diff mbox

[02/10] powerpc/xive: guest exploitation of the XIVE interrupt controller

Message ID 1502182579-990-3-git-send-email-clg@kaod.org (mailing list archive)
State Superseded
Headers show

Commit Message

Cédric Le Goater Aug. 8, 2017, 8:56 a.m. UTC 
This is the framework for using XIVE in a PowerVM guest. The support
is very similar to the native one in a much simpler form.

Instead of OPAL calls, a set of Hypervisors call are used to configure
the interrupt sources and the event/notification queues of the guest:

 - H_INT_GET_SOURCE_INFO

   used to obtain the address of the MMIO page of the Event State
   Buffer (PQ bits) entry associated with the source.

 - H_INT_SET_SOURCE_CONFIG

   assigns a source to a "target".

 - H_INT_GET_SOURCE_CONFIG

   determines to which "target" and "priority" is assigned to a source

 - H_INT_GET_QUEUE_INFO

   returns the address of the notification management page associated
   with the specified "target" and "priority".

 - H_INT_SET_QUEUE_CONFIG

   sets or resets the event queue for a given "target" and "priority".
   It is also used to set the notification config associated with the
   queue, only unconditional notification for the moment.  Reset is
   performed with a queue size of 0 and queueing is disabled in that
   case.

 - H_INT_GET_QUEUE_CONFIG

   returns the queue settings for a given "target" and "priority".

 - H_INT_RESET

   resets all of the partition's interrupt exploitation structures to
   their initial state, losing all configuration set via the hcalls
   H_INT_SET_SOURCE_CONFIG and H_INT_SET_QUEUE_CONFIG.

 - H_INT_SYNC

   issue a synchronisation on a source to make sure sure all
   notifications have reached their queue.

As for XICS, the XIVE interface for the guest is described in the
device tree under the interrupt controller node. A couple of new
properties are specific to XIVE :

 - "reg"

   contains the base address and size of the thread interrupt
   managnement areas (TIMA) for the user level for the OS level. Only
   the OS level is taken into account.

 - "ibm,xive-eq-sizes"

   the size of the event queues.

 - "ibm,xive-lisn-ranges"

   the interrupt numbers ranges assigned to the guest. These are
   allocated using a simple bitmap.

Tested with a QEMU XIVE model for pseries and with the Power
hypervisor

Signed-off-by: Cédric Le Goater <clg@kaod.org>
---
 arch/powerpc/include/asm/hvcall.h            |  13 +-
 arch/powerpc/include/asm/xive.h              |   2 +
 arch/powerpc/platforms/pseries/Kconfig       |   1 +
 arch/powerpc/platforms/pseries/hotplug-cpu.c |  10 +-
 arch/powerpc/platforms/pseries/kexec.c       |   6 +-
 arch/powerpc/platforms/pseries/setup.c       |   8 +-
 arch/powerpc/platforms/pseries/smp.c         |  32 +-
 arch/powerpc/sysdev/xive/Kconfig             |   5 +
 arch/powerpc/sysdev/xive/Makefile            |   1 +
 arch/powerpc/sysdev/xive/spapr.c             | 554 +++++++++++++++++++++++++++
 10 files changed, 619 insertions(+), 13 deletions(-)
 create mode 100644 arch/powerpc/sysdev/xive/spapr.c

Comments

David Gibson Aug. 9, 2017, 3:53 a.m. UTC | #1 
On Tue, Aug 08, 2017 at 10:56:12AM +0200, Cédric Le Goater wrote:
> This is the framework for using XIVE in a PowerVM guest. The support
> is very similar to the native one in a much simpler form.
> 
> Instead of OPAL calls, a set of Hypervisors call are used to configure
> the interrupt sources and the event/notification queues of the guest:
> 
>  - H_INT_GET_SOURCE_INFO
> 
>    used to obtain the address of the MMIO page of the Event State
>    Buffer (PQ bits) entry associated with the source.
> 
>  - H_INT_SET_SOURCE_CONFIG
> 
>    assigns a source to a "target".
> 
>  - H_INT_GET_SOURCE_CONFIG
> 
>    determines to which "target" and "priority" is assigned to a source
> 
>  - H_INT_GET_QUEUE_INFO
> 
>    returns the address of the notification management page associated
>    with the specified "target" and "priority".
> 
>  - H_INT_SET_QUEUE_CONFIG
> 
>    sets or resets the event queue for a given "target" and "priority".
>    It is also used to set the notification config associated with the
>    queue, only unconditional notification for the moment.  Reset is
>    performed with a queue size of 0 and queueing is disabled in that
>    case.
> 
>  - H_INT_GET_QUEUE_CONFIG
> 
>    returns the queue settings for a given "target" and "priority".
> 
>  - H_INT_RESET
> 
>    resets all of the partition's interrupt exploitation structures to
>    their initial state, losing all configuration set via the hcalls
>    H_INT_SET_SOURCE_CONFIG and H_INT_SET_QUEUE_CONFIG.
> 
>  - H_INT_SYNC
> 
>    issue a synchronisation on a source to make sure sure all
>    notifications have reached their queue.
> 
> As for XICS, the XIVE interface for the guest is described in the
> device tree under the interrupt controller node. A couple of new
> properties are specific to XIVE :
> 
>  - "reg"
> 
>    contains the base address and size of the thread interrupt
>    managnement areas (TIMA) for the user level for the OS level. Only
>    the OS level is taken into account.
> 
>  - "ibm,xive-eq-sizes"
> 
>    the size of the event queues.
> 
>  - "ibm,xive-lisn-ranges"
> 
>    the interrupt numbers ranges assigned to the guest. These are
>    allocated using a simple bitmap.
> 
> Tested with a QEMU XIVE model for pseries and with the Power
> hypervisor
> 
> Signed-off-by: Cédric Le Goater <clg@kaod.org>

I don't know XIVE well enough to review in detail, but I've made some
comments based on general knowledge.

> ---
>  arch/powerpc/include/asm/hvcall.h            |  13 +-
>  arch/powerpc/include/asm/xive.h              |   2 +
>  arch/powerpc/platforms/pseries/Kconfig       |   1 +
>  arch/powerpc/platforms/pseries/hotplug-cpu.c |  10 +-
>  arch/powerpc/platforms/pseries/kexec.c       |   6 +-
>  arch/powerpc/platforms/pseries/setup.c       |   8 +-
>  arch/powerpc/platforms/pseries/smp.c         |  32 +-
>  arch/powerpc/sysdev/xive/Kconfig             |   5 +
>  arch/powerpc/sysdev/xive/Makefile            |   1 +
>  arch/powerpc/sysdev/xive/spapr.c             | 554 +++++++++++++++++++++++++++
>  10 files changed, 619 insertions(+), 13 deletions(-)
>  create mode 100644 arch/powerpc/sysdev/xive/spapr.c
> 
> diff --git a/arch/powerpc/include/asm/hvcall.h b/arch/powerpc/include/asm/hvcall.h
> index 57d38b504ff7..3d34dc0869f6 100644
> --- a/arch/powerpc/include/asm/hvcall.h
> +++ b/arch/powerpc/include/asm/hvcall.h
> @@ -280,7 +280,18 @@
>  #define H_RESIZE_HPT_COMMIT	0x370
>  #define H_REGISTER_PROC_TBL	0x37C
>  #define H_SIGNAL_SYS_RESET	0x380
> -#define MAX_HCALL_OPCODE	H_SIGNAL_SYS_RESET
> +#define H_INT_GET_SOURCE_INFO   0x3A8
> +#define H_INT_SET_SOURCE_CONFIG 0x3AC
> +#define H_INT_GET_SOURCE_CONFIG 0x3B0
> +#define H_INT_GET_QUEUE_INFO    0x3B4
> +#define H_INT_SET_QUEUE_CONFIG  0x3B8
> +#define H_INT_GET_QUEUE_CONFIG  0x3BC
> +#define H_INT_SET_OS_REPORTING_LINE 0x3C0
> +#define H_INT_GET_OS_REPORTING_LINE 0x3C4
> +#define H_INT_ESB               0x3C8
> +#define H_INT_SYNC              0x3CC
> +#define H_INT_RESET             0x3D0
> +#define MAX_HCALL_OPCODE	H_INT_RESET
>  
>  /* H_VIOCTL functions */
>  #define H_GET_VIOA_DUMP_SIZE	0x01
> diff --git a/arch/powerpc/include/asm/xive.h b/arch/powerpc/include/asm/xive.h
> index c23ff4389ca2..1deb10032d61 100644
> --- a/arch/powerpc/include/asm/xive.h
> +++ b/arch/powerpc/include/asm/xive.h
> @@ -110,6 +110,7 @@ extern bool __xive_enabled;
>  
>  static inline bool xive_enabled(void) { return __xive_enabled; }
>  
> +extern bool xive_spapr_init(void);
>  extern bool xive_native_init(void);
>  extern void xive_smp_probe(void);
>  extern int  xive_smp_prepare_cpu(unsigned int cpu);
> @@ -147,6 +148,7 @@ extern int xive_native_get_vp_info(u32 vp_id, u32 *out_cam_id, u32 *out_chip_id)
>  
>  static inline bool xive_enabled(void) { return false; }
>  
> +static inline bool xive_spapr_init(void) { return false; }
>  static inline bool xive_native_init(void) { return false; }
>  static inline void xive_smp_probe(void) { }
>  extern inline int  xive_smp_prepare_cpu(unsigned int cpu) { return -EINVAL; }
> diff --git a/arch/powerpc/platforms/pseries/Kconfig b/arch/powerpc/platforms/pseries/Kconfig
> index 3a6dfd14f64b..71dd69d9ec64 100644
> --- a/arch/powerpc/platforms/pseries/Kconfig
> +++ b/arch/powerpc/platforms/pseries/Kconfig
> @@ -7,6 +7,7 @@ config PPC_PSERIES
>  	select PCI
>  	select PCI_MSI
>  	select PPC_XICS
> +	select PPC_XIVE_SPAPR
>  	select PPC_ICP_NATIVE
>  	select PPC_ICP_HV
>  	select PPC_ICS_RTAS
> diff --git a/arch/powerpc/platforms/pseries/hotplug-cpu.c b/arch/powerpc/platforms/pseries/hotplug-cpu.c
> index 6afd1efd3633..de0a5c1d7b29 100644
> --- a/arch/powerpc/platforms/pseries/hotplug-cpu.c
> +++ b/arch/powerpc/platforms/pseries/hotplug-cpu.c
> @@ -34,6 +34,7 @@
>  #include <asm/machdep.h>
>  #include <asm/vdso_datapage.h>
>  #include <asm/xics.h>
> +#include <asm/xive.h>
>  #include <asm/plpar_wrappers.h>
>  
>  #include "pseries.h"
> @@ -109,7 +110,9 @@ static void pseries_mach_cpu_die(void)
>  
>  	local_irq_disable();
>  	idle_task_exit();
> -	xics_teardown_cpu();
> +	/* Nothing TODO for xive ? */;

You have a XIVE teardown_cpu() method below - should you be calling
that here, even though it's a no-op for now?

