From: Marc Zyngier <marc.zyngier@arm.com>
To: Phil Elwell <phil@raspberrypi.org>,
Thomas Gleixner <tglx@linutronix.de>,
Jason Cooper <jason@lakedaemon.net>,
Rob Herring <robh+dt@kernel.org>,
Mark Rutland <mark.rutland@arm.com>,
Florian Fainelli <f.fainelli@gmail.com>,
Stefan Wahren <stefan.wahren@i2se.com>,
Eric Anholt <eric@anholt.net>,
Russell King <linux@armlinux.org.uk>,
Michael Turquette <mturquette@baylibre.com>,
Stephen Boyd <sboyd@codeaurora.org>,
devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-rpi-kernel@lists.infradead.org, linux-clk@vger.kernel.org
Subject: Re: [PATCH v2 3/4] irqchip: Add BCM2835 AUX interrupt controller
Date: Mon, 12 Jun 2017 15:59:46 +0100 [thread overview]
Message-ID: <56e314db-49a9-48fa-d7d9-877bd68aadfc@arm.com> (raw)
In-Reply-To: <c5ceca4d-81d1-9faf-bc4a-3ea3113abf36@raspberrypi.org>
On 12/06/17 15:25, Phil Elwell wrote:
> Devices in the BCM2835 AUX block share a common interrupt line, with a
> register indicating which devices have active IRQs. Expose this as a
> nested interrupt controller to avoid IRQ sharing problems (easily
> observed if UART1 and SPI1/2 are enabled simultaneously).
>
> Signed-off-by: Phil Elwell <phil@raspberrypi.org>
> ---
> drivers/irqchip/Makefile | 2 +-
> drivers/irqchip/irq-bcm2835-aux.c | 155 ++++++++++++++++++++++++++++++++++++++
> 2 files changed, 156 insertions(+), 1 deletion(-)
> create mode 100644 drivers/irqchip/irq-bcm2835-aux.c
>
> diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile
> index b64c59b..cf01920 100644
> --- a/drivers/irqchip/Makefile
> +++ b/drivers/irqchip/Makefile
> @@ -3,7 +3,7 @@ obj-$(CONFIG_IRQCHIP) += irqchip.o
> obj-$(CONFIG_ALPINE_MSI) += irq-alpine-msi.o
> obj-$(CONFIG_ATH79) += irq-ath79-cpu.o
> obj-$(CONFIG_ATH79) += irq-ath79-misc.o
> -obj-$(CONFIG_ARCH_BCM2835) += irq-bcm2835.o
> +obj-$(CONFIG_ARCH_BCM2835) += irq-bcm2835.o irq-bcm2835-aux.o
> obj-$(CONFIG_ARCH_BCM2835) += irq-bcm2836.o
> obj-$(CONFIG_ARCH_EXYNOS) += exynos-combiner.o
> obj-$(CONFIG_FARADAY_FTINTC010) += irq-ftintc010.o
> diff --git a/drivers/irqchip/irq-bcm2835-aux.c b/drivers/irqchip/irq-bcm2835-aux.c
> new file mode 100644
> index 0000000..545f12e
> --- /dev/null
> +++ b/drivers/irqchip/irq-bcm2835-aux.c
> @@ -0,0 +1,155 @@
> +/*
> + * Copyright (C) 2017 Raspberry Pi (Trading) Ltd.
> + *
> + * This program is free software; you can redistribute it and/or modify it
> + * under the terms and conditions of the GNU General Public License,
> + * version 2, as published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope it will be useful, but WITHOUT
> + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
> + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
> + * more details.
> + *
> + * You should have received a copy of the GNU General Public License
> + * along with this program. If not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#include <linux/interrupt.h>
> +#include <linux/irqdomain.h>
> +#include <linux/module.h>
> +#include <linux/of_irq.h>
> +#include <linux/platform_device.h>
> +#include <dt-bindings/interrupt-controller/bcm2835-aux-intc.h>
> +
> +#define BCM2835_AUXIRQ 0x00
> +
> +#define BCM2835_AUX_IRQ_UART_MASK BIT(BCM2835_AUX_IRQ_UART)
> +#define BCM2835_AUX_IRQ_SPI1_MASK BIT(BCM2835_AUX_IRQ_SPI1)
> +#define BCM2835_AUX_IRQ_SPI2_MASK BIT(BCM2835_AUX_IRQ_SPI2)
> +
> +#define BCM2835_AUX_IRQ_ALL_MASK \
> + (BCM2835_AUX_IRQ_UART_MASK | \
> + BCM2835_AUX_IRQ_SPI1_MASK | \
> + BCM2835_AUX_IRQ_SPI2_MASK)
> +
> +struct aux_irq_state {
> + void __iomem *status;
> + struct irq_domain *domain;
> +};
> +
> +static struct aux_irq_state aux_irq __read_mostly;
> +
> +static irqreturn_t bcm2835_aux_irq_handler(int irq, void *dev_id)
> +{
> + u32 stat = readl_relaxed(aux_irq.status);
> +
> + if (stat & BCM2835_AUX_IRQ_UART_MASK)
> + generic_handle_irq(irq_linear_revmap(aux_irq.domain,
> + BCM2835_AUX_IRQ_UART));
> +
> + if (stat & BCM2835_AUX_IRQ_SPI1_MASK)
> + generic_handle_irq(irq_linear_revmap(aux_irq.domain,
> + BCM2835_AUX_IRQ_SPI1));
> +
> + if (stat & BCM2835_AUX_IRQ_SPI2_MASK)
> + generic_handle_irq(irq_linear_revmap(aux_irq.domain,
> + BCM2835_AUX_IRQ_SPI2));
> +
> + return (stat & BCM2835_AUX_IRQ_ALL_MASK) ? IRQ_HANDLED : IRQ_NONE;
I could understand the use of a normal interrupt handler instead of a
chained handler, as the HW doesn't have any way of masking interrupts
(whoever designed this should be forced to fix each and every SoC with a
magnifier and a tiny drill) if it wasn't for this last line.
