_lsync_id == i)
> > +                     continue;
> > +
> > +             if (test_bit(i, priv->ids_enabled_bimap))
> > +                     __imsic_id_enable(i);
> > +             else
> > +                     __imsic_id_disable(i);
> > +     }
> > +     raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
> > +}
> > +
> > +static void imsic_ids_local_delivery(struct imsic_priv *priv, bool enable)
> > +{
> > +     if (enable) {
> > +             imsic_csr_write(IMSIC_EITHRESHOLD, IMSIC_ENABLE_EITHRESHOLD);
> > +             imsic_csr_write(IMSIC_EIDELIVERY, IMSIC_ENABLE_EIDELIVERY);
> > +     } else {
> > +             imsic_csr_write(IMSIC_EIDELIVERY, IMSIC_DISABLE_EIDELIVERY);
> > +             imsic_csr_write(IMSIC_EITHRESHOLD, IMSIC_DISABLE_EITHRESHOLD);
> > +     }
> > +}
> > +
> > +static int imsic_ids_alloc(struct imsic_priv *priv,
> > +                        unsigned int max_id, unsigned int order)
> > +{
> > +     int ret;
> > +     unsigned long flags;
> > +
> > +     if ((priv->global.nr_ids < max_id) ||
> > +         (max_id < BIT(order)))
> > +             return -EINVAL;
>
> Why do we need this check? Shouldn't that be guaranteed by
> construction?

Yes, these are redundant checks. I will remove it.

>
> > +
> > +     raw_spin_lock_irqsave(&priv->ids_lock, flags);
> > +     ret = bitmap_find_free_region(priv->ids_used_bimap,
> > +                                   max_id + 1, order);
> > +     raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
> > +
> > +     return ret;
> > +}
> > +
> > +static void imsic_ids_free(struct imsic_priv *priv, unsigned int base_id,
> > +                        unsigned int order)
> > +{
> > +     unsigned long flags;
> > +
> > +     raw_spin_lock_irqsave(&priv->ids_lock, flags);
> > +     bitmap_release_region(priv->ids_used_bimap, base_id, order);
> > +     raw_spin_unlock_irqrestore(&priv->ids_lock, flags);
> > +}
> > +
> > +static int __init imsic_ids_init(struct imsic_priv *priv)
> > +{
> > +     int i;
> > +     struct imsic_global_config *global = &priv->global;
> > +
> > +     raw_spin_lock_init(&priv->ids_lock);
> > +
> > +     /* Allocate used bitmap */
> > +     priv->ids_used_bimap = kcalloc(BITS_TO_LONGS(global->nr_ids + 1),
> > +                                     sizeof(unsigned long), GFP_KERNEL);
>
> How about bitmap_alloc?

Okay, I will use bitmap_zalloc() here.

>
> > +     if (!priv->ids_used_bimap)
> > +             return -ENOMEM;
> > +
> > +     /* Allocate enabled bitmap */
> > +     priv->ids_enabled_bimap = kcalloc(BITS_TO_LONGS(global->nr_ids + 1),
> > +                                        sizeof(unsigned long), GFP_KERNEL);
> > +     if (!priv->ids_enabled_bimap) {
> > +             kfree(priv->ids_used_bimap);
> > +             return -ENOMEM;
> > +     }
> > +
> > +     /* Allocate target CPU array */
> > +     priv->ids_target_cpu = kcalloc(global->nr_ids + 1,
> > +                                    sizeof(unsigned int), GFP_KERNEL);
> > +     if (!priv->ids_target_cpu) {
> > +             kfree(priv->ids_enabled_bimap);
> > +             kfree(priv->ids_used_bimap);
> > +             return -ENOMEM;
> > +     }
> > +     for (i = 0; i <= global->nr_ids; i++)
> > +             priv->ids_target_cpu[i] = UINT_MAX;
> > +
> > +     /* Reserve ID#0 because it is special and never implemented */
> > +     bitmap_set(priv->ids_used_bimap, 0, 1);
> > +
> > +     return 0;
> > +}
> > +
> > +static void __init imsic_ids_cleanup(struct imsic_priv *priv)
> > +{
> > +     kfree(priv->ids_target_cpu);
> > +     kfree(priv->ids_enabled_bimap);
> > +     kfree(priv->ids_used_bimap);
> > +}
> > +
> > +#ifdef CONFIG_SMP
> > +static void imsic_ipi_send(unsigned int cpu)
> > +{
> > +     struct imsic_handler *handler = per_cpu_ptr(&imsic_handlers, cpu);
> > +
> > +     if (!handler || !handler->priv || !handler->local.msi_va) {
> > +             pr_warn("CPU%d: handler not initialized\n", cpu);
> > +             return;
> > +     }
> > +
> > +     writel(handler->priv->ipi_id, handler->local.msi_va);
> > +}
> > +
> > +static void imsic_ipi_enable(struct imsic_priv *priv)
> > +{
> > +     __imsic_id_enable(priv->ipi_id);
> > +     __imsic_id_enable(priv->ipi_lsync_id);
> > +}
> > +
> > +static int __init imsic_ipi_domain_init(struct imsic_priv *priv)
> > +{
> > +     int virq;
> > +
> > +     /* Allocate interrupt identity for IPIs */
> > +     virq = imsic_ids_alloc(priv, priv->global.nr_ids, get_count_order(1));
> > +     if (virq < 0)
> > +             return virq;
> > +     priv->ipi_id = virq;
> > +
> > +     /* Create IMSIC IPI multiplexing */
> > +     virq = ipi_mux_create(BITS_PER_BYTE, imsic_ipi_send);
>
> Please! This BITS_PER_BYTE makes zero sense here. Have a proper define
> that says 8, and document *why* this is 8! You're not defining a type
> system, you're writing a irqchip driver.

