All of lore.kernel.org
 help / color / mirror / Atom feed
From: Marc Zyngier <marc.zyngier@arm.com>
To: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>,
	Jiang Liu <jiang.liu@linux.intel.com>,
	Jason Cooper <jason@lakedaemon.net>,
	"linux-acpi@vger.kernel.org" <linux-acpi@vger.kernel.org>,
	"linux-arm-kernel@lists.infradead.org"
	<linux-arm-kernel@lists.infradead.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Tomasz Nowicki <tomasz.nowicki@linaro.org>,
	"hanjun.guo@linaro.org" <hanjun.guo@linaro.org>,
	"Rafael J. Wysocki" <rjw@rjwysocki.net>,
	"suravee.suthikulpanit@amd.com" <suravee.suthikulpanit@amd.com>
Subject: Re: [PATCH 4/5] acpi: gsi: Use acpi_gsi_descriptor to allocate interrupts
Date: Wed, 22 Jul 2015 08:46:04 +0100	[thread overview]
Message-ID: <55AF4A3C.1010608@arm.com> (raw)
In-Reply-To: <20150721181646.GC32187@red-moon>

On 21/07/15 19:16, Lorenzo Pieralisi wrote:
> On Tue, Jul 21, 2015 at 11:07:59AM +0100, Marc Zyngier wrote:
>> Now that the irqdomain layer is a bit more ACPI friendly, add the
>> mapping code that allows irq_create_acpi_mapping to be called.
>>
>> As we only support the GIC so far, support is pretty limited.
>>
>> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
>> ---
>>  drivers/acpi/gsi.c | 54 ++++++++++++++++++++++++++++++++++++------------------
>>  1 file changed, 36 insertions(+), 18 deletions(-)
>>
>> diff --git a/drivers/acpi/gsi.c b/drivers/acpi/gsi.c
>> index 38208f2..005d843 100644
>> --- a/drivers/acpi/gsi.c
>> +++ b/drivers/acpi/gsi.c
>> @@ -33,6 +33,12 @@ static unsigned int acpi_gsi_get_irq_type(int trigger, int polarity)
>>  	}
>>  }
>>  
>> +static struct irq_domain *irq_find_acpi_domain(enum acpi_irq_model_id id,
>> +					       enum irq_domain_bus_token bus_token)
>> +{
>> +	return irq_find_matching_host((void *)id, bus_token);
>> +}
>> +
>>  /**
>>   * acpi_gsi_to_irq() - Retrieve the linux irq number for a given GSI
>>   * @gsi: GSI IRQ number to map
>> @@ -45,12 +51,10 @@ static unsigned int acpi_gsi_get_irq_type(int trigger, int polarity)
>>   */
>>  int acpi_gsi_to_irq(u32 gsi, unsigned int *irq)
>>  {
>> -	/*
>> -	 * Only default domain is supported at present, always find
>> -	 * the mapping corresponding to default domain by passing NULL
>> -	 * as irq_domain parameter
>> -	 */
>> -	*irq = irq_find_mapping(NULL, gsi);
>> +	struct irq_domain *d = irq_find_acpi_domain(acpi_irq_model,
>> +						    DOMAIN_BUS_ANY);
>> +
>> +	*irq = irq_find_mapping(d, gsi);
>>  	/*
>>  	 * *irq == 0 means no mapping, that should
>>  	 * be reported as a failure
>> @@ -72,23 +76,35 @@ EXPORT_SYMBOL_GPL(acpi_gsi_to_irq);
>>  int acpi_register_gsi(struct device *dev, u32 gsi, int trigger,
>>  		      int polarity)
>>  {
>> -	unsigned int irq;
>> +	struct acpi_gsi_descriptor data;
>>  	unsigned int irq_type = acpi_gsi_get_irq_type(trigger, polarity);
>> +	struct irq_domain *d = irq_find_acpi_domain(acpi_irq_model,
>> +						    DOMAIN_BUS_ANY);
>>  
>>  	/*
>> -	 * There is no way at present to look-up the IRQ domain on ACPI,
>> -	 * hence always create mapping referring to the default domain
>> -	 * by passing NULL as irq_domain parameter
>> +	 * Populate the GSI descriptor in a way that matches the way
>> +	 * the driver expects its of_phandle_args.
>>  	 */
>> -	irq = irq_create_mapping(NULL, gsi);
>> -	if (!irq)
>> +	switch (acpi_irq_model) {
>> +	case ACPI_IRQ_MODEL_GIC:
>> +		if (gsi >= 32) {
>> +			data.param[0] = 0;
>> +			data.param[1] = gsi - 32;
>> +			data.param[2] = irq_type;
>> +		} else {
>> +			data.param[0] = 1;
>> +			data.param[1] = gsi - 16;
>> +			data.param[2] = 0xff << 4 | irq_type;
>> +		}
>> +
>> +		data.param_count = 3;
>> +		break;
>> +	default:
>> +		pr_warn("Unknown acpi_irq_model = %d\n", acpi_irq_model);
>>  		return -EINVAL;
>> +	}
> 
> Nit: you could move the switch in a function (in this file) so that
> code that is irq model specific is self-contained and in a function
> on its own, if we ever have to add other irq models later I think it
> is easier to parse.

