devicetree.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Will Deacon <will.deacon-5wv7dgnIgG8@public.gmane.org>
To: Lorenzo Pieralisi <Lorenzo.Pieralisi-5wv7dgnIgG8@public.gmane.org>
Cc: Nicolas Pitre
	<nicolas.pitre-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>,
	Kukjin Kim <kgene.kim-Sze3O3UU22JBDgjK7y7TUQ@public.gmane.org>,
	Russell King <linux-lFZ/pmaqli7XmaaqVzeoHQ@public.gmane.org>,
	Pawel Moll <Pawel.Moll-5wv7dgnIgG8@public.gmane.org>,
	Catalin Marinas <Catalin.Marinas-5wv7dgnIgG8@public.gmane.org>,
	"devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org"
	<devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org>,
	Amit Kucheria
	<amit.kucheria-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>,
	"rob.herring-bsGFqQB8/DxBDgjK7y7TUQ@public.gmane.org"
	<rob.herring-bsGFqQB8/DxBDgjK7y7TUQ@public.gmane.org>,
	David Brown <davidb-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>,
	"linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org"
	<linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org>
Subject: Re: [RFC PATCH 4/4] ARM: gic: use a private mapping for CPU target interfaces
Date: Tue, 6 Nov 2012 22:16:51 +0000	[thread overview]
Message-ID: <20121106221651.GE21764@mudshark.cambridge.arm.com> (raw)
In-Reply-To: <1350393709-23546-5-git-send-email-lorenzo.pieralisi-5wv7dgnIgG8@public.gmane.org>

On Tue, Oct 16, 2012 at 02:21:48PM +0100, Lorenzo Pieralisi wrote:
> From: Nicolas Pitre <nicolas.pitre-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
> 
> The GIC interface numbering does not necessarily follow the logical
> CPU numbering, especially for complex topologies such as multi-cluster
> systems.
> 
> Fortunately we can easily probe the GIC to create a mapping as the
> Interrupt Processor Targets Registers for the first 32 interrupts are
> read-only, and each field returns a value that always corresponds to
> the processor reading the register.
> 
> Initially all mappings target all CPUs in case an IPI is required to
> boot secondary CPUs.  It is refined as those CPUs discover what their
> actual mapping is.
> 
> Signed-off-by: Nicolas Pitre <nico-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>

This is a really neat idea!

> ---
>  arch/arm/common/gic.c | 42 ++++++++++++++++++++++++++++++++++--------
>  1 file changed, 34 insertions(+), 8 deletions(-)
> 
> diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c
> index aa52699..1338a55 100644
> --- a/arch/arm/common/gic.c
> +++ b/arch/arm/common/gic.c
> @@ -70,6 +70,13 @@ struct gic_chip_data {
>  static DEFINE_RAW_SPINLOCK(irq_controller_lock);
>  
>  /*
> + * The GIC mapping of CPU interfaces does not necessarily match
> + * the logical CPU numbering.  Let's use a mapping as returned
> + * by the GIC itself.
> + */
> +static u8 gic_cpu_map[8] __read_mostly;

Can we have a #define for the number CPUs supported by the GIC? It gets
used a fair amount in this patch for loop bounds etc.

> +/*
>   * Supported arch specific GIC irq extension.
>   * Default make them NULL.
>   */
> @@ -242,7 +249,7 @@ static int gic_set_affinity(struct irq_data *d, const struct cpumask *mask_val,
>  		return -EINVAL;
>  
>  	mask = 0xff << shift;
> -	bit = 1 << (cpu_logical_map(cpu) + shift);
> +	bit = gic_cpu_map[cpu] << shift;
>  
>  	raw_spin_lock(&irq_controller_lock);
>  	val = readl_relaxed(reg) & ~mask;
> @@ -349,11 +356,6 @@ static void __init gic_dist_init(struct gic_chip_data *gic)
>  	u32 cpumask;
>  	unsigned int gic_irqs = gic->gic_irqs;
>  	void __iomem *base = gic_data_dist_base(gic);
> -	u32 cpu = cpu_logical_map(smp_processor_id());
> -
> -	cpumask = 1 << cpu;
> -	cpumask |= cpumask << 8;
> -	cpumask |= cpumask << 16;
>  
>  	writel_relaxed(0, base + GIC_DIST_CTRL);
>  
> @@ -366,6 +368,7 @@ static void __init gic_dist_init(struct gic_chip_data *gic)
>  	/*
>  	 * Set all global interrupts to this CPU only.
>  	 */
> +	cpumask = readl_relaxed(base + GIC_DIST_TARGET + 0);
>  	for (i = 32; i < gic_irqs; i += 4)
>  		writel_relaxed(cpumask, base + GIC_DIST_TARGET + i * 4 / 4);
>  
> @@ -389,9 +392,25 @@ static void __cpuinit gic_cpu_init(struct gic_chip_data *gic)
>  {
>  	void __iomem *dist_base = gic_data_dist_base(gic);
>  	void __iomem *base = gic_data_cpu_base(gic);
> +	unsigned int cpu_mask, cpu = smp_processor_id();
>  	int i;
>  
>  	/*
> +	 * Get what the GIC says our CPU mask is.
> +	 */
> +	BUG_ON(cpu >= 8);
> +	cpu_mask = readl_relaxed(dist_base + GIC_DIST_TARGET + 0);

Making the mask a u8 and using readb_relaxed here makes this bit of code
clearer to me (and the GIC apparently allows such an access to this
register).

