Re: [PATCH v2 5/6] irqchip: Add Loongson PCH MSI controller

From: Marc Zyngier
Date: Tue Apr 28 2020 - 09:07:47 EST


On Tue, 28 Apr 2020 14:32:44 +0800
Jiaxun Yang <jiaxun.yang@xxxxxxxxxxx> wrote:

> This controller appears on Loongson LS7A family of PCH to transform
> interrupts from PCI MSI into HyperTransport vectorized interrrupts
> and send them to procrssor's HT vector controller.
>
> Signed-off-by: Jiaxun Yang <jiaxun.yang@xxxxxxxxxxx>
> --
> v2:
> - Style clean-ups
> - Add ack callback
> - Use bitmap_find_free_region
> ---
> drivers/irqchip/Kconfig | 10 +
> drivers/irqchip/Makefile | 1 +
> drivers/irqchip/irq-loongson-pch-msi.c | 259 +++++++++++++++++++++++++
> 3 files changed, 270 insertions(+)
> create mode 100644 drivers/irqchip/irq-loongson-pch-msi.c
>
> diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig
> index 5524a621638c..0b6b826dd843 100644
> --- a/drivers/irqchip/Kconfig
> +++ b/drivers/irqchip/Kconfig
> @@ -549,4 +549,14 @@ config LOONGSON_PCH_PIC
> help
> Support for the Loongson PCH PIC Controller.
>
> +config LOONGSON_PCH_MSI
> + bool "Loongson PCH PIC Controller"
> + depends on MACH_LOONGSON64 || COMPILE_TEST
> + depends on PCI
> + default MACH_LOONGSON64
> + select IRQ_DOMAIN_HIERARCHY
> + select PCI_MSI
> + help
> + Support for the Loongson PCH MSI Controller.
> +
> endmenu
> diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile
> index acc72331cec8..3a4ce283189a 100644
> --- a/drivers/irqchip/Makefile
> +++ b/drivers/irqchip/Makefile
> @@ -109,3 +109,4 @@ obj-$(CONFIG_LOONGSON_LIOINTC) += irq-loongson-liointc.o
> obj-$(CONFIG_LOONGSON_HTPIC) += irq-loongson-htpic.o
> obj-$(CONFIG_LOONGSON_HTVEC) += irq-loongson-htvec.o
> obj-$(CONFIG_LOONGSON_PCH_PIC) += irq-loongson-pch-pic.o
> +obj-$(CONFIG_LOONGSON_PCH_MSI) += irq-loongson-pch-msi.o
> diff --git a/drivers/irqchip/irq-loongson-pch-msi.c b/drivers/irqchip/irq-loongson-pch-msi.c
> new file mode 100644
> index 000000000000..5b4d607a899e
> --- /dev/null
> +++ b/drivers/irqchip/irq-loongson-pch-msi.c
> @@ -0,0 +1,259 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Copyright (C) 2020, Jiaxun Yang <jiaxun.yang@xxxxxxxxxxx>
> + * Loongson PCH MSI support
> + */
> +
> +#define pr_fmt(fmt) "pch-msi: " fmt
> +
> +#include <linux/irqchip.h>
> +#include <linux/msi.h>
> +#include <linux/of.h>
> +#include <linux/of_address.h>
> +#include <linux/of_irq.h>
> +#include <linux/of_pci.h>
> +#include <linux/pci.h>
> +#include <linux/slab.h>
> +
> +struct pch_msi_data {
> + spinlock_t msi_map_lock;
> + phys_addr_t doorbell;
> + u32 irq_first; /* The vector number that MSIs starts */
> + u32 num_irqs; /* The number of vectors for MSIs */
> + unsigned long *msi_map;
> +};
> +
> +static void pch_msi_mask_msi_irq(struct irq_data *d)
> +{
> + pci_msi_mask_irq(d);
> + irq_chip_mask_parent(d);
> +}
> +
> +static void pch_msi_unmask_msi_irq(struct irq_data *d)
> +{
> + pci_msi_unmask_irq(d);
> + irq_chip_unmask_parent(d);
> +}
> +
> +static struct irq_chip pch_msi_irq_chip = {
> + .name = "PCH MSI",
> + .irq_mask = pch_msi_mask_msi_irq,
> + .irq_unmask = pch_msi_unmask_msi_irq,
> + .irq_ack = irq_chip_ack_parent,
> + .irq_set_affinity = irq_chip_set_affinity_parent,
> +};
> +
> +static int pch_msi_allocate_hwirq(struct pch_msi_data *priv, int num_req)
> +{
> + int first;
> +
> + spin_lock(&priv->msi_map_lock);

Why does it have to be a spinlock? As far as I can tell, we never
allocate MSIs from non-preemptible contexts.