Here, you're making sure that you always return IRQ_HANDLED if something
was pending, irrespective of whether it has been handled or not. How do
you recover when you have a screaming interrupt and no handler?
Why don't you simply request the interrupt as a shared one, and check
for the state in the handlers themselves? This way, the kernel will be
able to recover from a screaming interrupt by disabling it.
> +}
> +
> +static int bcm2835_aux_irq_xlate(struct irq_domain *d,
> + struct device_node *ctrlr,
> + const u32 *intspec, unsigned int intsize,
> + unsigned long *out_hwirq,
> + unsigned int *out_type)
> +{
> + if (WARN_ON(intsize != 1))
> + return -EINVAL;
> +
> + if (WARN_ON(intspec[0] >= BCM2835_AUX_IRQ_COUNT))
> + return -EINVAL;
> +
> + *out_hwirq = intspec[0];
> + *out_type = IRQ_TYPE_NONE;
> +
> + return 0;
> +}
> +
> +/*
> + * The irq_mask and irq_unmask function pointers are used without
> + * validity checks, so they must not be NULL. Create a dummy function
> + * with the expected type for use as a no-op.
> + */
> +static void bcm2835_aux_irq_dummy(struct irq_data *data)
> +{
> +}
> +
> +static struct irq_chip bcm2835_aux_irq_chip = {
> + .name = "bcm2835-aux_irq",
> + .irq_mask = bcm2835_aux_irq_dummy,
> + .irq_unmask = bcm2835_aux_irq_dummy,
> +};
> +
> +static const struct irq_domain_ops bcm2835_aux_irq_ops = {
> + .xlate = bcm2835_aux_irq_xlate
> +};
> +
> +static int bcm2835_aux_irq_probe(struct platform_device *pdev)
> +{
> + struct device *dev = &pdev->dev;
> + struct device_node *node = dev->of_node;
> + int parent_irq;
> + struct resource *res;
> + void __iomem *reg;
> + int i;
> +
> + res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> + reg = devm_ioremap_resource(dev, res);
> + if (IS_ERR(reg))
> + return PTR_ERR(reg);
> +
> + parent_irq = irq_of_parse_and_map(node, 0);
> + if (!parent_irq)
> + return -ENXIO;
> +
> + aux_irq.status = reg + BCM2835_AUXIRQ;
> + aux_irq.domain = irq_domain_add_linear(node,
> + BCM2835_AUX_IRQ_COUNT,
> + &bcm2835_aux_irq_ops,
> + NULL);
> + if (!aux_irq.domain)
> + return -ENXIO;
> +
> + for (i = 0; i < BCM2835_AUX_IRQ_COUNT; i++) {
> + unsigned int irq = irq_create_mapping(aux_irq.domain, i);
> +
> + if (irq == 0)
> + return -ENXIO;
> +
> + irq_set_chip_and_handler(irq, &bcm2835_aux_irq_chip,
> + handle_level_irq);
> + }
My initial question notwithstanding, why do you need any of this? This
should be done at map time, and the irq_create_mapping() call should
entirely be driven from DT.
> +
> + return devm_request_irq(dev, parent_irq, bcm2835_aux_irq_handler,
> + 0, "bcm2835-aux-intc", NULL);
> +}
> +
> +static const struct of_device_id bcm2835_aux_irq_of_match[] = {
> + { .compatible = "brcm,bcm2835-aux-intc", },
> + {},
> +};
> +MODULE_DEVICE_TABLE(of, bcm2835_aux_irq_of_match);
> +
> +static struct platform_driver bcm2835_aux_irq_driver = {
> + .driver = {
> + .name = "bcm2835-aux-intc",
> + .of_match_table = bcm2835_aux_irq_of_match,
> + },
> + .probe = bcm2835_aux_irq_probe,
> +};
> +builtin_platform_driver(bcm2835_aux_irq_driver);
> +
> +MODULE_AUTHOR("Phil Elwell <phil@raspberrypi.org>");
> +MODULE_DESCRIPTION("BCM2835 auxiliary peripheral interrupt driver");
> +MODULE_LICENSE("GPL v2");
>
Thanks,
M.
--
Jazz is not dead. It just smells funny...
next prev parent reply other threads:[~2017-06-12 14:59 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-06-12 14:25 [PATCH v2 3/4] irqchip: Add BCM2835 AUX interrupt controller Phil Elwell
2017-06-12 14:59 ` Marc Zyngier [this message]
[not found] ` <56e314db-49a9-48fa-d7d9-877bd68aadfc-5wv7dgnIgG8@public.gmane.org>
2017-06-12 15:21 ` Phil Elwell
2017-06-12 16:19 ` Marc Zyngier
2017-06-12 16:33 ` Rogier Wolff
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=56e314db-49a9-48fa-d7d9-877bd68aadfc@arm.com \
--to=marc.zyngier@arm.com \
--cc=devicetree@vger.kernel.org \
--cc=eric@anholt.net \
--cc=f.fainelli@gmail.com \
--cc=jason@lakedaemon.net \
--cc=linux-clk@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-rpi-kernel@lists.infradead.org \
--cc=linux@armlinux.org.uk \
--cc=mark.rutland@arm.com \
--cc=mturquette@baylibre.com \
--cc=phil@raspberrypi.org \
--cc=robh+dt@kernel.org \
--cc=sboyd@codeaurora.org \
--cc=stefan.wahren@i2se.com \
--cc=tglx@linutronix.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).