linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
To: Rob Herring <robh+dt@kernel.org>, Shawn Guo <shawnguo@kernel.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Jason Cooper <jason@lakedaemon.net>,
	Marc Zyngier <marc.zyngier@arm.com>
Cc: Andy Tang <andy.tang@nxp.com>,
	Alexander Stein <alexander.stein@systec-electronic.com>,
	Rasmus Villemoes <rasmus.villemoes@prevas.dk>,
	<linux-kernel@vger.kernel.org>
Subject: [PATCH v3 1/2] irqchip: add support for Layerscape external interrupt lines
Date: Mon, 22 Jan 2018 10:21:32 +0100	[thread overview]
Message-ID: <20180122092133.23177-1-rasmus.villemoes@prevas.dk> (raw)
In-Reply-To: <1513758631-19909-1-git-send-email-rasmus.villemoes@prevas.dk>

The LS1021A allows inverting the polarity of six interrupt lines
IRQ[0:5] via the scfg_intpcr register, effectively allowing
IRQ_TYPE_LEVEL_LOW and IRQ_TYPE_EDGE_FALLING for those. We just need to
check the type, set the relevant bit in INTPCR accordingly, and fixup
the type argument before calling the GIC's irq_set_type.

In fact, the power-on-reset value of the INTPCR register on the LS1021A
is so that all six lines have their polarity inverted. Hence any
hardware connected to those lines is unusable without this: If the line
is indeed active low, the generic GIC code will reject an irq spec with
IRQ_TYPE_LEVEL_LOW, while if the line is active high, we must obviously
disable the polarity inversion (writing 0 to the relevant bit) before
unmasking the interrupt.

Some other Layerscape SOCs (LS1043A, LS1046A) reportedly have a similar
feature, just with a different number of external interrupt lines (and a
different POR value for the INTPCR register). This driver should be
prepared for supporting those by properly filling out the device tree
node, but I don't have the full reference manual, nor the hardware to be
able to test it. I do know, from a tiny clipout from one of the other
reference manuals I was shown, that 1U<<n is the right formula to
use for setting/clearing the bit corresponding to the external IRQn, but
I don't know which interrupts on the GIC those lines represent.

There's also a little Kconfig/Kbuild issue: For now, I've let the driver
be built if CONFIG_SOC_LS1021A is set. For the others, it might make
sense to instead use CONFIG_ARCH_LAYERSCAPE, but SOC_LS1021A does not
select ARCH_LAYERSCAPE. The simplest solution is probably to do what is
done for irq-ls-scfg-msi.c: introduce a "def_bool y if SOC_LS1021A ||
ARCH_LAYERSCAPE" symbol. But I think that can wait until somebody with
the hardware actually tests that this works for the other platforms. At
that point, one can also add the extra IRQCHIP_DECLARE instances.

Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
---
Changes since v2 (all addressing comments from Rob Herring):

- use fsl,bit-reverse rather than bit-reverse
- make the dts node a child of the scfg node
- make the node name "interrupt-controller"

 drivers/irqchip/Makefile        |   1 +
 drivers/irqchip/irq-ls-extirq.c | 173 ++++++++++++++++++++++++++++++++++++++++
 2 files changed, 174 insertions(+)
 create mode 100644 drivers/irqchip/irq-ls-extirq.c

diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile
index b842dfdc903f..32d7160680fe 100644
--- a/drivers/irqchip/Makefile
+++ b/drivers/irqchip/Makefile
@@ -75,6 +75,7 @@ obj-$(CONFIG_MVEBU_ICU)			+= irq-mvebu-icu.o
 obj-$(CONFIG_MVEBU_ODMI)		+= irq-mvebu-odmi.o
 obj-$(CONFIG_MVEBU_PIC)			+= irq-mvebu-pic.o
 obj-$(CONFIG_LS_SCFG_MSI)		+= irq-ls-scfg-msi.o
+obj-$(CONFIG_SOC_LS1021A)		+= irq-ls-extirq.o
 obj-$(CONFIG_EZNPS_GIC)			+= irq-eznps.o
 obj-$(CONFIG_ARCH_ASPEED)		+= irq-aspeed-vic.o irq-aspeed-i2c-ic.o
 obj-$(CONFIG_STM32_EXTI) 		+= irq-stm32-exti.o
diff --git a/drivers/irqchip/irq-ls-extirq.c b/drivers/irqchip/irq-ls-extirq.c
new file mode 100644
index 000000000000..ac84ad053b51
--- /dev/null
+++ b/drivers/irqchip/irq-ls-extirq.c
@@ -0,0 +1,173 @@
+#define pr_fmt(fmt) "irq-ls-extirq: " fmt
+
+#include <linux/irq.h>
+#include <linux/irqchip.h>
+#include <linux/irqdomain.h>
+#include <linux/of.h>
+#include <linux/of_irq.h>
+#include <linux/of_address.h>
+#include <linux/mfd/syscon.h>
+#include <linux/regmap.h>
+#include <linux/slab.h>
+#include <dt-bindings/interrupt-controller/arm-gic.h>
+
+#define MAXIRQ 12
+
+struct extirq_chip_data {
+	struct regmap *syscon;
+	u32           intpcr;
+	bool          bit_reverse;
+	u32           nirq;
+	u32           parent_irq[MAXIRQ];
+};
+
+static int
+ls_extirq_set_type(struct irq_data *data, unsigned int type)
+{
+	irq_hw_number_t hwirq = data->hwirq;
+	struct extirq_chip_data *chip_data = data->chip_data;
+	u32 value, mask;
+
+	if (chip_data->bit_reverse)
+		mask = 1U << (31 - hwirq);
+	else
+		mask = 1U << hwirq;
+
+	switch (type) {
+	case IRQ_TYPE_LEVEL_LOW:
+		type = IRQ_TYPE_LEVEL_HIGH;
+		value = mask;
+		break;
+	case IRQ_TYPE_EDGE_FALLING:
+		type = IRQ_TYPE_EDGE_RISING;
+		value = mask;
+		break;
+	case IRQ_TYPE_LEVEL_HIGH:
+	case IRQ_TYPE_EDGE_RISING:
+		value = 0;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	regmap_update_bits(chip_data->syscon, chip_data->intpcr, mask, value);
+
+	data = data->parent_data;
+	return data->chip->irq_set_type(data, type);
+}
+
+static struct irq_chip extirq_chip = {
+	.name			= "extirq",
+	.irq_mask		= irq_chip_mask_parent,
+	.irq_unmask		= irq_chip_unmask_parent,
+	.irq_eoi		= irq_chip_eoi_parent,
+	.irq_set_type		= ls_extirq_set_type,
+	.irq_retrigger		= irq_chip_retrigger_hierarchy,
+	.irq_set_affinity	= irq_chip_set_affinity_parent,
+	.flags                  = IRQCHIP_SET_TYPE_MASKED,
+};
+
+static int
+ls_extirq_domain_translate(struct irq_domain *d, struct irq_fwspec *fwspec,
+			   unsigned long *hwirq, unsigned int *type)
+{
+	if (!is_of_node(fwspec->fwnode))
+		return -EINVAL;
+
+	if (fwspec->param_count != 3)
+		return -EINVAL;
+
+	/* No PPI should point to this domain */
+	if (fwspec->param[0] != GIC_SPI)
+		return -EINVAL;
+
+	*hwirq = fwspec->param[1];
+	*type = fwspec->param[2] & IRQ_TYPE_SENSE_MASK;
+	return 0;
+}
+
+static int
+ls_extirq_domain_alloc(struct irq_domain *domain, unsigned int virq,
+		       unsigned int nr_irqs, void *arg)
+{
+	irq_hw_number_t hwirq;
+	struct irq_fwspec *fwspec = arg;
+	struct irq_fwspec gic_fwspec;
+	struct extirq_chip_data *chip_data = domain->host_data;
+
+	if (fwspec->param_count != 3)
+		return -EINVAL;
+
+	if (fwspec->param[0] != GIC_SPI)
+		return -EINVAL;
+
+	hwirq = fwspec->param[1];
+	if (hwirq >= chip_data->nirq)
+		return -EINVAL;
+
+	irq_domain_set_hwirq_and_chip(domain, virq, hwirq, &extirq_chip,
+				      chip_data);
+
+	gic_fwspec.fwnode = domain->parent->fwnode;
+	gic_fwspec.param_count = 3;
+	gic_fwspec.param[0] = GIC_SPI;
+	gic_fwspec.param[1] = chip_data->parent_irq[hwirq];
+	gic_fwspec.param[2] = fwspec->param[2];
+
+	return irq_domain_alloc_irqs_parent(domain, virq, 1, &gic_fwspec);
+}
+
+static const struct irq_domain_ops extirq_domain_ops = {
+	.translate	= ls_extirq_domain_translate,
+	.alloc		= ls_extirq_domain_alloc,
+	.free		= irq_domain_free_irqs_common,
+};
+
+static int __init
+ls_extirq_of_init(struct device_node *node, struct device_node *parent)
+{
+
+	struct irq_domain *domain, *domain_parent;
+	struct extirq_chip_data *chip_data;
+	int ret;
+
+	domain_parent = irq_find_host(parent);
+	if (!domain_parent) {
+		pr_err("interrupt-parent not found\n");
+		return -EINVAL;
+	}
+
+	chip_data = kzalloc(sizeof(*chip_data), GFP_KERNEL);
+	if (!chip_data)
+		return -ENOMEM;
+
+	chip_data->syscon = syscon_node_to_regmap(node->parent);
+	ret = of_property_read_u32(node, "offset", &chip_data->intpcr);
+
+	if (IS_ERR(chip_data->syscon))
+		ret = PTR_ERR(chip_data->syscon);
+	if (ret)
+		goto out_free_chip;
+
+	ret = of_property_read_variable_u32_array(node, "interrupts", chip_data->parent_irq,
+						  1, ARRAY_SIZE(chip_data->parent_irq));
+	if (ret < 0)
+		goto out_free_chip;
+	chip_data->nirq = ret;
+	chip_data->bit_reverse = of_property_read_bool(node, "fsl,bit-reverse");
+
+	domain = irq_domain_add_hierarchy(domain_parent, 0, chip_data->nirq, node,
+					  &extirq_domain_ops, chip_data);
+	if (!domain) {
+		ret = -ENOMEM;
+		goto out_free_chip;
+	}
+
+	return 0;
+
+out_free_chip:
+	kfree(chip_data);
+	return ret;
+}
+
+IRQCHIP_DECLARE(ls1021a_extirq, "fsl,ls1021a-extirq", ls_extirq_of_init);
-- 
2.15.1