Okay, I will add a "#define" for the number of IPI with an explanation
for *why*.

>
> > +     if (virq <= 0) {
> > +             imsic_ids_free(priv, priv->ipi_id, get_count_order(1));
> > +             return (virq < 0) ? virq : -ENOMEM;
> > +     }
> > +
> > +     /* Set vIRQ range */
> > +     riscv_ipi_set_virq_range(virq, BITS_PER_BYTE, true);
> > +
> > +     /* Allocate interrupt identity for local enable/disable sync */
> > +     virq = imsic_ids_alloc(priv, priv->global.nr_ids, get_count_order(1));
> > +     if (virq < 0) {
> > +             imsic_ids_free(priv, priv->ipi_id, get_count_order(1));
> > +             return virq;
> > +     }
> > +     priv->ipi_lsync_id = virq;
> > +
> > +     return 0;
> > +}
> > +
> > +static void __init imsic_ipi_domain_cleanup(struct imsic_priv *priv)
> > +{
> > +     imsic_ids_free(priv, priv->ipi_lsync_id, get_count_order(1));
> > +     if (priv->ipi_id)
> > +             imsic_ids_free(priv, priv->ipi_id, get_count_order(1));
> > +}
> > +#else
> > +static void imsic_ipi_enable(struct imsic_priv *priv)
> > +{
> > +}
> > +
> > +static int __init imsic_ipi_domain_init(struct imsic_priv *priv)
> > +{
> > +     /* Clear the IPI ids because we are not using IPIs */
> > +     priv->ipi_id = 0;
> > +     priv->ipi_lsync_id = 0;
> > +     return 0;
> > +}
> > +
> > +static void __init imsic_ipi_domain_cleanup(struct imsic_priv *priv)
> > +{
> > +}
> > +#endif
> > +
> > +static void imsic_irq_mask(struct irq_data *d)
> > +{
> > +     imsic_id_disable(irq_data_get_irq_chip_data(d), d->hwirq);
> > +}
> > +
> > +static void imsic_irq_unmask(struct irq_data *d)
> > +{
> > +     imsic_id_enable(irq_data_get_irq_chip_data(d), d->hwirq);
> > +}
> > +
> > +static void imsic_irq_compose_msi_msg(struct irq_data *d,
> > +                                   struct msi_msg *msg)
> > +{
> > +     struct imsic_priv *priv = irq_data_get_irq_chip_data(d);
> > +     unsigned int cpu;
> > +     int err;
> > +
> > +     cpu = imsic_id_get_target(priv, d->hwirq);
> > +     WARN_ON(cpu == UINT_MAX);
> > +
> > +     err = imsic_get_cpu_msi_msg(cpu, d->hwirq, msg);
> > +     WARN_ON(err);
> > +
> > +     iommu_dma_compose_msi_msg(irq_data_get_msi_desc(d), msg);
> > +}
> > +
> > +#ifdef CONFIG_SMP
> > +static int imsic_irq_set_affinity(struct irq_data *d,
> > +                               const struct cpumask *mask_val,
> > +                               bool force)
> > +{
> > +     struct imsic_priv *priv = irq_data_get_irq_chip_data(d);
> > +     unsigned int target_cpu;
> > +     int rc;
> > +
> > +     rc = imsic_get_cpu(priv, mask_val, force, &target_cpu);
> > +     if (rc)
> > +             return rc;
> > +
> > +     imsic_id_set_target(priv, d->hwirq, target_cpu);
> > +     irq_data_update_effective_affinity(d, cpumask_of(target_cpu));
> > +
> > +     return IRQ_SET_MASK_OK;
> > +}
> > +#endif
> > +
> > +static struct irq_chip imsic_irq_base_chip = {
> > +     .name                   = "RISC-V IMSIC-BASE",
> > +     .irq_mask               = imsic_irq_mask,
> > +     .irq_unmask             = imsic_irq_unmask,
> > +#ifdef CONFIG_SMP
> > +     .irq_set_affinity       = imsic_irq_set_affinity,
> > +#endif
> > +     .irq_compose_msi_msg    = imsic_irq_compose_msi_msg,
> > +     .flags                  = IRQCHIP_SKIP_SET_WAKE |
> > +                               IRQCHIP_MASK_ON_SUSPEND,
> > +};
> > +
> > +static int imsic_irq_domain_alloc(struct irq_domain *domain,
> > +                               unsigned int virq,
> > +                               unsigned int nr_irqs,
> > +                               void *args)
> > +{
> > +     struct imsic_priv *priv = domain->host_data;
> > +     msi_alloc_info_t *info = args;
> > +     phys_addr_t msi_addr;
> > +     int i, hwirq, err = 0;
> > +     unsigned int cpu;
> > +
> > +     err = imsic_get_cpu(priv, &priv->lmask, false, &cpu);
> > +     if (err)
> > +             return err;
> > +
> > +     err = imsic_cpu_page_phys(cpu, 0, &msi_addr);
> > +     if (err)
> > +             return err;
> > +
> > +     hwirq = imsic_ids_alloc(priv, priv->global.nr_ids,
> > +                             get_count_order(nr_irqs));
> > +     if (hwirq < 0)
> > +             return hwirq;
> > +
> > +     err = iommu_dma_prepare_msi(info->desc, msi_addr);
> > +     if (err)
> > +             goto fail;
> > +
> > +     for (i = 0; i < nr_irqs; i++) {
> > +             imsic_id_set_target(priv, hwirq + i, cpu);
> > +             irq_domain_set_info(domain, virq + i, hwirq + i,
> > +                                 &imsic_irq_base_chip, priv,
> > +                                 handle_simple_irq, NULL, NULL);
> > +             irq_set_noprobe(virq + i);
> > +             irq_set_affinity(virq + i, &priv->lmask);
> > +     }
> > +
> > +     return 0;
> > +
> > +fail:
> > +     imsic_ids_free(priv, hwirq, get_count_order(nr_irqs));
> > +     return err;
> > +}
> > +
> > +static void imsic_irq_domain_free(struct irq_domain *domain,
> > +                               unsigned int virq,
> > +                               unsigned int nr_irqs)
> > +{
> > +     struct irq_data *d = irq_domain_get_irq_data(domain, virq);
> > +     struct imsic_priv *priv = domain->host_data;
> > +