Yup, makes sense.

Actually, having given it a bit more thoughts, there is not reason why
this translation function should be located in this file. The interrupt
controller driver, instead of doing a

acpi_irq_model = ACPI_IRQ_MODEL_BLAH;

could as well be doing

acpi_set_irq_model(ACPI_IRQ_MODEL_BLAH, blah_acpi_gsi_desc_populate);

making the whole thing completely interrupt-controller agnostic.

Thanks,

	M.

> 
> Lorenzo
> 
>>  
>> -	/* Set irq type if specified and different than the current one */
>> -	if (irq_type != IRQ_TYPE_NONE &&
>> -		irq_type != irq_get_trigger_type(irq))
>> -		irq_set_irq_type(irq, irq_type);
>> -	return irq;
>> +	return irq_create_acpi_mapping(d, &data);
>>  }
>>  EXPORT_SYMBOL_GPL(acpi_register_gsi);
>>  
>> @@ -98,7 +114,9 @@ EXPORT_SYMBOL_GPL(acpi_register_gsi);
>>   */
>>  void acpi_unregister_gsi(u32 gsi)
>>  {
>> -	int irq = irq_find_mapping(NULL, gsi);
>> +	struct irq_domain *d = irq_find_acpi_domain(acpi_irq_model,
>> +						    DOMAIN_BUS_ANY);
>> +	int irq = irq_find_mapping(d, gsi);
>>  
>>  	irq_dispose_mapping(irq);
>>  }
>> -- 
>> 2.1.4
>>
> 


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

WARNING: multiple messages have this Message-ID (diff)
From: marc.zyngier@arm.com (Marc Zyngier)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 4/5] acpi: gsi: Use acpi_gsi_descriptor to allocate interrupts
Date: Wed, 22 Jul 2015 08:46:04 +0100	[thread overview]
Message-ID: <55AF4A3C.1010608@arm.com> (raw)
In-Reply-To: <20150721181646.GC32187@red-moon>