> +
> + first = bitmap_find_free_region(priv->msi_map, priv->num_irqs,
> + get_count_order(num_req));
> + if (first < 0) {
> + spin_unlock(&priv->msi_map_lock);
> + return -ENOSPC;
> + }
> +
> + bitmap_set(priv->msi_map, first, num_req);

What is that for? bitmap_find_free_region has already done the work for
you.

> + spin_unlock(&priv->msi_map_lock);
> +
> + return priv->irq_first + first;
> +}
> +
> +static void pch_msi_free_hwirq(struct pch_msi_data *priv,
> + int hwirq, int num_req)
> +{
> + int first = hwirq - priv->irq_first;
> +
> + spin_lock(&priv->msi_map_lock);
> + bitmap_clear(priv->msi_map, first, num_req);

bitmap_clear doesn't reverse the effects of bitmap_find_free_region.

> + spin_unlock(&priv->msi_map_lock);
> +}
> +
> +static void pch_msi_compose_msi_msg(struct irq_data *data,
> + struct msi_msg *msg)
> +{
> + struct pch_msi_data *priv = irq_data_get_irq_chip_data(data);
> +
> + msg->address_hi = upper_32_bits(priv->doorbell);
> + msg->address_lo = lower_32_bits(priv->doorbell);
> + msg->data = data->hwirq;
> +}
> +
> +static struct msi_domain_info pch_msi_domain_info = {
> + .flags = MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS |
> + MSI_FLAG_MULTI_PCI_MSI | MSI_FLAG_PCI_MSIX,
> + .chip = &pch_msi_irq_chip,
> +};
> +
> +static struct irq_chip middle_irq_chip = {
> + .name = "PCH MSI Middle",

This "middle" thing doesn't mean anything. What it really implements is
a generic MSI irqchip. What you call PCH MSI seems to be PCI-MSI. Pretty
confusing.

> + .irq_mask = irq_chip_mask_parent,
> + .irq_unmask = irq_chip_unmask_parent,
> + .irq_ack = irq_chip_ack_parent,
> + .irq_set_affinity = irq_chip_set_affinity_parent,
> + .irq_compose_msi_msg = pch_msi_compose_msi_msg,
> +};
> +
> +static int pch_msi_parent_domain_alloc(struct irq_domain *domain,
> + unsigned int virq, int hwirq)
> +{
> + struct irq_fwspec fwspec;
> + int ret;
> +
> + fwspec.fwnode = domain->parent->fwnode;
> + fwspec.param_count = 1;
> + fwspec.param[0] = hwirq;
> +
> + ret = irq_domain_alloc_irqs_parent(domain, virq, 1, &fwspec);
> + if (ret)
> + return ret;
> +
> + irq_domain_set_hwirq_and_chip(domain, virq, hwirq,
> + &middle_irq_chip, NULL);
> +
> + return 0;
> +}
> +
> +static int pch_msi_middle_domain_alloc(struct irq_domain *domain,
> + unsigned int virq,
> + unsigned int nr_irqs, void *args)
> +{
> + struct pch_msi_data *priv = domain->host_data;
> + int hwirq, err, i;
> +
> + hwirq = pch_msi_allocate_hwirq(priv, nr_irqs);
> + if (hwirq < 0)
> + return hwirq;
> +
> + for (i = 0; i < nr_irqs; i++) {
> + err = pch_msi_parent_domain_alloc(domain, virq + i, hwirq + i);
> + if (err)
> + goto err_hwirq;
> +
> + irq_domain_set_hwirq_and_chip(domain, virq + i, hwirq + i,
> + &middle_irq_chip, priv);

So you're doing that twice per MSI. I think once is enough.

> + }
> +
> + return 0;
> +
> +err_hwirq:
> + while (--i >= 0)
> + irq_domain_free_irqs_parent(domain, virq, i);

This looks very wrong. Why isn't it just:

irq_domain_free_irqs_parent(domain, virq, i - 1);

?