> > +     imsic_ids_free(priv, d->hwirq, get_count_order(nr_irqs));
> > +     irq_domain_free_irqs_parent(domain, virq, nr_irqs);
> > +}
> > +
> > +static const struct irq_domain_ops imsic_base_domain_ops = {
> > +     .alloc          = imsic_irq_domain_alloc,
> > +     .free           = imsic_irq_domain_free,
> > +};
> > +
> > +#ifdef CONFIG_RISCV_IMSIC_PCI
> > +
> > +static void imsic_pci_mask_irq(struct irq_data *d)
> > +{
> > +     pci_msi_mask_irq(d);
> > +     irq_chip_mask_parent(d);
> > +}
> > +
> > +static void imsic_pci_unmask_irq(struct irq_data *d)
> > +{
> > +     pci_msi_unmask_irq(d);
> > +     irq_chip_unmask_parent(d);
> > +}
> > +
> > +static struct irq_chip imsic_pci_irq_chip = {
> > +     .name                   = "RISC-V IMSIC-PCI",
> > +     .irq_mask               = imsic_pci_mask_irq,
> > +     .irq_unmask             = imsic_pci_unmask_irq,
> > +     .irq_eoi                = irq_chip_eoi_parent,
> > +};
> > +
> > +static struct msi_domain_ops imsic_pci_domain_ops = {
> > +};
> > +
> > +static struct msi_domain_info imsic_pci_domain_info = {
> > +     .flags  = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS |
> > +                MSI_FLAG_PCI_MSIX | MSI_FLAG_MULTI_PCI_MSI),
> > +     .ops    = &imsic_pci_domain_ops,
> > +     .chip   = &imsic_pci_irq_chip,
> > +};
> > +
> > +#endif
> > +
> > +static struct irq_chip imsic_plat_irq_chip = {
> > +     .name                   = "RISC-V IMSIC-PLAT",
> > +};
> > +
> > +static struct msi_domain_ops imsic_plat_domain_ops = {
> > +};
> > +
> > +static struct msi_domain_info imsic_plat_domain_info = {
> > +     .flags  = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS),
> > +     .ops    = &imsic_plat_domain_ops,
> > +     .chip   = &imsic_plat_irq_chip,
> > +};
> > +
> > +static int __init imsic_irq_domains_init(struct imsic_priv *priv,
> > +                                      struct fwnode_handle *fwnode)
> > +{
> > +     /* Create Base IRQ domain */
> > +     priv->base_domain = irq_domain_create_tree(fwnode,
> > +                                             &imsic_base_domain_ops, priv);
> > +     if (!priv->base_domain) {
> > +             pr_err("Failed to create IMSIC base domain\n");
> > +             return -ENOMEM;
> > +     }
> > +     irq_domain_update_bus_token(priv->base_domain, DOMAIN_BUS_NEXUS);
> > +
> > +#ifdef CONFIG_RISCV_IMSIC_PCI
> > +     /* Create PCI MSI domain */
> > +     priv->pci_domain = pci_msi_create_irq_domain(fwnode,
> > +                                             &imsic_pci_domain_info,
> > +                                             priv->base_domain);
> > +     if (!priv->pci_domain) {
> > +             pr_err("Failed to create IMSIC PCI domain\n");
> > +             irq_domain_remove(priv->base_domain);
> > +             return -ENOMEM;
> > +     }
> > +#endif
> > +
> > +     /* Create Platform MSI domain */
> > +     priv->plat_domain = platform_msi_create_irq_domain(fwnode,
> > +                                             &imsic_plat_domain_info,
> > +                                             priv->base_domain);
> > +     if (!priv->plat_domain) {
> > +             pr_err("Failed to create IMSIC platform domain\n");
> > +             if (priv->pci_domain)
> > +                     irq_domain_remove(priv->pci_domain);
> > +             irq_domain_remove(priv->base_domain);
> > +             return -ENOMEM;
> > +     }
> > +
> > +     return 0;
> > +}
> > +
> > +/*
> > + * To handle an interrupt, we read the TOPEI CSR and write zero in one
> > + * instruction. If TOPEI CSR is non-zero then we translate TOPEI.ID to
> > + * Linux interrupt number and let Linux IRQ subsystem handle it.
> > + */
> > +static void imsic_handle_irq(struct irq_desc *desc)
> > +{
> > +     struct imsic_handler *handler = this_cpu_ptr(&imsic_handlers);
> > +     struct irq_chip *chip = irq_desc_get_chip(desc);
> > +     struct imsic_priv *priv = handler->priv;
> > +     irq_hw_number_t hwirq;
> > +     int err;
> > +
> > +     WARN_ON_ONCE(!handler->priv);
> > +
> > +     chained_irq_enter(chip, desc);
> > +
> > +     while ((hwirq = csr_swap(CSR_TOPEI, 0))) {
> > +             hwirq = hwirq >> TOPEI_ID_SHIFT;
> > +
> > +             if (hwirq == priv->ipi_id) {
> > +#ifdef CONFIG_SMP
> > +                     ipi_mux_process();
> > +#endif
> > +                     continue;
> > +             } else if (hwirq == priv->ipi_lsync_id) {
> > +                     imsic_ids_local_sync(priv);
> > +                     continue;
> > +             }
> > +
> > +             err = generic_handle_domain_irq(priv->base_domain, hwirq);
> > +             if (unlikely(err))
> > +                     pr_warn_ratelimited(
> > +                             "hwirq %lu mapping not found\n", hwirq);
> > +     }
> > +
> > +     chained_irq_exit(chip, desc);
> > +}
> > +
> > +static int imsic_starting_cpu(unsigned int cpu)
> > +{
> > +     struct imsic_handler *handler = this_cpu_ptr(&imsic_handlers);
> > +     struct imsic_priv *priv = handler->priv;
> > +
> > +     /* Enable per-CPU parent interrupt */
> > +     if (imsic_parent_irq)
> > +             enable_percpu_irq(imsic_parent_irq,
> > +                               irq_get_trigger_type(imsic_parent_irq));
>
> Shouldn't that be the default already?

