Message-ID: <56E6AC3E.3030607@arm.com>
Date:	Mon, 14 Mar 2016 12:19:10 +0000
From:	Julien Grall <julien.grall@....com>
To:	Christoffer Dall <christoffer.dall@...aro.org>
Cc:	kvmarm@...ts.cs.columbia.edu, marc.zyngier@....com,
	fu.wei@...aro.org, kvm@...r.kernel.org,
	linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
	wei@...hat.com, al.stone@...aro.org,
	Thomas Gleixner <tglx@...utronix.de>,
	Jason Cooper <jason@...edaemon.net>
Subject: Re: [PATCH v3 4/9] irqchip/gic-v2: Parse and export virtual GIC
 information

Hi Christoffer,

On 09/03/16 05:14, Christoffer Dall wrote:
> On Tue, Mar 08, 2016 at 11:29:28AM +0000, Julien Grall wrote:
>> For now, the firmware tables are parsed 2 times: once in the GIC
>> drivers, the other timer when initializing the vGIC. It means code
>> duplication and make more tedious to add the support for another
>> firmware table (like ACPI).
>>
>> Introduce a new structure and set of helpers to get/set the virtual GIC
>> information. Also fill up the structure for GICv2.
>>
>> Signed-off-by: Julien Grall <julien.grall@....com>
>>
>> ---
>> Cc: Thomas Gleixner <tglx@...utronix.de>
>> Cc: Jason Cooper <jason@...edaemon.net>
>> Cc: Marc Zyngier <marc.zyngier@....com>
>>
>>      Changes in v2:
>>          - Use 0 rather than a negative value to know when the maintenance IRQ
>>          is not present.
>>          - Use resource for vcpu and vctrl
>> ---
>>   drivers/irqchip/irq-gic-common.c       | 13 ++++++
>>   drivers/irqchip/irq-gic-common.h       |  3 ++
>>   drivers/irqchip/irq-gic.c              | 80 +++++++++++++++++++++++++++++++++-
>>   include/linux/irqchip/arm-gic-common.h | 33 ++++++++++++++
>>   4 files changed, 128 insertions(+), 1 deletion(-)
>>   create mode 100644 include/linux/irqchip/arm-gic-common.h
>>
>> diff --git a/drivers/irqchip/irq-gic-common.c b/drivers/irqchip/irq-gic-common.c
>> index f174ce0..704caf4 100644
>> --- a/drivers/irqchip/irq-gic-common.c
>> +++ b/drivers/irqchip/irq-gic-common.c
>> @@ -21,6 +21,19 @@
>>
>>   #include "irq-gic-common.h"
>>
>> +static const struct gic_kvm_info *gic_kvm_info;
>> +
>> +const struct gic_kvm_info *gic_get_kvm_info(void)
>> +{
>> +	return gic_kvm_info;
>> +}
>> +
>> +void gic_set_kvm_info(const struct gic_kvm_info *info)
>> +{
>> +	WARN(gic_kvm_info != NULL, "gic_kvm_info already set\n");
>
> why do we WARN here?  Wouldn't this be an obvious bug?

Right this is an obvious bug. I will switch to BUG_ON.

