linux-kernel - Re: [PATCH 3/3] irqchip: al-fic: Introducing support for MSI-X

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 5 Jun 2019 09:59:58 +0200
From:   Greg KH <gregkh@...uxfoundation.org>
To:     Talel Shenhar <talel@...zon.com>
Cc:     nicolas.ferre@...rochip.com, jason@...edaemon.net,
        marc.zyngier@....com, mark.rutland@....com,
        mchehab+samsung@...nel.org, robh+dt@...nel.org,
        davem@...emloft.net, shawn.lin@...k-chips.com, tglx@...utronix.de,
        devicetree@...r.kernel.org, linux-kernel@...r.kernel.org,
        dwmw@...zon.co.uk, benh@...nel.crashing.org, jonnyc@...zon.com,
        hhhawa@...zon.com, ronenk@...zon.com, hanochu@...zon.com,
        barakw@...zon.com
Subject: Re: [PATCH 3/3] irqchip: al-fic: Introducing support for MSI-X

On Wed, Jun 05, 2019 at 09:54:13AM +0300, Talel Shenhar wrote:
> The FIC supports either a (single) wired output, or generation of an MSI-X
> interrupt per input (for cases where it is embedded in a PCIe device,
> hence, allowing the PCIe drivers to call this API).
> This patch introduces the support for allowing the configuration of MSI-X
> instead of a wire interrupt.
> 
> Signed-off-by: Talel Shenhar <talel@...zon.com>
> ---
>  drivers/irqchip/irq-al-fic.c   | 48 +++++++++++++++++++++++++++++++++++++++---
>  include/linux/irqchip/al-fic.h |  2 ++
>  2 files changed, 47 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/irqchip/irq-al-fic.c b/drivers/irqchip/irq-al-fic.c
> index d881d42..e49b912 100644
> --- a/drivers/irqchip/irq-al-fic.c
> +++ b/drivers/irqchip/irq-al-fic.c
> @@ -19,6 +19,7 @@
>  #define AL_FIC_MASK		0x10
>  #define AL_FIC_CONTROL		0x28
>  
> +#define CONTROL_AUTO_CLEAR	BIT(2)
>  #define CONTROL_TRIGGER_RISING	BIT(3)
>  #define CONTROL_MASK_MSI_X	BIT(5)
>  
> @@ -193,9 +194,11 @@ struct irq_domain *al_fic_wire_get_domain(struct al_fic *fic)
>  }
>  EXPORT_SYMBOL_GPL(al_fic_wire_get_domain);
>  
> -static void al_fic_hw_init(struct al_fic *fic)
> +static void al_fic_hw_init(struct al_fic *fic,
> +			   int use_msi)
>  {
> -	u32 control = CONTROL_MASK_MSI_X;
> +	u32 control = (use_msi ? (CONTROL_AUTO_CLEAR | CONTROL_TRIGGER_RISING) :
> +		       CONTROL_MASK_MSI_X);
>  
>  	/* mask out all interrupts */
>  	writel(0xFFFFFFFF, fic->base + AL_FIC_MASK);
> @@ -240,7 +243,7 @@ struct al_fic *al_fic_wire_init(struct device_node *node,
>  	fic->parent_irq = parent_irq;
>  	fic->name = (name ?: "al-fic-wire");
>  
> -	al_fic_hw_init(fic);
> +	al_fic_hw_init(fic, false);
>  
>  	ret = al_fic_register(node, fic);
>  	if (ret) {
> @@ -260,6 +263,45 @@ struct al_fic *al_fic_wire_init(struct device_node *node,
>  EXPORT_SYMBOL_GPL(al_fic_wire_init);
>  
>  /**
> + * al_fic_msi_x_init() - initialize and configure fic in msi-x mode
> + * @base: mmio to fic register
> + * @name: name of the fic
> + *
> + * This API will configure the fic hardware to to work in msi-x mode.
> + * msi-x fic is to be configured for fics that are embedded inside AL PCIE EP.
> + * Those kind of fic are aware of the fact that they live inside PCIE and
> + * familiar with the MSI-X table which is configured as part of
> + * pci_enable_msix_range() and friends.
> + * Interrupt can be generated based on a positive edge or level - configuration
> + * is to be determined based on connected hardware to this fic.
> + *
> + * Returns pointer to fic context or ERR_PTR in case of error.
> + */
> +struct al_fic *al_fic_msi_x_init(void __iomem *base,
> +				 const char *name)
> +{
> +	struct al_fic *fic;
> +
> +	if (!base)
> +		return ERR_PTR(-EINVAL);
> +
> +	fic = kzalloc(sizeof(*fic), GFP_KERNEL);
> +	if (!fic)
> +		return ERR_PTR(-ENOMEM);
> +
> +	fic->base = base;
> +	fic->name = (name ?: "al-fic-full-fledged");
> +
> +	al_fic_hw_init(fic, true);
> +
> +	pr_debug("%s initialized successfully in Full-Fledged mode\n",
> +		 fic->name);
> +
> +	return fic;
> +}
> +EXPORT_SYMBOL_GPL(al_fic_msi_x_init);
> +
> +/**
>   * al_fic_cleanup() - free all resources allocated by fic
>   * @fic: pointer to fic context
>   *
> diff --git a/include/linux/irqchip/al-fic.h b/include/linux/irqchip/al-fic.h
> index 0833749..a2e89ff 100644
> --- a/include/linux/irqchip/al-fic.h
> +++ b/include/linux/irqchip/al-fic.h
> @@ -16,6 +16,8 @@ struct al_fic *al_fic_wire_init(struct device_node *node,
>  				void __iomem *base,
>  				const char *name,
>  				unsigned int parent_irq);
> +struct al_fic *al_fic_msi_x_init(void __iomem *base,
> +				 const char *name);

Who uses this new function?

thanks,

greg k-h