Re: [PATCH 06/13] irqchip: Add driver for Loongson-3 HyperTransport interrupt controller

[Marc Zyngier <maz@kernel.org>]

On 27/08/2019 09:52, Jiaxun Yang wrote:
> This controller appeared on Loongson-3 family of chips to receive interrupts
> from PCH chip.
> 
> Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
> ---
>  drivers/irqchip/Kconfig          |   8 ++
>  drivers/irqchip/Makefile         |   1 +
>  drivers/irqchip/irq-ls3-htintc.c | 145 +++++++++++++++++++++++++++++++
>  3 files changed, 154 insertions(+)
>  create mode 100644 drivers/irqchip/irq-ls3-htintc.c
> 
> diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig
> index 8d9eac5fd4a7..b3ce0f3e43ae 100644
> --- a/drivers/irqchip/Kconfig
> +++ b/drivers/irqchip/Kconfig
> @@ -480,6 +480,14 @@ config LS3_IOINTC
>  	help
>  	  Support for the Loongson-3 I/O Interrupt Controller.
>  
> +config LS3_HTINTC
> +	bool "Loongson3 HyperTransport Interrupt Controller"
> +	depends on MACH_LOONGSON64
> +	default y
> +	select IRQ_DOMAIN
> +	select GENERIC_IRQ_CHIP
> +	help
> +	  Support for the Loongson-3 HyperTransport Interrupt Controller.
>  endmenu
>  
>  config SIFIVE_PLIC
> diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile
> index 49ecb8d38138..0fda94c319e9 100644
> --- a/drivers/irqchip/Makefile
> +++ b/drivers/irqchip/Makefile
> @@ -103,3 +103,4 @@ obj-$(CONFIG_LS1X_IRQ)			+= irq-ls1x.o
>  obj-$(CONFIG_TI_SCI_INTR_IRQCHIP)	+= irq-ti-sci-intr.o
>  obj-$(CONFIG_TI_SCI_INTA_IRQCHIP)	+= irq-ti-sci-inta.o
>  obj-$(CONFIG_LS3_IOINTC)			+= irq-ls3-iointc.o
> +obj-$(CONFIG_LS3_HTINTC)			+= irq-ls3-htintc.o
> diff --git a/drivers/irqchip/irq-ls3-htintc.c b/drivers/irqchip/irq-ls3-htintc.c
> new file mode 100644
> index 000000000000..c53bbb0bd78c
> --- /dev/null
> +++ b/drivers/irqchip/irq-ls3-htintc.c
> @@ -0,0 +1,145 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + *  Copyright (C) 2019, Jiaxun Yang <jiaxun.yang@flygoat.com>
> + *  Loongson-3 HyperTransport IRQ support
> + */
> +
> +#include <linux/errno.h>
> +#include <linux/init.h>
> +#include <linux/types.h>
> +#include <linux/interrupt.h>
> +#include <linux/ioport.h>
> +#include <linux/irqchip.h>
> +#include <linux/of_address.h>
> +#include <linux/of_irq.h>
> +#include <linux/io.h>
> +#include <linux/irqchip/chained_irq.h>
> +
> +#define HTINTC_NUM_GC  4
> +#define HTINTC_GC_SIZE  0x4
> +#define HTINTC_NUM_HANDLER  4
> +#define HTINTC_HANDLER_SIZE  0x8
> +#define HTINTC_HANDLER_IRQ  64
> +
> +#define HTINTC_VECTOR_OFFSET   0x0
> +#define HTINTC_EN_OFFSET   0x20
> +
> +struct htintc_handler_priv {
> +	struct irq_domain	*domain;
> +	void __iomem		*handler_base;
> +};
> +
> +static void htintc_chained_handle_irq(struct irq_desc *desc)
> +{
> +	struct htintc_handler_priv *priv = irq_desc_get_handler_data(desc);
> +	struct irq_chip *chip = irq_desc_get_chip(desc);
> +	int i;
> +	bool handled;
> +
> +	chained_irq_enter(chip, desc);
> +
> +	for (i = 0; i < HTINTC_NUM_GC; i++) {
> +		uint32_t irqs = readl(priv->handler_base + HTINTC_GC_SIZE * i);
> +
> +		while (irqs) {
> +			int bit = __ffs(irqs);
> +
> +			generic_handle_irq(irq_find_mapping(priv->domain, bit + 32 * i));
> +			irqs &= ~BIT(bit);
> +			handled = true;
> +		}
> +	}
> +
> +	if (!handled)

Uninitialized variable. I'm pretty sure your compiler warns about that.