> +	if (!xive_enabled())
> +		xics_teardown_cpu();
>  
>  	if (get_preferred_offline_state(cpu) == CPU_STATE_INACTIVE) {
>  		set_cpu_current_state(cpu, CPU_STATE_INACTIVE);
> @@ -174,7 +177,10 @@ static int pseries_cpu_disable(void)
>  		boot_cpuid = cpumask_any(cpu_online_mask);
>  
>  	/* FIXME: abstract this to not be platform specific later on */
> -	xics_migrate_irqs_away();
> +	if (xive_enabled())
> +		xive_smp_disable_cpu();
> +	else
> +		xics_migrate_irqs_away();
>  	return 0;
>  }
>  
> diff --git a/arch/powerpc/platforms/pseries/kexec.c b/arch/powerpc/platforms/pseries/kexec.c
> index 6681ac97fb18..eeb13429d685 100644
> --- a/arch/powerpc/platforms/pseries/kexec.c
> +++ b/arch/powerpc/platforms/pseries/kexec.c
> @@ -15,6 +15,7 @@
>  #include <asm/firmware.h>
>  #include <asm/kexec.h>
>  #include <asm/xics.h>
> +#include <asm/xive.h>
>  #include <asm/smp.h>
>  #include <asm/plpar_wrappers.h>
>  
> @@ -51,5 +52,8 @@ void pseries_kexec_cpu_down(int crash_shutdown, int secondary)
>  		}
>  	}
>  
> -	xics_kexec_teardown_cpu(secondary);
> +	if (xive_enabled())
> +		xive_kexec_teardown_cpu(secondary);
> +	else
> +		xics_kexec_teardown_cpu(secondary);
>  }
> diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c
> index b5d86426e97b..a8531e012658 100644
> --- a/arch/powerpc/platforms/pseries/setup.c
> +++ b/arch/powerpc/platforms/pseries/setup.c
> @@ -57,6 +57,7 @@
>  #include <asm/nvram.h>
>  #include <asm/pmc.h>
>  #include <asm/xics.h>
> +#include <asm/xive.h>
>  #include <asm/ppc-pci.h>
>  #include <asm/i8259.h>
>  #include <asm/udbg.h>
> @@ -176,8 +177,11 @@ static void __init pseries_setup_i8259_cascade(void)
>  
>  static void __init pseries_init_irq(void)
>  {
> -	xics_init();
> -	pseries_setup_i8259_cascade();
> +	/* Try using a XIVE if available, otherwise use a XICS */
> +	if (!xive_spapr_init()) {
> +		xics_init();
> +		pseries_setup_i8259_cascade();
> +	}
>  }
>  
>  static void pseries_lpar_enable_pmcs(void)
> diff --git a/arch/powerpc/platforms/pseries/smp.c b/arch/powerpc/platforms/pseries/smp.c
> index 24785f63fb40..244bb9974963 100644
> --- a/arch/powerpc/platforms/pseries/smp.c
> +++ b/arch/powerpc/platforms/pseries/smp.c
> @@ -41,6 +41,7 @@
>  #include <asm/vdso_datapage.h>
>  #include <asm/cputhreads.h>
>  #include <asm/xics.h>
> +#include <asm/xive.h>
>  #include <asm/dbell.h>
>  #include <asm/plpar_wrappers.h>
>  #include <asm/code-patching.h>
> @@ -136,7 +137,9 @@ static inline int smp_startup_cpu(unsigned int lcpu)
>  
>  static void smp_setup_cpu(int cpu)
>  {
> -	if (cpu != boot_cpuid)
> +	if (xive_enabled())
> +		xive_smp_setup_cpu();
> +	else if (cpu != boot_cpuid)
>  		xics_setup_cpu();
>  
>  	if (firmware_has_feature(FW_FEATURE_SPLPAR))
> @@ -181,13 +184,22 @@ static int smp_pSeries_kick_cpu(int nr)
>  	return 0;
>  }
>  
> +static int pseries_smp_prepare_cpu(int cpu)
> +{
> +	if (xive_enabled())
> +		return xive_smp_prepare_cpu(cpu);
> +	return 0;
> +}
> +
> +/* Cause IPI as setup by the interrupt controller (xics or xive) */
> +static void (*ic_cause_ipi)(int cpu);
> +
>  static void smp_pseries_cause_ipi(int cpu)
>  {
> -	/* POWER9 should not use this handler */
>  	if (doorbell_try_core_ipi(cpu))
>  		return;
>  
> -	icp_ops->cause_ipi(cpu);
> +	ic_cause_ipi(cpu);

Wouldn't it make more sense to change smp_ops->cause_ipi, rather than
having a double indirection through smp_ops, then the ic_cause_ipi
global?