The imsic_parent_irq is already set before calling imsic_starting_cpu()
on each CPU so we can drop the if-check.

>
> > +     else
> > +             pr_warn("cpu%d: parent irq not available\n", cpu);
>
> And yet continue in sequence? Duh...

This warning is also not required.

>
> > +
> > +     /* Enable IPIs */
> > +     imsic_ipi_enable(priv);
> > +
> > +     /*
> > +      * Interrupts identities might have been enabled/disabled while
> > +      * this CPU was not running so sync-up local enable/disable state.
> > +      */
> > +     imsic_ids_local_sync(priv);
> > +
> > +     /* Locally enable interrupt delivery */
> > +     imsic_ids_local_delivery(priv, true);
> > +
> > +     return 0;
> > +}
> > +
> > +struct imsic_fwnode_ops {
> > +     u32 (*nr_parent_irq)(struct fwnode_handle *fwnode,
> > +                          void *fwopaque);
> > +     int (*parent_hartid)(struct fwnode_handle *fwnode,
> > +                          void *fwopaque, u32 index,
> > +                          unsigned long *out_hartid);
> > +     u32 (*nr_mmio)(struct fwnode_handle *fwnode, void *fwopaque);
> > +     int (*mmio_to_resource)(struct fwnode_handle *fwnode,
> > +                             void *fwopaque, u32 index,
> > +                             struct resource *res);
> > +     void __iomem *(*mmio_map)(struct fwnode_handle *fwnode,
> > +                               void *fwopaque, u32 index);
> > +     int (*read_u32)(struct fwnode_handle *fwnode,
> > +                     void *fwopaque, const char *prop, u32 *out_val);
> > +     bool (*read_bool)(struct fwnode_handle *fwnode,
> > +                       void *fwopaque, const char *prop);
> > +};
>
> Why do we need this sort of (terrible) indirection?