>
>> +	gic_kvm_info = info;
>> +}
>> +
>>   void gic_enable_quirks(u32 iidr, const struct gic_quirk *quirks,
>>   		void *data)
>>   {
>> diff --git a/drivers/irqchip/irq-gic-common.h b/drivers/irqchip/irq-gic-common.h
>> index fff697d..205e5fd 100644
>> --- a/drivers/irqchip/irq-gic-common.h
>> +++ b/drivers/irqchip/irq-gic-common.h
>> @@ -19,6 +19,7 @@
>>
>>   #include <linux/of.h>
>>   #include <linux/irqdomain.h>
>> +#include <linux/irqchip/arm-gic-common.h>
>>
>>   struct gic_quirk {
>>   	const char *desc;
>> @@ -35,4 +36,6 @@ void gic_cpu_config(void __iomem *base, void (*sync_access)(void));
>>   void gic_enable_quirks(u32 iidr, const struct gic_quirk *quirks,
>>   		void *data);
>>
>> +void gic_set_kvm_info(const struct gic_kvm_info *info);
>> +
>>   #endif /* _IRQ_GIC_COMMON_H */
>> diff --git a/drivers/irqchip/irq-gic.c b/drivers/irqchip/irq-gic.c
>> index fbde202..0c58112 100644
>> --- a/drivers/irqchip/irq-gic.c
>> +++ b/drivers/irqchip/irq-gic.c
>> @@ -102,6 +102,8 @@ static struct static_key supports_deactivate = STATIC_KEY_INIT_TRUE;
>>
>>   static struct gic_chip_data gic_data[CONFIG_ARM_GIC_MAX_NR] __read_mostly;
>>
>> +static struct gic_kvm_info gic_v2_kvm_info;
>> +
>>   #ifdef CONFIG_GIC_NON_BANKED
>>   static void __iomem *gic_get_percpu_base(union gic_base *base)
>>   {
>> @@ -1189,6 +1191,35 @@ static bool gic_check_eoimode(struct device_node *node, void __iomem **base)
>>   	return true;
>>   }
>>
>> +static void __init gic_of_setup_kvm_info(struct device_node *node)
>> +{
>> +	int ret;
>> +	struct resource r;
>
> I would prefer two struct resource variables more akin to how the KVM
> code did it already, and then name them something more understandable
> than 'r'.
>
> You could also argue that you can then only populate the gic_v2_kvm_info
> with all coherent info or nothing, instead of filling it out partially
> and then exiting.
>
>> +
>> +	gic_v2_kvm_info.type = GIC_V2;
>> +
>> +	gic_v2_kvm_info.maint_irq = irq_of_parse_and_map(node, 0);
>> +
>> +	ret = of_address_to_resource(node, 2, &r);
>> +	if (!ret)
>> +		gic_v2_kvm_info.vctrl = r;
>> +
>> +	ret = of_address_to_resource(node, 3, &r);
>
> here you're overwriting the error return value if the first call to
> of_address_to_resource failed ?
>
>> +	if (!ret) {
>> +		if (!PAGE_ALIGNED(r.start))
>> +			pr_warn("GICV physical address 0x%llx not page aligned\n",
>> +				(unsigned long long)r.start);
>
> how does KVM know that this went bad?
>
>> +		else if (!PAGE_ALIGNED(resource_size(&r)))
>> +			pr_warn("GICV size 0x%llx not a multiple of page size 0x%lx\n",
>> +				(unsigned long long)resource_size(&r),
>> +				PAGE_SIZE);
>
> same?
>
>> +		else
>> +			gic_v2_kvm_info.vcpu = r;
>> +	}
>> +
>> +	gic_set_kvm_info(&gic_v2_kvm_info);
>
> so here you're setting the kvm info even if one of the calls above
> fails?
>
> I think this function could leverage much more of the existing KVM
> implementation to avoid these kinds of mistakes.

I will rework this function.

>
>> +}
>> +
>>   int __init
>>   gic_of_init(struct device_node *node, struct device_node *parent)
>>   {
>> @@ -1218,8 +1249,10 @@ gic_of_init(struct device_node *node, struct device_node *parent)
>>
>>   	__gic_init_bases(gic_cnt, -1, dist_base, cpu_base, percpu_offset,
>>   			 &node->fwnode);
>> -	if (!gic_cnt)
>> +	if (!gic_cnt) {
>>   		gic_init_physaddr(node);
>> +		gic_of_setup_kvm_info(node);
>> +	}
>>
>>   	if (parent) {
>>   		irq = irq_of_parse_and_map(node, 0);
>> @@ -1248,6 +1281,10 @@ IRQCHIP_DECLARE(pl390, "arm,pl390", gic_of_init);
>>   static struct
>>   {
>>   	phys_addr_t cpu_phy_base;
>> +	u32 maint_irq;
>> +	int maint_irq_mode;
>> +	phys_addr_t vctrl_base;
>> +	phys_addr_t vcpu_base;
>>   } acpi_data __initdata;
>>
>>   static int __init
>> @@ -1272,6 +1309,12 @@ gic_acpi_parse_madt_cpu(struct acpi_subtable_header *header,
>>   		return -EINVAL;
>>
>>   	acpi_data.cpu_phy_base = gic_cpu_base;
>> +	acpi_data.maint_irq = processor->vgic_interrupt;
>> +	acpi_data.maint_irq_mode = (processor->flags & ACPI_MADT_VGIC_IRQ_MODE) ?
>> +				    ACPI_EDGE_SENSITIVE : ACPI_LEVEL_SENSITIVE;
>
> I can only assume that bit 2 set means edge, and the default is leve,
> but I really don't know how ACPI works here.

See Table 5.62 in ACPI 6.0.