On 21/07/15 19:16, Lorenzo Pieralisi wrote:
> On Tue, Jul 21, 2015 at 11:07:59AM +0100, Marc Zyngier wrote:
>> Now that the irqdomain layer is a bit more ACPI friendly, add the
>> mapping code that allows irq_create_acpi_mapping to be called.
>>
>> As we only support the GIC so far, support is pretty limited.
>>
>> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
>> ---
>>  drivers/acpi/gsi.c | 54 ++++++++++++++++++++++++++++++++++++------------------
>>  1 file changed, 36 insertions(+), 18 deletions(-)
>>
>> diff --git a/drivers/acpi/gsi.c b/drivers/acpi/gsi.c
>> index 38208f2..005d843 100644
>> --- a/drivers/acpi/gsi.c
>> +++ b/drivers/acpi/gsi.c
>> @@ -33,6 +33,12 @@ static unsigned int acpi_gsi_get_irq_type(int trigger, int polarity)
>>  	}
>>  }
>>  
>> +static struct irq_domain *irq_find_acpi_domain(enum acpi_irq_model_id id,
>> +					       enum irq_domain_bus_token bus_token)
>> +{
>> +	return irq_find_matching_host((void *)id, bus_token);
>> +}
>> +
>>  /**
>>   * acpi_gsi_to_irq() - Retrieve the linux irq number for a given GSI
>>   * @gsi: GSI IRQ number to map
>> @@ -45,12 +51,10 @@ static unsigned int acpi_gsi_get_irq_type(int trigger, int polarity)
>>   */
>>  int acpi_gsi_to_irq(u32 gsi, unsigned int *irq)
>>  {
>> -	/*
>> -	 * Only default domain is supported at present, always find
>> -	 * the mapping corresponding to default domain by passing NULL
>> -	 * as irq_domain parameter
>> -	 */
>> -	*irq = irq_find_mapping(NULL, gsi);
>> +	struct irq_domain *d = irq_find_acpi_domain(acpi_irq_model,
>> +						    DOMAIN_BUS_ANY);
>> +
>> +	*irq = irq_find_mapping(d, gsi);
>>  	/*
>>  	 * *irq == 0 means no mapping, that should
>>  	 * be reported as a failure
>> @@ -72,23 +76,35 @@ EXPORT_SYMBOL_GPL(acpi_gsi_to_irq);
>>  int acpi_register_gsi(struct device *dev, u32 gsi, int trigger,
>>  		      int polarity)
>>  {
>> -	unsigned int irq;
>> +	struct acpi_gsi_descriptor data;
>>  	unsigned int irq_type = acpi_gsi_get_irq_type(trigger, polarity);
>> +	struct irq_domain *d = irq_find_acpi_domain(acpi_irq_model,
>> +						    DOMAIN_BUS_ANY);
>>  
>>  	/*
>> -	 * There is no way at present to look-up the IRQ domain on ACPI,
>> -	 * hence always create mapping referring to the default domain
>> -	 * by passing NULL as irq_domain parameter
>> +	 * Populate the GSI descriptor in a way that matches the way
>> +	 * the driver expects its of_phandle_args.
>>  	 */
>> -	irq = irq_create_mapping(NULL, gsi);
>> -	if (!irq)
>> +	switch (acpi_irq_model) {
>> +	case ACPI_IRQ_MODEL_GIC:
>> +		if (gsi >= 32) {
>> +			data.param[0] = 0;
>> +			data.param[1] = gsi - 32;
>> +			data.param[2] = irq_type;
>> +		} else {
>> +			data.param[0] = 1;
>> +			data.param[1] = gsi - 16;
>> +			data.param[2] = 0xff << 4 | irq_type;
>> +		}
>> +
>> +		data.param_count = 3;
>> +		break;
>> +	default:
>> +		pr_warn("Unknown acpi_irq_model = %d\n", acpi_irq_model);
>>  		return -EINVAL;
>> +	}
> 
> Nit: you could move the switch in a function (in this file) so that
> code that is irq model specific is self-contained and in a function
> on its own, if we ever have to add other irq models later I think it
> is easier to parse.

Yup, makes sense.

Actually, having given it a bit more thoughts, there is not reason why
this translation function should be located in this file. The interrupt
controller driver, instead of doing a

acpi_irq_model = ACPI_IRQ_MODEL_BLAH;

could as well be doing

acpi_set_irq_model(ACPI_IRQ_MODEL_BLAH, blah_acpi_gsi_desc_populate);

making the whole thing completely interrupt-controller agnostic.

Thanks,

	M.

> 
> Lorenzo
> 
>>  
>> -	/* Set irq type if specified and different than the current one */
>> -	if (irq_type != IRQ_TYPE_NONE &&
>> -		irq_type != irq_get_trigger_type(irq))
>> -		irq_set_irq_type(irq, irq_type);
>> -	return irq;
>> +	return irq_create_acpi_mapping(d, &data);
>>  }
>>  EXPORT_SYMBOL_GPL(acpi_register_gsi);
>>  
>> @@ -98,7 +114,9 @@ EXPORT_SYMBOL_GPL(acpi_register_gsi);
>>   */
>>  void acpi_unregister_gsi(u32 gsi)
>>  {
>> -	int irq = irq_find_mapping(NULL, gsi);
>> +	struct irq_domain *d = irq_find_acpi_domain(acpi_irq_model,
>> +						    DOMAIN_BUS_ANY);
>> +	int irq = irq_find_mapping(d, gsi);
>>  
>>  	irq_dispose_mapping(irq);
>>  }
>> -- 
>> 2.1.4
>>
> 


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

  reply	other threads:[~2015-07-22  7:46 UTC|newest]