>  }
>  
>  static int pseries_cause_nmi_ipi(int cpu)
> @@ -213,12 +225,17 @@ static int pseries_cause_nmi_ipi(int cpu)
>  
>  static __init void pSeries_smp_probe(void)
>  {
> -	xics_smp_probe();
> +	if (xive_enabled())
> +		xive_smp_probe();
> +	else
> +		xics_smp_probe();
> +
> +	if (cpu_has_feature(CPU_FTR_DBELL)) {
> +		ic_cause_ipi = smp_ops->cause_ipi;
> +		WARN_ON(!ic_cause_ipi);
>  
> -	if (cpu_has_feature(CPU_FTR_DBELL))
>  		smp_ops->cause_ipi = smp_pseries_cause_ipi;
> -	else
> -		smp_ops->cause_ipi = icp_ops->cause_ipi;
> +	}
>  }
>  
>  static struct smp_ops_t pseries_smp_ops = {
> @@ -226,6 +243,7 @@ static struct smp_ops_t pseries_smp_ops = {
>  	.cause_ipi	= NULL,	/* Filled at runtime by pSeries_smp_probe() */
>  	.cause_nmi_ipi	= pseries_cause_nmi_ipi,
>  	.probe		= pSeries_smp_probe,
> +	.prepare_cpu	= pseries_smp_prepare_cpu,
>  	.kick_cpu	= smp_pSeries_kick_cpu,
>  	.setup_cpu	= smp_setup_cpu,
>  	.cpu_bootable	= smp_generic_cpu_bootable,
> diff --git a/arch/powerpc/sysdev/xive/Kconfig b/arch/powerpc/sysdev/xive/Kconfig
> index 12ccd7373d2f..3e3e25b5e30d 100644
> --- a/arch/powerpc/sysdev/xive/Kconfig
> +++ b/arch/powerpc/sysdev/xive/Kconfig
> @@ -9,3 +9,8 @@ config PPC_XIVE_NATIVE
>  	default n
>  	select PPC_XIVE
>  	depends on PPC_POWERNV
> +
> +config PPC_XIVE_SPAPR
> +	bool
> +	default n
> +	select PPC_XIVE
> diff --git a/arch/powerpc/sysdev/xive/Makefile b/arch/powerpc/sysdev/xive/Makefile
> index 3fab303fc169..536d6e5706e3 100644
> --- a/arch/powerpc/sysdev/xive/Makefile
> +++ b/arch/powerpc/sysdev/xive/Makefile
> @@ -2,3 +2,4 @@ subdir-ccflags-$(CONFIG_PPC_WERROR) := -Werror
>  
>  obj-y				+= common.o
>  obj-$(CONFIG_PPC_XIVE_NATIVE)	+= native.o
> +obj-$(CONFIG_PPC_XIVE_SPAPR)	+= spapr.o
> diff --git a/arch/powerpc/sysdev/xive/spapr.c b/arch/powerpc/sysdev/xive/spapr.c
> new file mode 100644
> index 000000000000..a3668815d5c1
> --- /dev/null
> +++ b/arch/powerpc/sysdev/xive/spapr.c
> @@ -0,0 +1,554 @@
> +/*
> + * Copyright 2016,2017 IBM Corporation.
> + *
> + * This program is free software; you can redistribute it and/or
> + * modify it under the terms of the GNU General Public License
> + * as published by the Free Software Foundation; either version
> + * 2 of the License, or (at your option) any later version.
> + */
> +
> +#define pr_fmt(fmt) "xive: " fmt
> +
> +#include <linux/types.h>
> +#include <linux/irq.h>
> +#include <linux/smp.h>
> +#include <linux/interrupt.h>
> +#include <linux/init.h>
> +#include <linux/of.h>
> +#include <linux/slab.h>
> +#include <linux/spinlock.h>
> +#include <linux/cpumask.h>
> +#include <linux/mm.h>
> +
> +#include <asm/prom.h>
> +#include <asm/io.h>
> +#include <asm/smp.h>
> +#include <asm/irq.h>
> +#include <asm/errno.h>
> +#include <asm/xive.h>
> +#include <asm/xive-regs.h>
> +#include <asm/hvcall.h>
> +
> +#include "xive-internal.h"
> +
> +static u32 xive_queue_shift;
> +
> +struct xive_irq_bitmap {
> +	unsigned long		*bitmap;
> +	unsigned int		base;
> +	unsigned int		count;
> +	spinlock_t		lock;
> +	struct list_head	list;
> +};
> +
> +static LIST_HEAD(xive_irq_bitmaps);
> +
> +static int xive_irq_bitmap_add(int base, int count)
> +{
> +	struct xive_irq_bitmap *xibm;
> +
> +	xibm = kzalloc(sizeof(*xibm), GFP_ATOMIC);
> +	if (!xibm)
> +		return -ENOMEM;
> +
> +	spin_lock_init(&xibm->lock);
> +	xibm->base = base;
> +	xibm->count = count;
> +	xibm->bitmap = kzalloc(xibm->count, GFP_KERNEL);
> +	list_add(&xibm->list, &xive_irq_bitmaps);
> +
> +	pr_info("Using IRQ range [%x-%x]", xibm->base,
> +		xibm->base + xibm->count - 1);
> +	return 0;
> +}
> +
> +static int __xive_irq_bitmap_alloc(struct xive_irq_bitmap *xibm)
> +{
> +	int irq;
> +
> +	irq = find_first_zero_bit(xibm->bitmap, xibm->count);
> +	if (irq != xibm->count) {
> +		set_bit(irq, xibm->bitmap);
> +		irq += xibm->base;
> +	} else {
> +		irq = -ENOMEM;
> +	}
> +
> +	return irq;
> +}
> +
> +static int xive_irq_bitmap_alloc(void)
> +{
> +	struct xive_irq_bitmap *xibm;
> +	unsigned long flags;
> +	int irq = -ENOENT;
> +
> +	list_for_each_entry(xibm, &xive_irq_bitmaps, list) {
> +		spin_lock_irqsave(&xibm->lock, flags);
> +		irq = __xive_irq_bitmap_alloc(xibm);
> +		spin_unlock_irqrestore(&xibm->lock, flags);
> +		if (irq >= 0)
> +			break;
> +	}
> +	return irq;
> +}
> +
> +static void xive_irq_bitmap_free(int irq)
> +{
> +	unsigned long flags;
> +	struct xive_irq_bitmap *xibm;
> +
> +	list_for_each_entry(xibm, &xive_irq_bitmaps, list) {
> +		if ((irq >= xibm->base) && (irq < xibm->base + xibm->count)) {
> +			spin_lock_irqsave(&xibm->lock, flags);
> +			clear_bit(irq - xibm->base, xibm->bitmap);
> +			spin_unlock_irqrestore(&xibm->lock, flags);
> +			break;
> +		}
> +	}
> +}
> +
> +static long plpar_int_get_source_info(unsigned long flags,
> +				      unsigned long lisn,
> +				      unsigned long *src_flags,
> +				      unsigned long *eoi_page,
> +				      unsigned long *trig_page,
> +				      unsigned long *esb_shift)
> +{
> +	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
> +	long rc;
> +
> +	rc = plpar_hcall(H_INT_GET_SOURCE_INFO, retbuf, flags, lisn);
> +	if (rc) {
> +		pr_err("H_INT_GET_SOURCE_INFO lisn=%ld failed %ld\n", lisn, rc);
> +		return rc;
> +	}
> +
> +	*src_flags = retbuf[0];
> +	*eoi_page  = retbuf[1];
> +	*trig_page = retbuf[2];
> +	*esb_shift = retbuf[3];
> +
> +	pr_devel("H_INT_GET_SOURCE_INFO flags=%lx eoi=%lx trig=%lx shift=%lx\n",
> +		retbuf[0], retbuf[1], retbuf[2], retbuf[3]);
> +
> +	return 0;
> +}
> +
> +#define XIVE_SRC_SET_EISN (1ull << (63 - 62))
> +#define XIVE_SRC_MASK     (1ull << (63 - 63)) /* unused */
> +
> +static long plpar_int_set_source_config(unsigned long flags,
> +					unsigned long lisn,
> +					unsigned long target,
> +					unsigned long prio,
> +					unsigned long sw_irq)
> +{
> +	long rc;
> +
> +
> +	pr_devel("H_INT_SET_SOURCE_CONFIG flags=%lx lisn=%lx target=%lx prio=%lx sw_irq=%lx\n",
> +		flags, lisn, target, prio, sw_irq);
> +
> +
> +	rc = plpar_hcall_norets(H_INT_SET_SOURCE_CONFIG, flags, lisn,
> +				target, prio, sw_irq);
> +	if (rc) {
> +		pr_err("H_INT_SET_SOURCE_CONFIG lisn=%ld failed %ld\n",
> +		       lisn, rc);
> +		return rc;
> +	}
> +
> +	return 0;
> +}
> +
> +static long plpar_int_get_queue_info(unsigned long flags,
> +				     unsigned long target,
> +				     unsigned long priority,
> +				     unsigned long *esn_page,
> +				     unsigned long *esn_size)
> +{
> +	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
> +	long rc;
> +
> +	rc = plpar_hcall(H_INT_GET_QUEUE_INFO, retbuf, flags, target, priority);
> +	if (rc) {
> +		pr_err("H_INT_GET_QUEUE_INFO cpu=%ld prio=%ld failed %ld\n",
> +		       target, priority, rc);
> +		return rc;
> +	}
> +
> +	*esn_page = retbuf[0];
> +	*esn_size = retbuf[1];
> +
> +	pr_devel("H_INT_GET_QUEUE_INFO page=%lx size=%lx\n",
> +		retbuf[0], retbuf[1]);
> +
> +	return 0;
> +}
> +
> +#define XIVE_EQ_ALWAYS_NOTIFY (1ull << (63 - 63))
> +
> +static long plpar_int_set_queue_config(unsigned long flags,
> +				       unsigned long target,
> +				       unsigned long priority,
> +				       unsigned long qpage,
> +				       unsigned long qsize)
> +{
> +	long rc;
> +
> +	pr_devel("H_INT_SET_QUEUE_CONFIG flags=%lx target=%lx priority=%lx qpage=%lx qsize=%lx\n",
> +		flags,  target, priority, qpage, qsize);
> +
> +	rc = plpar_hcall_norets(H_INT_SET_QUEUE_CONFIG, flags, target,
> +				priority, qpage, qsize);
> +	if (rc) {
> +		pr_err("H_INT_SET_QUEUE_CONFIG cpu=%ld prio=%ld qpage=%lx returned %ld\n",
> +		       target, priority, qpage, rc);
> +		return  rc;
> +	}
> +
> +	return 0;
> +}
> +
> +static long plpar_int_sync(unsigned long flags)
> +{
> +	long rc;
> +
> +	rc = plpar_hcall_norets(H_INT_SYNC, flags);
> +	if (rc) {
> +		pr_err("H_INT_SYNC returned %ld\n", rc);
> +		return  rc;
> +	}
> +
> +	return 0;
> +}
> +
> +#define XIVE_SRC_H_INT_ESB     (1ull << (63 - 60)) /* TODO */
> +#define XIVE_SRC_LSI           (1ull << (63 - 61))
> +#define XIVE_SRC_TRIGGER       (1ull << (63 - 62))
> +#define XIVE_SRC_STORE_EOI     (1ull << (63 - 63))
> +
> +static int xive_spapr_populate_irq_data(u32 hw_irq, struct xive_irq_data *data)
> +{
> +	long rc;
> +	unsigned long flags;
> +	unsigned long eoi_page;
> +	unsigned long trig_page;
> +	unsigned long esb_shift;
> +
> +	memset(data, 0, sizeof(*data));
> +
> +	rc = plpar_int_get_source_info(0, hw_irq, &flags, &eoi_page, &trig_page,
> +				       &esb_shift);
> +	if (rc)
> +		return  -EINVAL;
> +
> +	if (flags & XIVE_SRC_STORE_EOI)
> +		data->flags  |= XIVE_IRQ_FLAG_STORE_EOI;
> +	if (flags & XIVE_SRC_LSI)
> +		data->flags  |= XIVE_IRQ_FLAG_LSI;
> +	data->eoi_page  = eoi_page;
> +	data->esb_shift = esb_shift;
> +	data->trig_page = trig_page;
> +
> +	/*
> +	 * No chip-id for the sPAPR backend. This has an impact how we
> +	 * pick a target. See xive_pick_irq_target().
> +	 */
> +	data->src_chip = XIVE_INVALID_CHIP_ID;
> +
> +	data->eoi_mmio = ioremap(data->eoi_page, 1u << data->esb_shift);
> +	if (!data->eoi_mmio) {
> +		pr_err("Failed to map EOI page for irq 0x%x\n", hw_irq);
> +		return -ENOMEM;
> +	}
> +
> +	/* Full function page supports trigger */
> +	if (flags & XIVE_SRC_TRIGGER) {
> +		data->trig_mmio = data->eoi_mmio;
> +		return 0;
> +	}
> +
> +	data->trig_mmio = ioremap(data->trig_page, 1u << data->esb_shift);
> +	if (!data->trig_mmio) {
> +		pr_err("Failed to map trigger page for irq 0x%x\n", hw_irq);
> +		return -ENOMEM;
> +	}
> +	return 0;
> +}
> +
> +static int xive_spapr_configure_irq(u32 hw_irq, u32 target, u8 prio, u32 sw_irq)
> +{
> +	long rc;
> +
> +	rc = plpar_int_set_source_config(XIVE_SRC_SET_EISN, hw_irq, target,
> +					 prio, sw_irq);
> +
> +	return rc == 0 ? 0 : -ENXIO;
> +}
> +
> +/* This can be called multiple time to change a queue configuration */
> +static int xive_spapr_configure_queue(u32 target, struct xive_q *q, u8 prio,
> +				   __be32 *qpage, u32 order)
> +{
> +	s64 rc = 0;
> +	unsigned long esn_page;
> +	unsigned long esn_size;
> +	u64 flags, qpage_phys;
> +
> +	/* If there's an actual queue page, clean it */
> +	if (order) {
> +		if (WARN_ON(!qpage))
> +			return -EINVAL;
> +		qpage_phys = __pa(qpage);
> +	} else {
> +		qpage_phys = 0;
> +	}
> +
> +	/* Initialize the rest of the fields */
> +	q->msk = order ? ((1u << (order - 2)) - 1) : 0;
> +	q->idx = 0;
> +	q->toggle = 0;
> +
> +	rc = plpar_int_get_queue_info(0, target, prio, &esn_page, &esn_size);
> +	if (rc) {
> +		pr_err("Error %lld getting queue info prio %d\n", rc, prio);
> +		rc = -EIO;
> +		goto fail;
> +	}
> +
> +	/* TODO: add support for the notification page */
> +	q->eoi_phys = esn_page;
> +
> +	/* Default is to always notify */
> +	flags = XIVE_EQ_ALWAYS_NOTIFY;
> +
> +	/* Configure and enable the queue in HW */
> +	rc = plpar_int_set_queue_config(flags, target, prio, qpage_phys, order);
> +	if (rc) {
> +		pr_err("Error %lld setting queue for prio %d\n", rc, prio);
> +		rc = -EIO;
> +	} else {
> +		q->qpage = qpage;
> +	}
> +fail:
> +	return rc;
> +}
> +
> +static int xive_spapr_setup_queue(unsigned int cpu, struct xive_cpu *xc,
> +				  u8 prio)
> +{
> +	struct xive_q *q = &xc->queue[prio];
> +	unsigned int alloc_order;
> +	struct page *pages;
> +	__be32 *qpage;
> +
> +	alloc_order = (xive_queue_shift > PAGE_SHIFT) ?
> +		(xive_queue_shift - PAGE_SHIFT) : 0;
> +	pages = alloc_pages_node(cpu_to_node(cpu), GFP_KERNEL, alloc_order);
> +	if (!pages)
> +		return -ENOMEM;
> +	qpage = (__be32 *)page_address(pages);
> +	memset(qpage, 0, 1 << xive_queue_shift);
> +
> +	return xive_spapr_configure_queue(cpu, q, prio, qpage,
> +					  xive_queue_shift);
> +}
> +
> +static void xive_spapr_cleanup_queue(unsigned int cpu, struct xive_cpu *xc,
> +				  u8 prio)
> +{
> +	struct xive_q *q = &xc->queue[prio];
> +	unsigned int alloc_order;
> +	long rc;
> +
> +	rc = plpar_int_set_queue_config(0, cpu, prio, 0, 0);
> +	if (rc)
> +		pr_err("Error %ld setting queue for prio %d\n", rc, prio);
> +
> +	alloc_order = (xive_queue_shift > PAGE_SHIFT) ?
> +		(xive_queue_shift - PAGE_SHIFT) : 0;