> +		spurious_interrupt();
> +
> +	chained_irq_exit(chip, desc);
> +}
> +
> +int __init ls3_htintc_of_init(struct device_node *node,
> +					   struct device_node *parent)
> +{
> +	struct irq_chip_generic *gc;
> +	struct irq_chip_type *ct;
> +	struct htintc_handler_priv *priv;
> +	struct irq_domain *domain;
> +	void __iomem *base;
> +	int parent_irq[HTINTC_NUM_HANDLER], err = 0;
> +	int i;
> +
> +

Spurious empty line.

> +	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
> +	if (!priv)
> +		return -ENOMEM;
> +
> +	base = of_iomap(node, 0);
> +	if (!base) {
> +		err = -ENODEV;
> +		goto out_free_priv;
> +	}
> +
> +	for (i = 0; i < HTINTC_NUM_HANDLER; i++) {
> +		parent_irq[i] = irq_of_parse_and_map(node, i);
> +		if (!parent_irq[i]) {
> +			pr_err("ls3-htintc: unable to get parent irq %d\n", i);
> +			err =  -ENODEV;
> +			goto out_iounmap;
> +		}
> +	}
> +	/* Set up an IRQ domain */
> +	domain = irq_domain_add_linear(node, 32 * HTINTC_NUM_GC,
> +		&irq_generic_chip_ops, NULL);

Please align the second line on the opening bracket from the previous line.

> +	if (!domain) {
> +		pr_err("ls3-htintc: cannot add IRQ domain\n");
> +		err = -ENOMEM;
> +		goto out_iounmap;
> +	}
> +
> +	for (i = 0; i < HTINTC_NUM_HANDLER; i++) {
> +		/* Mask all interrupts */
> +		writeq(0x0, base + HTINTC_EN_OFFSET + HTINTC_HANDLER_SIZE * i);
> +	}
> +
> +	err = irq_alloc_domain_generic_chips(domain, 32, 1,
> +		node->full_name, handle_fasteoi_irq,

handle_fast_eoi? Where is your irq_eoi callback? That looks wrong given
that you're using a mask/ack/unmask sequence.

> +		IRQ_NOPROBE, 0, IRQ_GC_INIT_MASK_CACHE);
> +	if (err) {
> +		pr_err("ls3-htintc: unable to register IRQ domain\n");
> +		err = -ENOMEM;
> +		goto out_free_domain;
> +	}
> +
> +	for (i = 0; i < HTINTC_NUM_GC; i++) {
> +		gc = irq_get_domain_generic_chip(domain, i * 32);
> +		gc->reg_base = base + HTINTC_GC_SIZE * i;
> +		gc->domain = domain;
> +
> +		ct = gc->chip_types;
> +		ct->regs.mask = HTINTC_EN_OFFSET;
> +		ct->regs.ack = HTINTC_VECTOR_OFFSET;
> +		ct->chip.irq_unmask = irq_gc_mask_set_bit;
> +		ct->chip.irq_mask = irq_gc_mask_clr_bit;
> +		ct->chip.irq_ack = irq_gc_ack_set_bit;
> +	}
> +
> +	priv->domain = domain;
> +	priv->handler_base = base + HTINTC_VECTOR_OFFSET;
> +
> +	for (i = 0; i < HTINTC_NUM_HANDLER; i++) {
> +		irq_set_chained_handler_and_data(parent_irq[i],
> +			htintc_chained_handle_irq, priv);
> +	}
> +
> +	return 0;
> +
> +out_free_domain:
> +	irq_domain_remove(domain);
> +out_iounmap:
> +	iounmap(base);
> +out_free_priv:
> +	kfree(priv);
> +
> +	return err;
> +}
> +
> +IRQCHIP_DECLARE(ls3_htintc, "loongson,ls3-htintc", ls3_htintc_of_init);
> 

	M.
-- 
Jazz is not dead, it just smells funny...