  parent reply	other threads:[~2018-01-22  9:22 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-12-04 15:11 polarity inversion on LS1021a Rasmus Villemoes
2017-12-04 15:23 ` Marc Zyngier
2017-12-08 14:33   ` [RFC] irqchip: add support for LS1021A external interrupt lines Rasmus Villemoes
2017-12-08 15:11     ` Alexander Stein
2017-12-08 16:09       ` Marc Zyngier
2017-12-11  9:08         ` Rasmus Villemoes
2017-12-11  9:45           ` Alexander Stein
2017-12-11 10:02             ` Alexander Stein
2017-12-11 13:45               ` Rasmus Villemoes
2017-12-11 14:06                 ` Rasmus Villemoes
2017-12-11 14:38                   ` Alexander Stein
2017-12-08 16:02     ` Marc Zyngier
2017-12-11  9:30       ` Rasmus Villemoes
2017-12-11 18:29         ` Marc Zyngier
2017-12-12 23:28     ` Rob Herring
2017-12-15 22:55       ` Rasmus Villemoes
2017-12-21 22:45         ` Rob Herring
2017-12-20  8:30     ` [PATCH v2 1/2] irqchip: add support for Layerscape " Rasmus Villemoes
2017-12-20  8:30       ` [PATCH v2 2/2] dt/bindings: Add bindings for Layerscape external irqs Rasmus Villemoes
2017-12-21 22:44         ` Rob Herring
2018-01-22  9:21       ` Rasmus Villemoes [this message]
2018-01-22  9:21         ` [PATCH v3 " Rasmus Villemoes
2018-01-24 15:28           ` Marc Zyngier
2018-01-25 15:02         ` [PATCH v4 1/2] irqchip: add support for Layerscape external interrupt lines Rasmus Villemoes
2018-01-25 15:02           ` [PATCH v4 2/2] dt/bindings: Add bindings for Layerscape external irqs Rasmus Villemoes
2018-02-05  6:07             ` Rob Herring
2018-02-08 15:08               ` Rasmus Villemoes
2018-02-09  9:47                 ` Marc Zyngier
2018-02-23 21:08           ` [PATCH v5 0/2] irqchip: add support for Layerscape external interrupt lines Rasmus Villemoes
2018-02-23 21:08             ` [PATCH v5 1/2] " Rasmus Villemoes
2018-03-01 12:16               ` Thomas Gleixner
2018-05-04  7:44                 ` Rasmus Villemoes
2019-09-17  9:39                   ` Kurt Kanzenbach
2018-02-23 21:09             ` [PATCH v5 2/2] dt/bindings: Add bindings for Layerscape external irqs Rasmus Villemoes
2018-03-02 19:49               ` Rob Herring
2018-05-04  8:07                 ` Rasmus Villemoes
2017-12-04 15:31 ` polarity inversion on LS1021a Alexander Stein
2017-12-04 15:37   ` Marc Zyngier
2017-12-04 16:04     ` Alexander Stein

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=20180122092133.23177-1-rasmus.villemoes@prevas.dk \
    --to=rasmus.villemoes@prevas.dk \
    --cc=alexander.stein@systec-electronic.com \
    --cc=andy.tang@nxp.com \
    --cc=jason@lakedaemon.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=marc.zyngier@arm.com \
    --cc=robh+dt@kernel.org \
    --cc=shawnguo@kernel.org \
    --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).