Maybe a helper inline for this calculation, since it's used in both
setup_queue() and cleanup_queue?

> +	free_pages((unsigned long)q->qpage, alloc_order);
> +	q->qpage = NULL;
> +}
> +
> +static bool xive_spapr_match(struct device_node *node)
> +{
> +	return 1;
> +}
> +
> +#ifdef CONFIG_SMP
> +static int xive_spapr_get_ipi(unsigned int cpu, struct xive_cpu *xc)
> +{
> +	int irq = xive_irq_bitmap_alloc();
> +
> +	if (irq < 0) {
> +		pr_err("Failed to allocate IPI on CPU %d\n", cpu);
> +		return -ENXIO;
> +	}
> +
> +	xc->hw_ipi = irq;
> +	return 0;
> +}
> +
> +static void xive_spapr_put_ipi(unsigned int cpu, struct xive_cpu *xc)
> +{
> +	xive_irq_bitmap_free(xc->hw_ipi);
> +}
> +#endif /* CONFIG_SMP */
> +
> +static void xive_spapr_shutdown(void)
> +{
> +	long rc;
> +
> +	rc = plpar_hcall_norets(H_INT_RESET, 0);
> +	if (rc)
> +		pr_err("H_INT_RESET failed %ld\n", rc);
> +}
> +
> +static void xive_spapr_update_pending(struct xive_cpu *xc)
> +{
> +	u8 nsr, cppr;
> +	u16 ack;
> +
> +	/* Perform the acknowledge hypervisor to register cycle */
> +	ack = be16_to_cpu(__raw_readw(xive_tima + TM_SPC_ACK_OS_REG));

Why do you need the raw_readw() + be16_to_cpu + mb, rather than one of
the higher level IO helpers?

> +	/* Synchronize subsequent queue accesses */
> +	mb();
> +
> +	/*
> +	 * Grab the CPPR and the "NSR" field which indicates the source
> +	 * of the hypervisor interrupt (if any)
> +	 */
> +	cppr = ack & 0xff;
> +	nsr = ack >> 8;
> +
> +	if (nsr & TM_QW1_NSR_EO) {
> +		if (cppr == 0xff)
> +			return;
> +		/* Mark the priority pending */
> +		xc->pending_prio |= 1 << cppr;
> +
> +		/*
> +		 * A new interrupt should never have a CPPR less favored
> +		 * than our current one.
> +		 */
> +		if (cppr >= xc->cppr)
> +			pr_err("CPU %d odd ack CPPR, got %d at %d\n",
> +			       smp_processor_id(), cppr, xc->cppr);
> +
> +		/* Update our idea of what the CPPR is */
> +		xc->cppr = cppr;
> +	}
> +}
> +
> +static void xive_spapr_eoi(u32 hw_irq)
> +{
> +	/* Not used */;
> +}

Do you really want empty stub functions, or would it be better to have
the method caller check for NULL in the ops structure?

> +
> +static void xive_spapr_setup_cpu(unsigned int cpu, struct xive_cpu *xc)
> +{
> +	/* Only some debug on the TIMA settings */
> +	pr_debug("(HW value: %08x %08x %08x)\n",
> +		 in_be32(xive_tima + TM_QW1_OS + TM_WORD0),
> +		 in_be32(xive_tima + TM_QW1_OS + TM_WORD1),
> +		 in_be32(xive_tima + TM_QW1_OS + TM_WORD2));
> +}
> +
> +static void xive_spapr_teardown_cpu(unsigned int cpu, struct xive_cpu *xc)
> +{
> +	/* Nothing to do */;
> +}
> +
> +static void xive_spapr_sync_source(u32 hw_irq)
> +{
> +	/* Specs are unclear on what this is doing */
> +	plpar_int_sync(0);
> +}
> +
> +static const struct xive_ops xive_spapr_ops = {
> +	.populate_irq_data	= xive_spapr_populate_irq_data,
> +	.configure_irq		= xive_spapr_configure_irq,
> +	.setup_queue		= xive_spapr_setup_queue,
> +	.cleanup_queue		= xive_spapr_cleanup_queue,
> +	.match			= xive_spapr_match,
> +	.shutdown		= xive_spapr_shutdown,
> +	.update_pending		= xive_spapr_update_pending,
> +	.eoi			= xive_spapr_eoi,
> +	.setup_cpu		= xive_spapr_setup_cpu,
> +	.teardown_cpu		= xive_spapr_teardown_cpu,
> +	.sync_source		= xive_spapr_sync_source,
> +#ifdef CONFIG_SMP
> +	.get_ipi		= xive_spapr_get_ipi,
> +	.put_ipi		= xive_spapr_put_ipi,
> +#endif /* CONFIG_SMP */
> +	.name			= "spapr",
> +};
> +
> +bool xive_spapr_init(void)
> +{
> +	struct device_node *np;
> +	struct resource r;
> +	void __iomem *tima;
> +	struct property *prop;
> +	u8 max_prio = 7;
> +	u32 val;
> +	u32 len;
> +	const __be32 *reg;
> +	int i;
> +
> +	if (xive_cmdline_disabled)
> +		return false;
> +
> +	pr_devel("%s()\n", __func__);
> +	np = of_find_compatible_node(NULL, NULL, "ibm,power-ivpe");
> +	if (!np) {
> +		pr_devel("not found !\n");
> +		return false;
> +	}
> +	pr_devel("Found %s\n", np->full_name);
> +
> +	/* Resource 1 is the OS ring TIMA */
> +	if (of_address_to_resource(np, 1, &r)) {
> +		pr_err("Failed to get thread mgmnt area resource\n");
> +		return false;
> +	}
> +	tima = ioremap(r.start, resource_size(&r));
> +	if (!tima) {
> +		pr_err("Failed to map thread mgmnt area\n");
> +		return false;
> +	}
> +
> +	/* Feed the IRQ number allocator with the ranges given in the DT */
> +	reg = of_get_property(np, "ibm,xive-lisn-ranges", &len);
> +	if (!reg) {
> +		pr_err("Failed to read 'ibm,xive-lisn-ranges' property\n");
> +		return false;
> +	}
> +
> +	if (len % (2 * sizeof(u32)) != 0) {
> +		pr_err("invalid 'ibm,xive-lisn-ranges' property\n");
> +		return false;
> +	}
> +
> +	for (i = 0; i < len / (2 * sizeof(u32)); i++, reg += 2)
> +		xive_irq_bitmap_add(be32_to_cpu(reg[0]),
> +				    be32_to_cpu(reg[1]));
> +
> +	/* Iterate the EQ sizes and pick one */
> +	of_property_for_each_u32(np, "ibm,xive-eq-sizes", prop, reg, val) {
> +		xive_queue_shift = val;
> +		if (val == PAGE_SHIFT)
> +			break;
> +	}
> +
> +	/* Initialize XIVE core with our backend */
> +	if (!xive_core_init(&xive_spapr_ops, tima, TM_QW1_OS, max_prio))
> +		return false;
> +
> +	pr_info("Using %dkB queues\n", 1 << (xive_queue_shift - 10));
> +	return true;
> +}
Cédric Le Goater Aug. 9, 2017, 8:48 a.m. UTC | #2 
On 08/09/2017 05:53 AM, David Gibson wrote:
> On Tue, Aug 08, 2017 at 10:56:12AM +0200, Cédric Le Goater wrote:
>> This is the framework for using XIVE in a PowerVM guest. The support
>> is very similar to the native one in a much simpler form.
>>
>> Instead of OPAL calls, a set of Hypervisors call are used to configure
>> the interrupt sources and the event/notification queues of the guest:
>>
>>  - H_INT_GET_SOURCE_INFO
>>
>>    used to obtain the address of the MMIO page of the Event State
>>    Buffer (PQ bits) entry associated with the source.
>>
>>  - H_INT_SET_SOURCE_CONFIG
>>
>>    assigns a source to a "target".
>>
>>  - H_INT_GET_SOURCE_CONFIG
>>
>>    determines to which "target" and "priority" is assigned to a source
>>
>>  - H_INT_GET_QUEUE_INFO
>>
>>    returns the address of the notification management page associated
>>    with the specified "target" and "priority".
>>
>>  - H_INT_SET_QUEUE_CONFIG
>>
>>    sets or resets the event queue for a given "target" and "priority".
>>    It is also used to set the notification config associated with the
>>    queue, only unconditional notification for the moment.  Reset is
>>    performed with a queue size of 0 and queueing is disabled in that
>>    case.
>>
>>  - H_INT_GET_QUEUE_CONFIG
>>
>>    returns the queue settings for a given "target" and "priority".
>>
>>  - H_INT_RESET
>>
>>    resets all of the partition's interrupt exploitation structures to
>>    their initial state, losing all configuration set via the hcalls
>>    H_INT_SET_SOURCE_CONFIG and H_INT_SET_QUEUE_CONFIG.
>>
>>  - H_INT_SYNC
>>
>>    issue a synchronisation on a source to make sure sure all
>>    notifications have reached their queue.
>>
>> As for XICS, the XIVE interface for the guest is described in the
>> device tree under the interrupt controller node. A couple of new
>> properties are specific to XIVE :
>>
>>  - "reg"
>>
>>    contains the base address and size of the thread interrupt
>>    managnement areas (TIMA) for the user level for the OS level. Only
>>    the OS level is taken into account.
>>
>>  - "ibm,xive-eq-sizes"
>>
>>    the size of the event queues.
>>
>>  - "ibm,xive-lisn-ranges"
>>
>>    the interrupt numbers ranges assigned to the guest. These are
>>    allocated using a simple bitmap.
>>
>> Tested with a QEMU XIVE model for pseries and with the Power
>> hypervisor
>>
>> Signed-off-by: Cédric Le Goater <clg@kaod.org>
> 
> I don't know XIVE well enough to review in detail, but I've made some
> comments based on general knowledge.