Okay, I will replace this indirection with fwnode APIs.

>
> > +
> > +static int __init imsic_init(struct imsic_fwnode_ops *fwops,
> > +                          struct fwnode_handle *fwnode,
> > +                          void *fwopaque)
> > +{
> > +     struct resource res;
> > +     phys_addr_t base_addr;
> > +     int rc, nr_parent_irqs;
> > +     struct imsic_mmio *mmio;
> > +     struct imsic_priv *priv;
> > +     struct irq_domain *domain;
> > +     struct imsic_handler *handler;
> > +     struct imsic_global_config *global;
> > +     u32 i, tmp, nr_handlers = 0;
> > +
> > +     if (imsic_init_done) {
> > +             pr_err("%pfwP: already initialized hence ignoring\n",
> > +                     fwnode);
> > +             return -ENODEV;
> > +     }
> > +
> > +     if (!riscv_isa_extension_available(NULL, SxAIA)) {
> > +             pr_err("%pfwP: AIA support not available\n", fwnode);
> > +             return -ENODEV;
> > +     }
> > +
> > +     priv = kzalloc(sizeof(*priv), GFP_KERNEL);
> > +     if (!priv)
> > +             return -ENOMEM;
> > +     global = &priv->global;
> > +
> > +     /* Find number of parent interrupts */
> > +     nr_parent_irqs = fwops->nr_parent_irq(fwnode, fwopaque);
> > +     if (!nr_parent_irqs) {
> > +             pr_err("%pfwP: no parent irqs available\n", fwnode);
> > +             return -EINVAL;
> > +     }
> > +
> > +     /* Find number of guest index bits in MSI address */
> > +     rc = fwops->read_u32(fwnode, fwopaque, "riscv,guest-index-bits",
> > +                          &global->guest_index_bits);
> > +     if (rc)
> > +             global->guest_index_bits = 0;
> > +     tmp = BITS_PER_LONG - IMSIC_MMIO_PAGE_SHIFT;
> > +     if (tmp < global->guest_index_bits) {
> > +             pr_err("%pfwP: guest index bits too big\n", fwnode);
> > +             return -EINVAL;
> > +     }
> > +
> > +     /* Find number of HART index bits */
> > +     rc = fwops->read_u32(fwnode, fwopaque, "riscv,hart-index-bits",
> > +                          &global->hart_index_bits);
> > +     if (rc) {
> > +             /* Assume default value */
> > +             global->hart_index_bits = __fls(nr_parent_irqs);
> > +             if (BIT(global->hart_index_bits) < nr_parent_irqs)
> > +                     global->hart_index_bits++;
> > +     }
> > +     tmp = BITS_PER_LONG - IMSIC_MMIO_PAGE_SHIFT -
> > +           global->guest_index_bits;
> > +     if (tmp < global->hart_index_bits) {
> > +             pr_err("%pfwP: HART index bits too big\n", fwnode);
> > +             return -EINVAL;
> > +     }
> > +
> > +     /* Find number of group index bits */
> > +     rc = fwops->read_u32(fwnode, fwopaque, "riscv,group-index-bits",
> > +                          &global->group_index_bits);
> > +     if (rc)
> > +             global->group_index_bits = 0;
> > +     tmp = BITS_PER_LONG - IMSIC_MMIO_PAGE_SHIFT -
> > +           global->guest_index_bits - global->hart_index_bits;
> > +     if (tmp < global->group_index_bits) {
> > +             pr_err("%pfwP: group index bits too big\n", fwnode);
> > +             return -EINVAL;
> > +     }
> > +
> > +     /*
> > +      * Find first bit position of group index.
> > +      * If not specified assumed the default APLIC-IMSIC configuration.
> > +      */
> > +     rc = fwops->read_u32(fwnode, fwopaque, "riscv,group-index-shift",
> > +                          &global->group_index_shift);
> > +     if (rc)
> > +             global->group_index_shift = IMSIC_MMIO_PAGE_SHIFT * 2;
> > +     tmp = global->group_index_bits + global->group_index_shift - 1;
> > +     if (tmp >= BITS_PER_LONG) {
> > +             pr_err("%pfwP: group index shift too big\n", fwnode);
> > +             return -EINVAL;
> > +     }
> > +
> > +     /* Find number of interrupt identities */
> > +     rc = fwops->read_u32(fwnode, fwopaque, "riscv,num-ids",
> > +                          &global->nr_ids);
> > +     if (rc) {
> > +             pr_err("%pfwP: number of interrupt identities not found\n",
> > +                     fwnode);
> > +             return rc;
> > +     }
> > +     if ((global->nr_ids < IMSIC_MIN_ID) ||
> > +         (global->nr_ids >= IMSIC_MAX_ID) ||
> > +         ((global->nr_ids & IMSIC_MIN_ID) != IMSIC_MIN_ID)) {
> > +             pr_err("%pfwP: invalid number of interrupt identities\n",
> > +                     fwnode);
> > +             return -EINVAL;
> > +     }
> > +
> > +     /* Find number of guest interrupt identities */
> > +     if (fwops->read_u32(fwnode, fwopaque, "riscv,num-guest-ids",
> > +                         &global->nr_guest_ids))
> > +             global->nr_guest_ids = global->nr_ids;
> > +     if ((global->nr_guest_ids < IMSIC_MIN_ID) ||
> > +         (global->nr_guest_ids >= IMSIC_MAX_ID) ||
> > +         ((global->nr_guest_ids & IMSIC_MIN_ID) != IMSIC_MIN_ID)) {
> > +             pr_err("%pfwP: invalid number of guest interrupt identities\n",
> > +                     fwnode);
> > +             return -EINVAL;
> > +     }
>
> Please split the whole guest stuff out. It is totally unused!