Thread overview: 63+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-21 10:07 [PATCH 0/5] Making the generic ACPI GSI layer irqdomain aware Marc Zyngier
2015-07-21 10:07 ` Marc Zyngier
2015-07-21 10:07 ` Marc Zyngier
2015-07-21 10:07 ` [PATCH 1/5] genirq: irqdomain: Use an accessor for the of_node field Marc Zyngier
2015-07-21 10:07   ` Marc Zyngier
2015-07-21 10:07   ` Marc Zyngier
2015-07-22  7:35   ` Hanjun Guo
2015-07-22  7:35     ` Hanjun Guo
2015-07-22  7:52     ` Marc Zyngier
2015-07-22  7:52       ` Marc Zyngier
2015-07-22  7:58       ` Thomas Gleixner
2015-07-22  7:58         ` Thomas Gleixner
2015-07-22 12:43         ` Marc Zyngier
2015-07-22 12:43           ` Marc Zyngier
2015-07-22  7:57     ` Thomas Gleixner
2015-07-22  7:57       ` Thomas Gleixner
2015-07-21 10:07 ` [PATCH 2/5] genirq: irqdomain: Remove irqdomain dependency on struct device_node Marc Zyngier
2015-07-21 10:07   ` Marc Zyngier
2015-07-21 10:07   ` Marc Zyngier
2015-07-21 17:56   ` Lorenzo Pieralisi
2015-07-21 17:56     ` Lorenzo Pieralisi
2015-07-22  8:13     ` Hanjun Guo
2015-07-22  8:13       ` Hanjun Guo
2015-07-23  8:59       ` Suravee Suthikulpanit
2015-07-23  8:59         ` Suravee Suthikulpanit
2015-07-23  8:59         ` Suravee Suthikulpanit
2015-07-21 10:07 ` [PATCH 3/5] genirq: irqdomain: Add irq_create_acpi_mappings Marc Zyngier
2015-07-21 10:07   ` Marc Zyngier
2015-07-21 10:07   ` Marc Zyngier
2015-07-21 10:07 ` [PATCH 4/5] acpi: gsi: Use acpi_gsi_descriptor to allocate interrupts Marc Zyngier
2015-07-21 10:07   ` Marc Zyngier
2015-07-21 10:07   ` Marc Zyngier
2015-07-21 18:16   ` Lorenzo Pieralisi
2015-07-21 18:16     ` Lorenzo Pieralisi
2015-07-22  7:46     ` Marc Zyngier [this message]
2015-07-22  7:46       ` Marc Zyngier
2015-07-21 10:08 ` [PATCH 5/5] irqchip: GIC: Switch ACPI support to stacked domains Marc Zyngier
2015-07-21 10:08   ` Marc Zyngier
2015-07-21 10:08   ` Marc Zyngier
2015-07-21 12:34   ` Graeme Gregory
2015-07-21 12:34     ` Graeme Gregory
2015-07-21 13:03     ` Marc Zyngier
2015-07-21 13:03       ` Marc Zyngier
2015-07-21 18:05   ` Lorenzo Pieralisi
2015-07-21 18:05     ` Lorenzo Pieralisi
2015-07-21 18:12     ` Marc Zyngier
2015-07-21 18:12       ` Marc Zyngier
2015-07-22  8:35       ` Hanjun Guo
2015-07-22  8:35         ` Hanjun Guo
2015-07-22  8:35         ` Hanjun Guo
2015-07-22  8:45         ` Hanjun Guo
2015-07-22  8:45           ` Hanjun Guo
2015-07-22  8:45           ` Hanjun Guo
2015-07-22  8:53         ` Marc Zyngier
2015-07-22  8:53           ` Marc Zyngier
2015-07-22  8:53           ` Marc Zyngier
2015-07-22  9:33           ` Hanjun Guo
2015-07-22  9:33             ` Hanjun Guo
2015-07-22  9:33             ` Hanjun Guo
2015-07-21 12:35 ` [PATCH 0/5] Making the generic ACPI GSI layer irqdomain aware Graeme Gregory
2015-07-21 12:35   ` Graeme Gregory
2015-07-22  6:45 ` Hanjun Guo
2015-07-22  6:45   ` Hanjun Guo

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=55AF4A3C.1010608@arm.com \
    --to=marc.zyngier@arm.com \
    --cc=hanjun.guo@linaro.org \
    --cc=jason@lakedaemon.net \
    --cc=jiang.liu@linux.intel.com \
    --cc=linux-acpi@vger.kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lorenzo.pieralisi@arm.com \
    --cc=rjw@rjwysocki.net \
    --cc=suravee.suthikulpanit@amd.com \
    --cc=tglx@linutronix.de \
    --cc=tomasz.nowicki@linaro.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.