Thanks for taking a look. 
 
> 
>> ---
>>  arch/powerpc/include/asm/hvcall.h            |  13 +-
>>  arch/powerpc/include/asm/xive.h              |   2 +
>>  arch/powerpc/platforms/pseries/Kconfig       |   1 +
>>  arch/powerpc/platforms/pseries/hotplug-cpu.c |  10 +-
>>  arch/powerpc/platforms/pseries/kexec.c       |   6 +-
>>  arch/powerpc/platforms/pseries/setup.c       |   8 +-
>>  arch/powerpc/platforms/pseries/smp.c         |  32 +-
>>  arch/powerpc/sysdev/xive/Kconfig             |   5 +
>>  arch/powerpc/sysdev/xive/Makefile            |   1 +
>>  arch/powerpc/sysdev/xive/spapr.c             | 554 +++++++++++++++++++++++++++
>>  10 files changed, 619 insertions(+), 13 deletions(-)
>>  create mode 100644 arch/powerpc/sysdev/xive/spapr.c
>>
>> diff --git a/arch/powerpc/include/asm/hvcall.h b/arch/powerpc/include/asm/hvcall.h
>> index 57d38b504ff7..3d34dc0869f6 100644
>> --- a/arch/powerpc/include/asm/hvcall.h
>> +++ b/arch/powerpc/include/asm/hvcall.h
>> @@ -280,7 +280,18 @@
>>  #define H_RESIZE_HPT_COMMIT	0x370
>>  #define H_REGISTER_PROC_TBL	0x37C
>>  #define H_SIGNAL_SYS_RESET	0x380
>> -#define MAX_HCALL_OPCODE	H_SIGNAL_SYS_RESET
>> +#define H_INT_GET_SOURCE_INFO   0x3A8
>> +#define H_INT_SET_SOURCE_CONFIG 0x3AC
>> +#define H_INT_GET_SOURCE_CONFIG 0x3B0
>> +#define H_INT_GET_QUEUE_INFO    0x3B4
>> +#define H_INT_SET_QUEUE_CONFIG  0x3B8
>> +#define H_INT_GET_QUEUE_CONFIG  0x3BC
>> +#define H_INT_SET_OS_REPORTING_LINE 0x3C0
>> +#define H_INT_GET_OS_REPORTING_LINE 0x3C4
>> +#define H_INT_ESB               0x3C8
>> +#define H_INT_SYNC              0x3CC
>> +#define H_INT_RESET             0x3D0
>> +#define MAX_HCALL_OPCODE	H_INT_RESET
>>  
>>  /* H_VIOCTL functions */
>>  #define H_GET_VIOA_DUMP_SIZE	0x01
>> diff --git a/arch/powerpc/include/asm/xive.h b/arch/powerpc/include/asm/xive.h
>> index c23ff4389ca2..1deb10032d61 100644
>> --- a/arch/powerpc/include/asm/xive.h
>> +++ b/arch/powerpc/include/asm/xive.h
>> @@ -110,6 +110,7 @@ extern bool __xive_enabled;
>>  
>>  static inline bool xive_enabled(void) { return __xive_enabled; }
>>  
>> +extern bool xive_spapr_init(void);
>>  extern bool xive_native_init(void);
>>  extern void xive_smp_probe(void);
>>  extern int  xive_smp_prepare_cpu(unsigned int cpu);
>> @@ -147,6 +148,7 @@ extern int xive_native_get_vp_info(u32 vp_id, u32 *out_cam_id, u32 *out_chip_id)
>>  
>>  static inline bool xive_enabled(void) { return false; }
>>  
>> +static inline bool xive_spapr_init(void) { return false; }
>>  static inline bool xive_native_init(void) { return false; }
>>  static inline void xive_smp_probe(void) { }
>>  extern inline int  xive_smp_prepare_cpu(unsigned int cpu) { return -EINVAL; }
>> diff --git a/arch/powerpc/platforms/pseries/Kconfig b/arch/powerpc/platforms/pseries/Kconfig
>> index 3a6dfd14f64b..71dd69d9ec64 100644
>> --- a/arch/powerpc/platforms/pseries/Kconfig
>> +++ b/arch/powerpc/platforms/pseries/Kconfig
>> @@ -7,6 +7,7 @@ config PPC_PSERIES
>>  	select PCI
>>  	select PCI_MSI
>>  	select PPC_XICS
>> +	select PPC_XIVE_SPAPR
>>  	select PPC_ICP_NATIVE
>>  	select PPC_ICP_HV
>>  	select PPC_ICS_RTAS
>> diff --git a/arch/powerpc/platforms/pseries/hotplug-cpu.c b/arch/powerpc/platforms/pseries/hotplug-cpu.c
>> index 6afd1efd3633..de0a5c1d7b29 100644
>> --- a/arch/powerpc/platforms/pseries/hotplug-cpu.c
>> +++ b/arch/powerpc/platforms/pseries/hotplug-cpu.c
>> @@ -34,6 +34,7 @@
>>  #include <asm/machdep.h>
>>  #include <asm/vdso_datapage.h>
>>  #include <asm/xics.h>
>> +#include <asm/xive.h>
>>  #include <asm/plpar_wrappers.h>
>>  
>>  #include "pseries.h"
>> @@ -109,7 +110,9 @@ static void pseries_mach_cpu_die(void)
>>  
>>  	local_irq_disable();
>>  	idle_task_exit();
>> -	xics_teardown_cpu();
>> +	/* Nothing TODO for xive ? */;
> 
> You have a XIVE teardown_cpu() method below - should you be calling
> that here, even though it's a no-op for now?

yes. I agree.

>> +	if (!xive_enabled())
>> +		xics_teardown_cpu();
>>  
>>  	if (get_preferred_offline_state(cpu) == CPU_STATE_INACTIVE) {
>>  		set_cpu_current_state(cpu, CPU_STATE_INACTIVE);
>> @@ -174,7 +177,10 @@ static int pseries_cpu_disable(void)
>>  		boot_cpuid = cpumask_any(cpu_online_mask);
>>  
>>  	/* FIXME: abstract this to not be platform specific later on */
>> -	xics_migrate_irqs_away();
>> +	if (xive_enabled())
>> +		xive_smp_disable_cpu();
>> +	else
>> +		xics_migrate_irqs_away();
>>  	return 0;
>>  }
>>  
>> diff --git a/arch/powerpc/platforms/pseries/kexec.c b/arch/powerpc/platforms/pseries/kexec.c
>> index 6681ac97fb18..eeb13429d685 100644
>> --- a/arch/powerpc/platforms/pseries/kexec.c
>> +++ b/arch/powerpc/platforms/pseries/kexec.c
>> @@ -15,6 +15,7 @@
>>  #include <asm/firmware.h>
>>  #include <asm/kexec.h>
>>  #include <asm/xics.h>
>> +#include <asm/xive.h>
>>  #include <asm/smp.h>
>>  #include <asm/plpar_wrappers.h>
>>  
>> @@ -51,5 +52,8 @@ void pseries_kexec_cpu_down(int crash_shutdown, int secondary)
>>  		}
>>  	}
>>  
>> -	xics_kexec_teardown_cpu(secondary);
>> +	if (xive_enabled())
>> +		xive_kexec_teardown_cpu(secondary);
>> +	else
>> +		xics_kexec_teardown_cpu(secondary);
>>  }
>> diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c
>> index b5d86426e97b..a8531e012658 100644
>> --- a/arch/powerpc/platforms/pseries/setup.c
>> +++ b/arch/powerpc/platforms/pseries/setup.c
>> @@ -57,6 +57,7 @@
>>  #include <asm/nvram.h>
>>  #include <asm/pmc.h>
>>  #include <asm/xics.h>
>> +#include <asm/xive.h>
>>  #include <asm/ppc-pci.h>
>>  #include <asm/i8259.h>
>>  #include <asm/udbg.h>
>> @@ -176,8 +177,11 @@ static void __init pseries_setup_i8259_cascade(void)
>>  
>>  static void __init pseries_init_irq(void)
>>  {
>> -	xics_init();
>> -	pseries_setup_i8259_cascade();
>> +	/* Try using a XIVE if available, otherwise use a XICS */
>> +	if (!xive_spapr_init()) {
>> +		xics_init();
>> +		pseries_setup_i8259_cascade();
>> +	}
>>  }
>>  
>>  static void pseries_lpar_enable_pmcs(void)
>> diff --git a/arch/powerpc/platforms/pseries/smp.c b/arch/powerpc/platforms/pseries/smp.c
>> index 24785f63fb40..244bb9974963 100644
>> --- a/arch/powerpc/platforms/pseries/smp.c
>> +++ b/arch/powerpc/platforms/pseries/smp.c
>> @@ -41,6 +41,7 @@
>>  #include <asm/vdso_datapage.h>
>>  #include <asm/cputhreads.h>
>>  #include <asm/xics.h>
>> +#include <asm/xive.h>
>>  #include <asm/dbell.h>
>>  #include <asm/plpar_wrappers.h>
>>  #include <asm/code-patching.h>
>> @@ -136,7 +137,9 @@ static inline int smp_startup_cpu(unsigned int lcpu)
>>  
>>  static void smp_setup_cpu(int cpu)
>>  {
>> -	if (cpu != boot_cpuid)
>> +	if (xive_enabled())
>> +		xive_smp_setup_cpu();
>> +	else if (cpu != boot_cpuid)
>>  		xics_setup_cpu();
>>  
>>  	if (firmware_has_feature(FW_FEATURE_SPLPAR))
>> @@ -181,13 +184,22 @@ static int smp_pSeries_kick_cpu(int nr)
>>  	return 0;
>>  }
>>  
>> +static int pseries_smp_prepare_cpu(int cpu)
>> +{
>> +	if (xive_enabled())
>> +		return xive_smp_prepare_cpu(cpu);
>> +	return 0;
>> +}
>> +
>> +/* Cause IPI as setup by the interrupt controller (xics or xive) */
>> +static void (*ic_cause_ipi)(int cpu);
>> +
>>  static void smp_pseries_cause_ipi(int cpu)
>>  {
>> -	/* POWER9 should not use this handler */
>>  	if (doorbell_try_core_ipi(cpu))
>>  		return;
>>  
>> -	icp_ops->cause_ipi(cpu);
>> +	ic_cause_ipi(cpu);
> 
> Wouldn't it make more sense to change smp_ops->cause_ipi, rather than
> having a double indirection through smp_ops, then the ic_cause_ipi
> global?

we need to retain the original setting of smp_ops->cause_ipi 
somewhere as it can be set from xive_smp_probe() to : 

	icp_ops->cause_ipi 

or from xics_smp_probe() to :

	xive_cause_ipi()

I am not sure we can do any better ? 