The number of guest IDs is used by KVM RISC-V AIA support which
is in the pipeline. The KVM RISC-V only need imsic_get_global_config()
and imsic_get_local_config(). The "nr_guest_ids" is part of the
IMSIC global config.

>
> I've stopped reading. This needs structure, cleanups and a bit of
> taste. Not a lot of that here at the moment.
>
>         M.
>
> --
> Without deviation from the norm, progress is not possible.

It took a while to address all your comments and I also got
preempted by other stuff. Sorry for the delay.

Regards,
Anup

Re: [PATCH v2 5/9] irqchip: Add RISC-V incoming MSI controller driver

Posted by Marc Zyngier 2 years, 4 months ago

On Mon, 01 May 2023 09:28:16 +0100,
Anup Patel <anup@brainfault.org> wrote:
> 
> On Fri, Jan 13, 2023 at 3:40 PM Marc Zyngier <maz@kernel.org> wrote:
> >
> > On Tue, 03 Jan 2023 14:14:05 +0000,
> > Anup Patel <apatel@ventanamicro.com> wrote:
> > >
> > > The RISC-V advanced interrupt architecture (AIA) specification defines
> > > a new MSI controller for managing MSIs on a RISC-V platform. This new
> > > MSI controller is referred to as incoming message signaled interrupt
> > > controller (IMSIC) which manages MSI on per-HART (or per-CPU) basis.
> > > (For more details refer https://github.com/riscv/riscv-aia)
> >
> > And how about IPIs, which this driver seems to be concerned about?
> 
> Okay, I will mention about IPIs in the commit description.
> 
> >
> > >
> > > This patch adds an irqchip driver for RISC-V IMSIC found on RISC-V
> > > platforms.
> > >
> > > Signed-off-by: Anup Patel <apatel@ventanamicro.com>
> > > ---
> > >  drivers/irqchip/Kconfig             |   14 +-
> > >  drivers/irqchip/Makefile            |    1 +
> > >  drivers/irqchip/irq-riscv-imsic.c   | 1174 +++++++++++++++++++++++++++
> > >  include/linux/irqchip/riscv-imsic.h |   92 +++
> > >  4 files changed, 1280 insertions(+), 1 deletion(-)
> > >  create mode 100644 drivers/irqchip/irq-riscv-imsic.c
> > >  create mode 100644 include/linux/irqchip/riscv-imsic.h
> > >
> > > diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig
> > > index 9e65345ca3f6..a1315189a595 100644
> > > --- a/drivers/irqchip/Kconfig
> > > +++ b/drivers/irqchip/Kconfig
> > > @@ -29,7 +29,6 @@ config ARM_GIC_V2M
> > >
> > >  config GIC_NON_BANKED
> > >       bool
> > > -
> > >  config ARM_GIC_V3
> > >       bool
> > >       select IRQ_DOMAIN_HIERARCHY
> > > @@ -548,6 +547,19 @@ config SIFIVE_PLIC
> > >       select IRQ_DOMAIN_HIERARCHY
> > >       select GENERIC_IRQ_EFFECTIVE_AFF_MASK if SMP
> > >
> > > +config RISCV_IMSIC
> > > +     bool
> > > +     depends on RISCV
> > > +     select IRQ_DOMAIN_HIERARCHY
> > > +     select GENERIC_MSI_IRQ_DOMAIN
> > > +
> > > +config RISCV_IMSIC_PCI
> > > +     bool
> > > +     depends on RISCV_IMSIC
> > > +     depends on PCI
> > > +     depends on PCI_MSI
> > > +     default RISCV_IMSIC
> >
> > This should definitely tell you that this driver needs splitting.
> 
> The code under "#ifdef CONFIG_RISCV_IMSIC_PCI" is hardly 40 lines
> so I felt it was too small to deserve its own source file.

It at least needs its own patch.