> +	gic_cpu_map[cpu] = cpu_mask;
> +
> +	/*
> +	 * Clear our mask from the other map entries in case they're
> +	 * still undefined.
> +	 */
> +	for (i = 0; i < 8; i++)
> +		if (i != cpu)
> +			gic_cpu_map[i] &= ~cpu_mask;
> +
> +	/*
>  	 * Deal with the banked PPI and SGI interrupts - disable all
>  	 * PPI interrupts, ensure all SGI interrupts are enabled.
>  	 */
> @@ -646,7 +665,7 @@ void __init gic_init_bases(unsigned int gic_nr, int irq_start,
>  {
>  	irq_hw_number_t hwirq_base;
>  	struct gic_chip_data *gic;
> -	int gic_irqs, irq_base;
> +	int gic_irqs, irq_base, i;
>  
>  	BUG_ON(gic_nr >= MAX_GIC_NR);
>  
> @@ -683,6 +702,13 @@ void __init gic_init_bases(unsigned int gic_nr, int irq_start,
>  	}
>  
>  	/*
> +	 * Initialize the CPU interface map to all CPUs.
> +	 * It will be refined as each CPU probes its ID.
> +	 */
> +	for (i = 0; i < 8; i++)
> +		gic_cpu_map[i] = 0xff;
> +
> +	/*
>  	 * For primary GICs, skip over SGIs.
>  	 * For secondary GICs, skip over PPIs, too.
>  	 */
> @@ -737,7 +763,7 @@ void gic_raise_softirq(const struct cpumask *mask, unsigned int irq)
>  
>  	/* Convert our logical CPU mask into a physical one. */
>  	for_each_cpu(cpu, mask)
> -		map |= 1 << cpu_logical_map(cpu);
> +		map |= gic_cpu_map[cpu];
>  
>  	/*
>  	 * Ensure that stores to Normal memory are visible to the

With those changes:

Acked-by: Will Deacon <will.deacon-5wv7dgnIgG8@public.gmane.org>

Will

  parent reply	other threads:[~2012-11-06 22:16 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-10-16 13:21 [RFC PATCH 0/4] ARM: multi-cluster aware boot protocol Lorenzo Pieralisi
2012-10-16 13:21 ` [RFC PATCH 1/4] ARM: kernel: add device tree init map function Lorenzo Pieralisi
     [not found]   ` <1350393709-23546-2-git-send-email-lorenzo.pieralisi-5wv7dgnIgG8@public.gmane.org>
2012-10-16 20:42     ` Rob Herring
2012-10-17 10:48       ` Lorenzo Pieralisi
2012-11-06 21:50     ` Will Deacon
2012-11-07 10:23       ` Lorenzo Pieralisi
     [not found]         ` <20121107102349.GC17831-7AyDDHkRsp3ZROr8t4l/smS4ubULX0JqMm0uRHvK7Nw@public.gmane.org>
2012-11-07 11:05           ` Will Deacon
2012-11-07 12:00             ` Lorenzo Pieralisi
     [not found]               ` <20121107120052.GD17831-7AyDDHkRsp3ZROr8t4l/smS4ubULX0JqMm0uRHvK7Nw@public.gmane.org>
2012-11-07 15:35                 ` Will Deacon
2012-11-07 17:43                   ` Lorenzo Pieralisi
2012-10-16 13:21 ` [RFC PATCH 2/4] ARM: kernel: add cpu logical map DT init in setup_arch Lorenzo Pieralisi
     [not found]   ` <1350393709-23546-3-git-send-email-lorenzo.pieralisi-5wv7dgnIgG8@public.gmane.org>
2012-11-06 21:52     ` Will Deacon
2012-10-16 13:21 ` [RFC PATCH 3/4] ARM: kernel: add logical mappings look-up Lorenzo Pieralisi
     [not found]   ` <1350393709-23546-4-git-send-email-lorenzo.pieralisi-5wv7dgnIgG8@public.gmane.org>
2012-11-06 22:00     ` Will Deacon
2012-10-16 13:21 ` [RFC PATCH 4/4] ARM: gic: use a private mapping for CPU target interfaces Lorenzo Pieralisi
     [not found]   ` <1350393709-23546-5-git-send-email-lorenzo.pieralisi-5wv7dgnIgG8@public.gmane.org>
2012-11-06 22:16     ` Will Deacon [this message]
     [not found]       ` <20121106221651.GE21764-MRww78TxoiP5vMa5CHWGZ34zcgK1vI+I0E9HWUfgJXw@public.gmane.org>
2012-11-06 22:59         ` Nicolas Pitre
     [not found]           ` <alpine.LFD.2.02.1211061755260.21033-QuJgVwGFrdf/9pzu0YdTqQ@public.gmane.org>
2012-11-07 10:23             ` Will Deacon
     [not found]               ` <20121107102357.GD23305-MRww78TxoiP5vMa5CHWGZ34zcgK1vI+I0E9HWUfgJXw@public.gmane.org>
2012-11-07 15:11                 ` Nicolas Pitre

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=20121106221651.GE21764@mudshark.cambridge.arm.com \
    --to=will.deacon-5wv7dgnigg8@public.gmane.org \
    --cc=Catalin.Marinas-5wv7dgnIgG8@public.gmane.org \
    --cc=Lorenzo.Pieralisi-5wv7dgnIgG8@public.gmane.org \
    --cc=Pawel.Moll-5wv7dgnIgG8@public.gmane.org \
    --cc=amit.kucheria-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org \
    --cc=davidb-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org \
    --cc=devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org \
    --cc=kgene.kim-Sze3O3UU22JBDgjK7y7TUQ@public.gmane.org \
    --cc=linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org \
    --cc=linux-lFZ/pmaqli7XmaaqVzeoHQ@public.gmane.org \
    --cc=nicolas.pitre-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org \
    --cc=rob.herring-bsGFqQB8/DxBDgjK7y7TUQ@public.gmane.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 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).