>>  }
>>  
>>  static int pseries_cause_nmi_ipi(int cpu)
>> @@ -213,12 +225,17 @@ static int pseries_cause_nmi_ipi(int cpu)
>>  
>>  static __init void pSeries_smp_probe(void)
>>  {
>> -	xics_smp_probe();
>> +	if (xive_enabled())
>> +		xive_smp_probe();
>> +	else
>> +		xics_smp_probe();
>> +
>> +	if (cpu_has_feature(CPU_FTR_DBELL)) {
>> +		ic_cause_ipi = smp_ops->cause_ipi;
>> +		WARN_ON(!ic_cause_ipi);
>>  
>> -	if (cpu_has_feature(CPU_FTR_DBELL))
>>  		smp_ops->cause_ipi = smp_pseries_cause_ipi;
>> -	else
>> -		smp_ops->cause_ipi = icp_ops->cause_ipi;
>> +	}
>>  }
>>  
>>  static struct smp_ops_t pseries_smp_ops = {
>> @@ -226,6 +243,7 @@ static struct smp_ops_t pseries_smp_ops = {
>>  	.cause_ipi	= NULL,	/* Filled at runtime by pSeries_smp_probe() */
>>  	.cause_nmi_ipi	= pseries_cause_nmi_ipi,
>>  	.probe		= pSeries_smp_probe,
>> +	.prepare_cpu	= pseries_smp_prepare_cpu,
>>  	.kick_cpu	= smp_pSeries_kick_cpu,
>>  	.setup_cpu	= smp_setup_cpu,
>>  	.cpu_bootable	= smp_generic_cpu_bootable,
>> diff --git a/arch/powerpc/sysdev/xive/Kconfig b/arch/powerpc/sysdev/xive/Kconfig
>> index 12ccd7373d2f..3e3e25b5e30d 100644
>> --- a/arch/powerpc/sysdev/xive/Kconfig
>> +++ b/arch/powerpc/sysdev/xive/Kconfig
>> @@ -9,3 +9,8 @@ config PPC_XIVE_NATIVE
>>  	default n
>>  	select PPC_XIVE
>>  	depends on PPC_POWERNV
>> +
>> +config PPC_XIVE_SPAPR
>> +	bool
>> +	default n
>> +	select PPC_XIVE
>> diff --git a/arch/powerpc/sysdev/xive/Makefile b/arch/powerpc/sysdev/xive/Makefile
>> index 3fab303fc169..536d6e5706e3 100644
>> --- a/arch/powerpc/sysdev/xive/Makefile
>> +++ b/arch/powerpc/sysdev/xive/Makefile
>> @@ -2,3 +2,4 @@ subdir-ccflags-$(CONFIG_PPC_WERROR) := -Werror
>>  
>>  obj-y				+= common.o
>>  obj-$(CONFIG_PPC_XIVE_NATIVE)	+= native.o
>> +obj-$(CONFIG_PPC_XIVE_SPAPR)	+= spapr.o
>> diff --git a/arch/powerpc/sysdev/xive/spapr.c b/arch/powerpc/sysdev/xive/spapr.c
>> new file mode 100644
>> index 000000000000..a3668815d5c1
>> --- /dev/null
>> +++ b/arch/powerpc/sysdev/xive/spapr.c
>> @@ -0,0 +1,554 @@
>> +/*
>> + * Copyright 2016,2017 IBM Corporation.
>> + *
>> + * This program is free software; you can redistribute it and/or
>> + * modify it under the terms of the GNU General Public License
>> + * as published by the Free Software Foundation; either version
>> + * 2 of the License, or (at your option) any later version.
>> + */
>> +
>> +#define pr_fmt(fmt) "xive: " fmt
>> +
>> +#include <linux/types.h>
>> +#include <linux/irq.h>
>> +#include <linux/smp.h>
>> +#include <linux/interrupt.h>
>> +#include <linux/init.h>
>> +#include <linux/of.h>
>> +#include <linux/slab.h>
>> +#include <linux/spinlock.h>
>> +#include <linux/cpumask.h>
>> +#include <linux/mm.h>
>> +
>> +#include <asm/prom.h>
>> +#include <asm/io.h>
>> +#include <asm/smp.h>
>> +#include <asm/irq.h>
>> +#include <asm/errno.h>
>> +#include <asm/xive.h>
>> +#include <asm/xive-regs.h>
>> +#include <asm/hvcall.h>
>> +
>> +#include "xive-internal.h"
>> +
>> +static u32 xive_queue_shift;
>> +
>> +struct xive_irq_bitmap {
>> +	unsigned long		*bitmap;
>> +	unsigned int		base;
>> +	unsigned int		count;
>> +	spinlock_t		lock;
>> +	struct list_head	list;
>> +};
>> +
>> +static LIST_HEAD(xive_irq_bitmaps);
>> +
>> +static int xive_irq_bitmap_add(int base, int count)
>> +{
>> +	struct xive_irq_bitmap *xibm;
>> +
>> +	xibm = kzalloc(sizeof(*xibm), GFP_ATOMIC);
>> +	if (!xibm)
>> +		return -ENOMEM;
>> +
>> +	spin_lock_init(&xibm->lock);
>> +	xibm->base = base;
>> +	xibm->count = count;
>> +	xibm->bitmap = kzalloc(xibm->count, GFP_KERNEL);
>> +	list_add(&xibm->list, &xive_irq_bitmaps);
>> +
>> +	pr_info("Using IRQ range [%x-%x]", xibm->base,
>> +		xibm->base + xibm->count - 1);
>> +	return 0;
>> +}
>> +
>> +static int __xive_irq_bitmap_alloc(struct xive_irq_bitmap *xibm)
>> +{
>> +	int irq;
>> +
>> +	irq = find_first_zero_bit(xibm->bitmap, xibm->count);
>> +	if (irq != xibm->count) {
>> +		set_bit(irq, xibm->bitmap);
>> +		irq += xibm->base;
>> +	} else {
>> +		irq = -ENOMEM;
>> +	}
>> +
>> +	return irq;
>> +}
>> +
>> +static int xive_irq_bitmap_alloc(void)
>> +{
>> +	struct xive_irq_bitmap *xibm;
>> +	unsigned long flags;
>> +	int irq = -ENOENT;
>> +
>> +	list_for_each_entry(xibm, &xive_irq_bitmaps, list) {
>> +		spin_lock_irqsave(&xibm->lock, flags);
>> +		irq = __xive_irq_bitmap_alloc(xibm);
>> +		spin_unlock_irqrestore(&xibm->lock, flags);
>> +		if (irq >= 0)
>> +			break;
>> +	}
>> +	return irq;
>> +}
>> +
>> +static void xive_irq_bitmap_free(int irq)
>> +{
>> +	unsigned long flags;
>> +	struct xive_irq_bitmap *xibm;
>> +
>> +	list_for_each_entry(xibm, &xive_irq_bitmaps, list) {
>> +		if ((irq >= xibm->base) && (irq < xibm->base + xibm->count)) {
>> +			spin_lock_irqsave(&xibm->lock, flags);
>> +			clear_bit(irq - xibm->base, xibm->bitmap);
>> +			spin_unlock_irqrestore(&xibm->lock, flags);
>> +			break;
>> +		}
>> +	}
>> +}
>> +
>> +static long plpar_int_get_source_info(unsigned long flags,
>> +				      unsigned long lisn,
>> +				      unsigned long *src_flags,
>> +				      unsigned long *eoi_page,
>> +				      unsigned long *trig_page,
>> +				      unsigned long *esb_shift)
>> +{
>> +	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
>> +	long rc;
>> +
>> +	rc = plpar_hcall(H_INT_GET_SOURCE_INFO, retbuf, flags, lisn);
>> +	if (rc) {
>> +		pr_err("H_INT_GET_SOURCE_INFO lisn=%ld failed %ld\n", lisn, rc);
>> +		return rc;
>> +	}
>> +
>> +	*src_flags = retbuf[0];
>> +	*eoi_page  = retbuf[1];
>> +	*trig_page = retbuf[2];
>> +	*esb_shift = retbuf[3];
>> +
>> +	pr_devel("H_INT_GET_SOURCE_INFO flags=%lx eoi=%lx trig=%lx shift=%lx\n",
>> +		retbuf[0], retbuf[1], retbuf[2], retbuf[3]);
>> +
>> +	return 0;
>> +}
>> +
>> +#define XIVE_SRC_SET_EISN (1ull << (63 - 62))
>> +#define XIVE_SRC_MASK     (1ull << (63 - 63)) /* unused */
>> +
>> +static long plpar_int_set_source_config(unsigned long flags,
>> +					unsigned long lisn,
>> +					unsigned long target,
>> +					unsigned long prio,
>> +					unsigned long sw_irq)
>> +{
>> +	long rc;
>> +
>> +
>> +	pr_devel("H_INT_SET_SOURCE_CONFIG flags=%lx lisn=%lx target=%lx prio=%lx sw_irq=%lx\n",
>> +		flags, lisn, target, prio, sw_irq);
>> +
>> +
>> +	rc = plpar_hcall_norets(H_INT_SET_SOURCE_CONFIG, flags, lisn,
>> +				target, prio, sw_irq);
>> +	if (rc) {
>> +		pr_err("H_INT_SET_SOURCE_CONFIG lisn=%ld failed %ld\n",
>> +		       lisn, rc);
>> +		return rc;
>> +	}
>> +
>> +	return 0;
>> +}
>> +
>> +static long plpar_int_get_queue_info(unsigned long flags,
>> +				     unsigned long target,
>> +				     unsigned long priority,
>> +				     unsigned long *esn_page,
>> +				     unsigned long *esn_size)
>> +{
>> +	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
>> +	long rc;
>> +
>> +	rc = plpar_hcall(H_INT_GET_QUEUE_INFO, retbuf, flags, target, priority);
>> +	if (rc) {
>> +		pr_err("H_INT_GET_QUEUE_INFO cpu=%ld prio=%ld failed %ld\n",
>> +		       target, priority, rc);
>> +		return rc;
>> +	}
>> +
>> +	*esn_page = retbuf[0];
>> +	*esn_size = retbuf[1];
>> +
>> +	pr_devel("H_INT_GET_QUEUE_INFO page=%lx size=%lx\n",
>> +		retbuf[0], retbuf[1]);
>> +
>> +	return 0;
>> +}
>> +
>> +#define XIVE_EQ_ALWAYS_NOTIFY (1ull << (63 - 63))
>> +
>> +static long plpar_int_set_queue_config(unsigned long flags,
>> +				       unsigned long target,
>> +				       unsigned long priority,
>> +				       unsigned long qpage,
>> +				       unsigned long qsize)
>> +{
>> +	long rc;
>> +
>> +	pr_devel("H_INT_SET_QUEUE_CONFIG flags=%lx target=%lx priority=%lx qpage=%lx qsize=%lx\n",
>> +		flags,  target, priority, qpage, qsize);
>> +
>> +	rc = plpar_hcall_norets(H_INT_SET_QUEUE_CONFIG, flags, target,
>> +				priority, qpage, qsize);
>> +	if (rc) {
>> +		pr_err("H_INT_SET_QUEUE_CONFIG cpu=%ld prio=%ld qpage=%lx returned %ld\n",
>> +		       target, priority, qpage, rc);
>> +		return  rc;
>> +	}
>> +
>> +	return 0;
>> +}
>> +
>> +static long plpar_int_sync(unsigned long flags)
>> +{
>> +	long rc;
>> +
>> +	rc = plpar_hcall_norets(H_INT_SYNC, flags);
>> +	if (rc) {
>> +		pr_err("H_INT_SYNC returned %ld\n", rc);
>> +		return  rc;
>> +	}
>> +
>> +	return 0;
>> +}
>> +
>> +#define XIVE_SRC_H_INT_ESB     (1ull << (63 - 60)) /* TODO */
>> +#define XIVE_SRC_LSI           (1ull << (63 - 61))
>> +#define XIVE_SRC_TRIGGER       (1ull << (63 - 62))
>> +#define XIVE_SRC_STORE_EOI     (1ull << (63 - 63))
>> +
>> +static int xive_spapr_populate_irq_data(u32 hw_irq, struct xive_irq_data *data)
>> +{
>> +	long rc;
>> +	unsigned long flags;
>> +	unsigned long eoi_page;
>> +	unsigned long trig_page;
>> +	unsigned long esb_shift;
>> +
>> +	memset(data, 0, sizeof(*data));
>> +
>> +	rc = plpar_int_get_source_info(0, hw_irq, &flags, &eoi_page, &trig_page,
>> +				       &esb_shift);
>> +	if (rc)
>> +		return  -EINVAL;
>> +
>> +	if (flags & XIVE_SRC_STORE_EOI)
>> +		data->flags  |= XIVE_IRQ_FLAG_STORE_EOI;
>> +	if (flags & XIVE_SRC_LSI)
>> +		data->flags  |= XIVE_IRQ_FLAG_LSI;
>> +	data->eoi_page  = eoi_page;
>> +	data->esb_shift = esb_shift;
>> +	data->trig_page = trig_page;
>> +
>> +	/*
>> +	 * No chip-id for the sPAPR backend. This has an impact how we
>> +	 * pick a target. See xive_pick_irq_target().
>> +	 */
>> +	data->src_chip = XIVE_INVALID_CHIP_ID;
>> +
>> +	data->eoi_mmio = ioremap(data->eoi_page, 1u << data->esb_shift);
>> +	if (!data->eoi_mmio) {
>> +		pr_err("Failed to map EOI page for irq 0x%x\n", hw_irq);
>> +		return -ENOMEM;
>> +	}
>> +
>> +	/* Full function page supports trigger */
>> +	if (flags & XIVE_SRC_TRIGGER) {
>> +		data->trig_mmio = data->eoi_mmio;
>> +		return 0;
>> +	}
>> +
>> +	data->trig_mmio = ioremap(data->trig_page, 1u << data->esb_shift);
>> +	if (!data->trig_mmio) {
>> +		pr_err("Failed to map trigger page for irq 0x%x\n", hw_irq);
>> +		return -ENOMEM;
>> +	}
>> +	return 0;
>> +}
>> +
>> +static int xive_spapr_configure_irq(u32 hw_irq, u32 target, u8 prio, u32 sw_irq)
>> +{
>> +	long rc;
>> +
>> +	rc = plpar_int_set_source_config(XIVE_SRC_SET_EISN, hw_irq, target,
>> +					 prio, sw_irq);
>> +
>> +	return rc == 0 ? 0 : -ENXIO;
>> +}
>> +
>> +/* This can be called multiple time to change a queue configuration */
>> +static int xive_spapr_configure_queue(u32 target, struct xive_q *q, u8 prio,
>> +				   __be32 *qpage, u32 order)
>> +{
>> +	s64 rc = 0;
>> +	unsigned long esn_page;
>> +	unsigned long esn_size;
>> +	u64 flags, qpage_phys;
>> +
>> +	/* If there's an actual queue page, clean it */
>> +	if (order) {
>> +		if (WARN_ON(!qpage))
>> +			return -EINVAL;
>> +		qpage_phys = __pa(qpage);
>> +	} else {
>> +		qpage_phys = 0;
>> +	}
>> +
>> +	/* Initialize the rest of the fields */
>> +	q->msk = order ? ((1u << (order - 2)) - 1) : 0;
>> +	q->idx = 0;
>> +	q->toggle = 0;
>> +
>> +	rc = plpar_int_get_queue_info(0, target, prio, &esn_page, &esn_size);
>> +	if (rc) {
>> +		pr_err("Error %lld getting queue info prio %d\n", rc, prio);
>> +		rc = -EIO;
>> +		goto fail;
>> +	}
>> +
>> +	/* TODO: add support for the notification page */
>> +	q->eoi_phys = esn_page;
>> +
>> +	/* Default is to always notify */
>> +	flags = XIVE_EQ_ALWAYS_NOTIFY;
>> +
>> +	/* Configure and enable the queue in HW */
>> +	rc = plpar_int_set_queue_config(flags, target, prio, qpage_phys, order);
>> +	if (rc) {
>> +		pr_err("Error %lld setting queue for prio %d\n", rc, prio);
>> +		rc = -EIO;
>> +	} else {
>> +		q->qpage = qpage;
>> +	}
>> +fail:
>> +	return rc;
>> +}
>> +
>> +static int xive_spapr_setup_queue(unsigned int cpu, struct xive_cpu *xc,
>> +				  u8 prio)
>> +{
>> +	struct xive_q *q = &xc->queue[prio];
>> +	unsigned int alloc_order;
>> +	struct page *pages;
>> +	__be32 *qpage;
>> +
>> +	alloc_order = (xive_queue_shift > PAGE_SHIFT) ?
>> +		(xive_queue_shift - PAGE_SHIFT) : 0;
>> +	pages = alloc_pages_node(cpu_to_node(cpu), GFP_KERNEL, alloc_order);
>> +	if (!pages)
>> +		return -ENOMEM;
>> +	qpage = (__be32 *)page_address(pages);
>> +	memset(qpage, 0, 1 << xive_queue_shift);
>> +
>> +	return xive_spapr_configure_queue(cpu, q, prio, qpage,
>> +					  xive_queue_shift);
>> +}
>> +
>> +static void xive_spapr_cleanup_queue(unsigned int cpu, struct xive_cpu *xc,
>> +				  u8 prio)
>> +{
>> +	struct xive_q *q = &xc->queue[prio];
>> +	unsigned int alloc_order;
>> +	long rc;
>> +
>> +	rc = plpar_int_set_queue_config(0, cpu, prio, 0, 0);
>> +	if (rc)
>> +		pr_err("Error %ld setting queue for prio %d\n", rc, prio);
>> +
>> +	alloc_order = (xive_queue_shift > PAGE_SHIFT) ?
>> +		(xive_queue_shift - PAGE_SHIFT) : 0;
> 
> Maybe a helper inline for this calculation, since it's used in both
> setup_queue() and cleanup_queue?