> 
> >
> > > +
> > >  config EXYNOS_IRQ_COMBINER
> > >       bool "Samsung Exynos IRQ combiner support" if COMPILE_TEST
> > >       depends on (ARCH_EXYNOS && ARM) || COMPILE_TEST
> > > diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile
> > > index 87b49a10962c..22c723cc6ec8 100644
> > > --- a/drivers/irqchip/Makefile
> > > +++ b/drivers/irqchip/Makefile
> > > @@ -96,6 +96,7 @@ obj-$(CONFIG_QCOM_MPM)                      += irq-qcom-mpm.o
> > >  obj-$(CONFIG_CSKY_MPINTC)            += irq-csky-mpintc.o
> > >  obj-$(CONFIG_CSKY_APB_INTC)          += irq-csky-apb-intc.o
> > >  obj-$(CONFIG_RISCV_INTC)             += irq-riscv-intc.o
> > > +obj-$(CONFIG_RISCV_IMSIC)            += irq-riscv-imsic.o
> > >  obj-$(CONFIG_SIFIVE_PLIC)            += irq-sifive-plic.o
> > >  obj-$(CONFIG_IMX_IRQSTEER)           += irq-imx-irqsteer.o
> > >  obj-$(CONFIG_IMX_INTMUX)             += irq-imx-intmux.o
> > > diff --git a/drivers/irqchip/irq-riscv-imsic.c b/drivers/irqchip/irq-riscv-imsic.c
> > > new file mode 100644
> > > index 000000000000..4c16b66738d6
> > > --- /dev/null
> > > +++ b/drivers/irqchip/irq-riscv-imsic.c
> > > @@ -0,0 +1,1174 @@
> > > +// SPDX-License-Identifier: GPL-2.0
> > > +/*
> > > + * Copyright (C) 2021 Western Digital Corporation or its affiliates.
> > > + * Copyright (C) 2022 Ventana Micro Systems Inc.
> > > + */
> > > +
> > > +#define pr_fmt(fmt) "riscv-imsic: " fmt
> > > +#include <linux/bitmap.h>
> > > +#include <linux/cpu.h>
> > > +#include <linux/interrupt.h>
> > > +#include <linux/io.h>
> > > +#include <linux/iommu.h>
> > > +#include <linux/irq.h>
> > > +#include <linux/irqchip.h>
> > > +#include <linux/irqchip/chained_irq.h>
> > > +#include <linux/irqchip/riscv-imsic.h>
> > > +#include <linux/irqdomain.h>
> > > +#include <linux/module.h>
> > > +#include <linux/msi.h>
> > > +#include <linux/of.h>
> > > +#include <linux/of_address.h>
> > > +#include <linux/of_irq.h>
> > > +#include <linux/pci.h>
> > > +#include <linux/platform_device.h>
> > > +#include <linux/spinlock.h>
> > > +#include <linux/smp.h>
> > > +#include <asm/hwcap.h>
> > > +
> > > +#define IMSIC_DISABLE_EIDELIVERY     0
> > > +#define IMSIC_ENABLE_EIDELIVERY              1
> > > +#define IMSIC_DISABLE_EITHRESHOLD    1
> > > +#define IMSIC_ENABLE_EITHRESHOLD     0
> > > +
> > > +#define imsic_csr_write(__c, __v)    \
> > > +do {                                 \
> > > +     csr_write(CSR_ISELECT, __c);    \
> > > +     csr_write(CSR_IREG, __v);       \
> > > +} while (0)
> > > +
> > > +#define imsic_csr_read(__c)          \
> > > +({                                   \
> > > +     unsigned long __v;              \
> > > +     csr_write(CSR_ISELECT, __c);    \
> > > +     __v = csr_read(CSR_IREG);       \
> > > +     __v;                            \
> > > +})
> > > +
> > > +#define imsic_csr_set(__c, __v)              \
> > > +do {                                 \
> > > +     csr_write(CSR_ISELECT, __c);    \
> > > +     csr_set(CSR_IREG, __v);         \
> > > +} while (0)
> > > +
> > > +#define imsic_csr_clear(__c, __v)    \
> > > +do {                                 \
> > > +     csr_write(CSR_ISELECT, __c);    \
> > > +     csr_clear(CSR_IREG, __v);       \
> > > +} while (0)
> > > +
> > > +struct imsic_mmio {
> > > +     phys_addr_t pa;
> > > +     void __iomem *va;
> > > +     unsigned long size;
> > > +};
> > > +
> > > +struct imsic_priv {
> > > +     /* Global configuration common for all HARTs */
> > > +     struct imsic_global_config global;
> > > +
> > > +     /* MMIO regions */
> > > +     u32 num_mmios;
> > > +     struct imsic_mmio *mmios;
> > > +
> > > +     /* Global state of interrupt identities */
> > > +     raw_spinlock_t ids_lock;
> > > +     unsigned long *ids_used_bimap;
> > > +     unsigned long *ids_enabled_bimap;
> > > +     unsigned int *ids_target_cpu;
> > > +
> > > +     /* Mask for connected CPUs */
> > > +     struct cpumask lmask;
> > > +
> > > +     /* IPI interrupt identity */
> > > +     u32 ipi_id;
> > > +     u32 ipi_lsync_id;
> > > +
> > > +     /* IRQ domains */
> > > +     struct irq_domain *base_domain;
> > > +     struct irq_domain *pci_domain;
> > > +     struct irq_domain *plat_domain;
> > > +};
> > > +
> > > +struct imsic_handler {
> > > +     /* Local configuration for given HART */
> > > +     struct imsic_local_config local;
> > > +
> > > +     /* Pointer to private context */
> > > +     struct imsic_priv *priv;
> > > +};
> > > +
> > > +static bool imsic_init_done;
> > > +
> > > +static int imsic_parent_irq;
> > > +static DEFINE_PER_CPU(struct imsic_handler, imsic_handlers);
> > > +
> > > +const struct imsic_global_config *imsic_get_global_config(void)
> > > +{
> > > +     struct imsic_handler *handler = this_cpu_ptr(&imsic_handlers);
> > > +
> > > +     if (!handler || !handler->priv)
> > > +             return NULL;
> > > +
> > > +     return &handler->priv->global;
> > > +}
> > > +EXPORT_SYMBOL_GPL(imsic_get_global_config);
> > > +
> > > +const struct imsic_local_config *imsic_get_local_config(unsigned int cpu)
> > > +{
> > > +     struct imsic_handler *handler = per_cpu_ptr(&imsic_handlers, cpu);
> > > +
> > > +     if (!handler || !handler->priv)
> > > +             return NULL;
> >
> > How can this happen?
> 
> These are redundant checks. I will drop.
> 
> >
> > > +
> > > +     return &handler->local;
> > > +}
> > > +EXPORT_SYMBOL_GPL(imsic_get_local_config);
> >
> > Why are these symbols exported? They have no user, so they shouldn't
> > even exist here. I also seriously doubt there is a valid use case for
> > exposing this information to the rest of the kernel.
> 
> The imsic_get_global_config() is used by APLIC driver and KVM RISC-V
> module whereas imsic_get_local_config() is only used by KVM RISC-V.
> 
> The KVM RISC-V AIA irqchip patches are available in riscv_kvm_aia_v1
> branch at: https://github.com/avpatel/linux.git. I have not posted KVM RISC-V
> patches due various interdependencies.