> +
> + pch_msi_free_hwirq(priv, hwirq, nr_irqs);

And when you look at the whole error handling (once fixed), it is
exactly pch_msi_middle_domain_free(). So why not calling this directly?

> + return err;
> +}
> +
> +static void pch_msi_middle_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 pch_msi_data *priv = irq_data_get_irq_chip_data(d);
> +
> + irq_domain_free_irqs_parent(domain, virq, nr_irqs);
> + pch_msi_free_hwirq(priv, d->hwirq, nr_irqs);
> +}
> +
> +static const struct irq_domain_ops pch_msi_middle_domain_ops = {
> + .alloc = pch_msi_middle_domain_alloc,
> + .free = pch_msi_middle_domain_free,
> +};
> +
> +static int pch_msi_init_domains(struct pch_msi_data *priv,
> + struct device_node *node,
> + struct device_node *parent)
> +{
> + struct irq_domain *middle_domain, *msi_domain,
> *parent_domain; +
> + parent_domain = irq_find_host(parent);
> + if (!parent_domain) {
> + pr_err("Failed to find the parent domain\n");
> + return -ENXIO;
> + }

Can't you check this early in the probe routine and bail out?

> +
> + middle_domain = irq_domain_add_linear(NULL, priv->num_irqs,
> + &pch_msi_middle_domain_ops,
> + priv);

NULL isn't an acceptable parameter. This irqdomain really belongs to
the node. See irq_domain_update_bus_token() to specialize the domain so
that its allocation doesn't clash with the PCI domain that you allocate
below (DOMAIN_BUS_NEXUS is used by quite a few drivers).

Also, please use irq_domain_create_linear instead, in order
to be consistent with the rest of the API usage in this file.

> + if (!middle_domain) {
> + pr_err("Failed to create the MSI middle domain\n");
> + return -ENOMEM;
> + }
> +
> + middle_domain->parent = parent_domain;
> +
> + msi_domain = pci_msi_create_irq_domain(of_node_to_fwnode(node),
> + &pch_msi_domain_info,
> + middle_domain);
> + if (!msi_domain) {
> + pr_err("Failed to create MSI domain\n");
> + irq_domain_remove(middle_domain);
> + return -ENOMEM;
> + }
> +
> + return 0;
> +}
> +
> +static int pch_msi_init(struct device_node *node,
> + struct device_node *parent)
> +{
> + struct pch_msi_data *priv;
> + struct resource res;
> + int ret;
> +
> + priv = kzalloc(sizeof(*priv), GFP_KERNEL);
> + if (!priv)
> + return -ENOMEM;
> +
> + spin_lock_init(&priv->msi_map_lock);
> +
> + ret = of_address_to_resource(node, 0, &res);
> + if (ret) {
> + pr_err("Failed to allocate resource\n");
> + goto err_priv;
> + }
> +
> + priv->doorbell = res.start;
> +
> + if (of_property_read_u32(node, "loongson,msi-base-vec",
> + &priv->irq_first)) {
> + pr_err("Unable to parse MSI vec base\n");
> + ret = -EINVAL;
> + goto err_priv;
> + }
> +
> + if (of_property_read_u32(node, "loongson,msi-num-vecs",
> + &priv->num_irqs)) {
> + pr_err("Unable to parse MSI vec number\n");
> + ret = -EINVAL;
> + goto err_priv;
> + }
> +
> + priv->msi_map = kcalloc(BITS_TO_LONGS(priv->num_irqs),
> + sizeof(*priv->msi_map),
> + GFP_KERNEL);

We have bitmap_alloc() that should already do the right thing.

> + if (!priv->msi_map) {
> + ret = -ENOMEM;
> + goto err_priv;
> + }
> +
> + pr_debug("Registering %d MSIs, starting at %d\n",
> + priv->num_irqs, priv->irq_first);
> +
> + ret = pch_msi_init_domains(priv, node, parent);
> + if (ret)
> + goto err_map;
> +
> + return 0;
> +
> +err_map:
> + kfree(priv->msi_map);
> +err_priv:
> + kfree(priv);
> + return ret;
> +}
> +
> +IRQCHIP_DECLARE(pch_msi, "loongson,pch-msi-1.0", pch_msi_init);


Thanks,

M.
--
Jazz is not dead. It just smells funny...