yes. 

The xive_spapr_setup_queue() and xive_native_setup_queue() 
routines have a lot in common. So do xive_spapr_setup_queue() 
and xive_native_setup_queue(). I will try to move some of
the common code in the xive frontend.


>> +	free_pages((unsigned long)q->qpage, alloc_order);
>> +	q->qpage = NULL;
>> +}
>> +
>> +static bool xive_spapr_match(struct device_node *node)
>> +{
>> +	return 1;
>> +}
>> +
>> +#ifdef CONFIG_SMP
>> +static int xive_spapr_get_ipi(unsigned int cpu, struct xive_cpu *xc)
>> +{
>> +	int irq = xive_irq_bitmap_alloc();
>> +
>> +	if (irq < 0) {
>> +		pr_err("Failed to allocate IPI on CPU %d\n", cpu);
>> +		return -ENXIO;
>> +	}
>> +
>> +	xc->hw_ipi = irq;
>> +	return 0;
>> +}
>> +
>> +static void xive_spapr_put_ipi(unsigned int cpu, struct xive_cpu *xc)
>> +{
>> +	xive_irq_bitmap_free(xc->hw_ipi);
>> +}
>> +#endif /* CONFIG_SMP */
>> +
>> +static void xive_spapr_shutdown(void)
>> +{
>> +	long rc;
>> +
>> +	rc = plpar_hcall_norets(H_INT_RESET, 0);
>> +	if (rc)
>> +		pr_err("H_INT_RESET failed %ld\n", rc);
>> +}
>> +
>> +static void xive_spapr_update_pending(struct xive_cpu *xc)
>> +{
>> +	u8 nsr, cppr;
>> +	u16 ack;
>> +
>> +	/* Perform the acknowledge hypervisor to register cycle */
>> +	ack = be16_to_cpu(__raw_readw(xive_tima + TM_SPC_ACK_OS_REG));
> 
> Why do you need the raw_readw() + be16_to_cpu + mb, rather than one of
> the higher level IO helpers?