Then the symbols can wait, can't they? It'd make more sense if the
KVM-dependent bits were brought together with the KVM patches.

Even better, you'd use some level of abstraction between KVM and the
irqchip code. GIC makes some heavy use of irq_set_vcpu_affinity() as a
private API with KVM, and I'd suggest you look into something similar.

[...]

> > > +#ifdef CONFIG_SMP
> > > +static void __imsic_id_smp_sync(struct imsic_priv *priv)
> > > +{
> > > +     struct imsic_handler *handler;
> > > +     struct cpumask amask;
> > > +     int cpu;
> > > +
> > > +     cpumask_and(&amask, &priv->lmask, cpu_online_mask);
> >
> > Can't this race against a CPU going down?
> 
> Yes, it can race if a CPU goes down while we are in this function
> but this won't be a problem because the imsic_starting_cpu()
> will unconditionally do imsic_ids_local_sync() when the CPU is
> brought-up again. I will add a multiline comment block explaining
> this.

I'd rather you avoid the race instead of papering over it.

> 
> >
> > > +     for_each_cpu(cpu, &amask) {
> > > +             if (cpu == smp_processor_id())
> > > +                     continue;
> > > +
> > > +             handler = per_cpu_ptr(&imsic_handlers, cpu);
> > > +             if (!handler || !handler->priv || !handler->local.msi_va) {
> > > +                     pr_warn("CPU%d: handler not initialized\n", cpu);
> >
> > How many times are you going to do that? On each failing synchronisation?
> 
> My bad for adding these paranoid checks. I remove these checks
> wherever possible.
> 
> >
> > > +                     continue;
> > > +             }
> > > +
> > > +             writel(handler->priv->ipi_lsync_id, handler->local.msi_va);
> >
> > As I understand it, this is a "behind the scenes" IPI. Why isn't that
> > a *real* IPI?
> 
> Yes, that's correct. The ID enable bits are per-CPU accessible only
> via CSRs hence we have a special "behind the scenes" IPI to
> synchronize state of ID enable bits.

My question still stands: why isn't this a *real*, Linux visible IPI?
This sideband signalling makes everything hard to follow, hard to
debug, and screws up accounting.

> > Please split the whole guest stuff out. It is totally unused!
> 
> The number of guest IDs is used by KVM RISC-V AIA support which
> is in the pipeline. The KVM RISC-V only need imsic_get_global_config()
> and imsic_get_local_config(). The "nr_guest_ids" is part of the
> IMSIC global config.

And yet it isn't needed for a minimal driver, which what I'd like to
see at first. Shoving the kitchen sink into an initial patch isn't a
great way to get it merged.

      M.

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