>
>> +	acpi_data.vctrl_base = processor->gich_base_address;
>> +	acpi_data.vcpu_base = processor->gicv_base_address;
>> +
>>   	cpu_base_assigned = 1;
>>   	return 0;
>>   }
>> @@ -1302,6 +1345,39 @@ static bool __init gic_validate_dist(struct acpi_subtable_header *header,
>>
>>   #define ACPI_GICV2_DIST_MEM_SIZE	(SZ_4K)
>>   #define ACPI_GIC_CPU_IF_MEM_SIZE	(SZ_8K)
>> +#define ACPI_GICV2_VCTRL_MEM_SIZE	(SZ_4K)
>> +#define ACPI_GICV2_VCPU_MEM_SIZE	(SZ_8K)
>> +
>> +static void __init gic_acpi_setup_kvm_info(void)
>> +{
>> +	int irq;
>> +
>> +	gic_v2_kvm_info.type = GIC_V2;
>> +
>> +	irq = acpi_register_gsi(NULL, acpi_data.maint_irq,
>> +				acpi_data.maint_irq_mode,
>> +				ACPI_ACTIVE_HIGH);
>> +	if (irq > 0)
>> +		gic_v2_kvm_info.maint_irq = irq;
>> +
>> +	if (acpi_data.vctrl_base) {
>> +		struct resource *vctrl = &gic_v2_kvm_info.vctrl;
>> +
>> +		vctrl->flags = IORESOURCE_MEM;
>> +		vctrl->start = acpi_data.vctrl_base;
>> +		vctrl->end = vctrl->start + ACPI_GICV2_VCTRL_MEM_SIZE - 1;
>> +	}
>> +
>> +	if (acpi_data.vcpu_base) {
>> +		struct resource *vcpu = &gic_v2_kvm_info.vcpu;
>> +
>> +		vcpu->flags = IORESOURCE_MEM;
>> +		vcpu->start = acpi_data.vcpu_base;
>> +		vcpu->end = vcpu->start + ACPI_GICV2_VCPU_MEM_SIZE - 1;
>> +	}
>> +
>> +	gic_set_kvm_info(&gic_v2_kvm_info);
>
> again, if something fails above, you're still setting the struct
> pointer.
>
> how will KVM know?

I will rework the function.

>
>> +}
>>
>>   static int __init gic_v2_acpi_init(struct acpi_subtable_header *header,
>>   				   const unsigned long end)
>> @@ -1359,6 +1435,8 @@ static int __init gic_v2_acpi_init(struct acpi_subtable_header *header,
>>   	if (IS_ENABLED(CONFIG_ARM_GIC_V2M))
>>   		gicv2m_init(NULL, gic_data[0].domain);
>>
>> +	gic_acpi_setup_kvm_info();
>> +
>>   	return 0;
>>   }
>>   IRQCHIP_ACPI_DECLARE(gic_v2, ACPI_MADT_TYPE_GENERIC_DISTRIBUTOR,
>> diff --git a/include/linux/irqchip/arm-gic-common.h b/include/linux/irqchip/arm-gic-common.h
>> new file mode 100644
>> index 0000000..f7723b9
>> --- /dev/null
>> +++ b/include/linux/irqchip/arm-gic-common.h
>> @@ -0,0 +1,33 @@
>> +/*
>> + * include/linux/irqchip/arm-gic-common.h
>> + *
>> + * Copyright (C) 2016 ARM Limited, All Rights Reserved.
>> + *
>> + * This program is free software; you can redistribute it and/or modify
>> + * it under the terms of the GNU General Public License version 2 as
>> + * published by the Free Software Foundation.
>> + */
>> +#ifndef __LINUX_IRQCHIP_ARM_GIC_COMMON_H
>> +#define __LINUX_IRQCHIP_ARM_GIC_COMMON_H
>> +
>> +#include <linux/types.h>
>> +#include <linux/ioport.h>
>> +
>> +enum gic_type {
>> +	GIC_V2,
>> +};
>> +
>> +struct gic_kvm_info {
>> +	/* GIC type */
>> +	enum gic_type	type;
>> +	/* Physical address & size of virtual cpu interface */
>
> is this a bitwise '&' or did you want to save two ascii characters by
> not writing 'and' ?  If the latter, please write 'and'.

I will write 'and'

>
>> +	struct resource vcpu;
>> +	/* Interrupt number */
>
> if you're going to comment on the fact that this is an interrupt number,
> I would be more interested in knowing if it's a Linux IRQ number or the
> INTID...
>
>> +	unsigned int	maint_irq;
>> +	/* Physical address & size of virtual control interface */
>
> same here

Ditto

>
>> +	struct resource vctrl;
>
> I'm actually not convinced the comments on this struct help a lot, but
> ok.

I can rename to "Virtual control interface".

Regards,

-- 
Julien Grall

Powered by blists - more mailing lists