This is one of the many ways to do MMIOs on the TIMA. This memory 
region defines a set of offsets and sizes for which loads and 
stores have different effects. 

See the arch/powerpc/include/asm/xive-regs.h file and 
arch/powerpc/kvm/book3s_xive_template.c for some more usage.

maybe we could add some helpers. 
     
> 
>> +	/* Synchronize subsequent queue accesses */
>> +	mb();
>> +
>> +	/*
>> +	 * Grab the CPPR and the "NSR" field which indicates the source
>> +	 * of the hypervisor interrupt (if any)
>> +	 */
>> +	cppr = ack & 0xff;
>> +	nsr = ack >> 8;
>> +
>> +	if (nsr & TM_QW1_NSR_EO) {
>> +		if (cppr == 0xff)
>> +			return;
>> +		/* Mark the priority pending */
>> +		xc->pending_prio |= 1 << cppr;
>> +
>> +		/*
>> +		 * A new interrupt should never have a CPPR less favored
>> +		 * than our current one.
>> +		 */
>> +		if (cppr >= xc->cppr)
>> +			pr_err("CPU %d odd ack CPPR, got %d at %d\n",
>> +			       smp_processor_id(), cppr, xc->cppr);
>> +
>> +		/* Update our idea of what the CPPR is */
>> +		xc->cppr = cppr;
>> +	}
>> +}
>> +
>> +static void xive_spapr_eoi(u32 hw_irq)
>> +{
>> +	/* Not used */;
>> +}
> 
> Do you really want empty stub functions, or would it be better to have
> the method caller check for NULL in the ops structure?

Ben had some comment saying that we could use this eoi op for 
interrupts needing H_INT_ESB. I am not sure this is the case 
anymore with the framework I propose in the following patches.
That said, I will have to implement in QEMU interrupt sources 
needing this flag in order to test. phyp won't implement it.

Nevertheless I can add a check for NULL.

Thanks,c

C.

>> +
>> +static void xive_spapr_setup_cpu(unsigned int cpu, struct xive_cpu *xc)
>> +{
>> +	/* Only some debug on the TIMA settings */
>> +	pr_debug("(HW value: %08x %08x %08x)\n",
>> +		 in_be32(xive_tima + TM_QW1_OS + TM_WORD0),
>> +		 in_be32(xive_tima + TM_QW1_OS + TM_WORD1),
>> +		 in_be32(xive_tima + TM_QW1_OS + TM_WORD2));
>> +}
>> +
>> +static void xive_spapr_teardown_cpu(unsigned int cpu, struct xive_cpu *xc)
>> +{
>> +	/* Nothing to do */;
>> +}
>> +
>> +static void xive_spapr_sync_source(u32 hw_irq)
>> +{
>> +	/* Specs are unclear on what this is doing */
>> +	plpar_int_sync(0);
>> +}
>> +
>> +static const struct xive_ops xive_spapr_ops = {
>> +	.populate_irq_data	= xive_spapr_populate_irq_data,
>> +	.configure_irq		= xive_spapr_configure_irq,
>> +	.setup_queue		= xive_spapr_setup_queue,
>> +	.cleanup_queue		= xive_spapr_cleanup_queue,
>> +	.match			= xive_spapr_match,
>> +	.shutdown		= xive_spapr_shutdown,
>> +	.update_pending		= xive_spapr_update_pending,
>> +	.eoi			= xive_spapr_eoi,
>> +	.setup_cpu		= xive_spapr_setup_cpu,
>> +	.teardown_cpu		= xive_spapr_teardown_cpu,
>> +	.sync_source		= xive_spapr_sync_source,
>> +#ifdef CONFIG_SMP
>> +	.get_ipi		= xive_spapr_get_ipi,
>> +	.put_ipi		= xive_spapr_put_ipi,
>> +#endif /* CONFIG_SMP */
>> +	.name			= "spapr",
>> +};
>> +
>> +bool xive_spapr_init(void)
>> +{
>> +	struct device_node *np;
>> +	struct resource r;
>> +	void __iomem *tima;
>> +	struct property *prop;
>> +	u8 max_prio = 7;
>> +	u32 val;
>> +	u32 len;
>> +	const __be32 *reg;
>> +	int i;
>> +
>> +	if (xive_cmdline_disabled)
>> +		return false;
>> +
>> +	pr_devel("%s()\n", __func__);
>> +	np = of_find_compatible_node(NULL, NULL, "ibm,power-ivpe");
>> +	if (!np) {
>> +		pr_devel("not found !\n");
>> +		return false;
>> +	}
>> +	pr_devel("Found %s\n", np->full_name);
>> +
>> +	/* Resource 1 is the OS ring TIMA */
>> +	if (of_address_to_resource(np, 1, &r)) {
>> +		pr_err("Failed to get thread mgmnt area resource\n");
>> +		return false;
>> +	}
>> +	tima = ioremap(r.start, resource_size(&r));
>> +	if (!tima) {
>> +		pr_err("Failed to map thread mgmnt area\n");
>> +		return false;
>> +	}
>> +
>> +	/* Feed the IRQ number allocator with the ranges given in the DT */
>> +	reg = of_get_property(np, "ibm,xive-lisn-ranges", &len);
>> +	if (!reg) {
>> +		pr_err("Failed to read 'ibm,xive-lisn-ranges' property\n");
>> +		return false;
>> +	}
>> +
>> +	if (len % (2 * sizeof(u32)) != 0) {
>> +		pr_err("invalid 'ibm,xive-lisn-ranges' property\n");
>> +		return false;
>> +	}
>> +
>> +	for (i = 0; i < len / (2 * sizeof(u32)); i++, reg += 2)
>> +		xive_irq_bitmap_add(be32_to_cpu(reg[0]),
>> +				    be32_to_cpu(reg[1]));
>> +
>> +	/* Iterate the EQ sizes and pick one */
>> +	of_property_for_each_u32(np, "ibm,xive-eq-sizes", prop, reg, val) {
>> +		xive_queue_shift = val;
>> +		if (val == PAGE_SHIFT)
>> +			break;
>> +	}
>> +
>> +	/* Initialize XIVE core with our backend */
>> +	if (!xive_core_init(&xive_spapr_ops, tima, TM_QW1_OS, max_prio))
>> +		return false;
>> +
>> +	pr_info("Using %dkB queues\n", 1 << (xive_queue_shift - 10));
>> +	return true;
>> +}
>
David Gibson Aug. 10, 2017, 4:28 a.m. UTC | #3 
On Wed, Aug 09, 2017 at 10:48:48AM +0200, Cédric Le Goater wrote:
> On 08/09/2017 05:53 AM, David Gibson wrote:
> > On Tue, Aug 08, 2017 at 10:56:12AM +0200, Cédric Le Goater wrote:
> >> This is the framework for using XIVE in a PowerVM guest. The support
> >> is very similar to the native one in a much simpler form.
> >>
> >> Instead of OPAL calls, a set of Hypervisors call are used to configure
> >> the interrupt sources and the event/notification queues of the guest:
> >>
> >>  - H_INT_GET_SOURCE_INFO
> >>
> >>    used to obtain the address of the MMIO page of the Event State
> >>    Buffer (PQ bits) entry associated with the source.
> >>
> >>  - H_INT_SET_SOURCE_CONFIG
> >>
> >>    assigns a source to a "target".
> >>
> >>  - H_INT_GET_SOURCE_CONFIG
> >>
> >>    determines to which "target" and "priority" is assigned to a source
> >>
> >>  - H_INT_GET_QUEUE_INFO
> >>
> >>    returns the address of the notification management page associated
> >>    with the specified "target" and "priority".
> >>
> >>  - H_INT_SET_QUEUE_CONFIG
> >>
> >>    sets or resets the event queue for a given "target" and "priority".
> >>    It is also used to set the notification config associated with the
> >>    queue, only unconditional notification for the moment.  Reset is
> >>    performed with a queue size of 0 and queueing is disabled in that
> >>    case.
> >>
> >>  - H_INT_GET_QUEUE_CONFIG
> >>
> >>    returns the queue settings for a given "target" and "priority".
> >>
> >>  - H_INT_RESET
> >>
> >>    resets all of the partition's interrupt exploitation structures to
> >>    their initial state, losing all configuration set via the hcalls
> >>    H_INT_SET_SOURCE_CONFIG and H_INT_SET_QUEUE_CONFIG.
> >>
> >>  - H_INT_SYNC
> >>
> >>    issue a synchronisation on a source to make sure sure all
> >>    notifications have reached their queue.
> >>
> >> As for XICS, the XIVE interface for the guest is described in the
> >> device tree under the interrupt controller node. A couple of new
> >> properties are specific to XIVE :
> >>
> >>  - "reg"
> >>
> >>    contains the base address and size of the thread interrupt
> >>    managnement areas (TIMA) for the user level for the OS level. Only
> >>    the OS level is taken into account.
> >>
> >>  - "ibm,xive-eq-sizes"
> >>
> >>    the size of the event queues.
> >>
> >>  - "ibm,xive-lisn-ranges"
> >>
> >>    the interrupt numbers ranges assigned to the guest. These are
> >>    allocated using a simple bitmap.
> >>
> >> Tested with a QEMU XIVE model for pseries and with the Power
> >> hypervisor
> >>
> >> Signed-off-by: Cédric Le Goater <clg@kaod.org>
> > 
> > I don't know XIVE well enough to review in detail, but I've made some
> > comments based on general knowledge.
> 
> Thanks for taking a look.

np

[snip]
> >> +/* Cause IPI as setup by the interrupt controller (xics or xive) */
> >> +static void (*ic_cause_ipi)(int cpu);
> >> +
> >>  static void smp_pseries_cause_ipi(int cpu)
> >>  {
> >> -	/* POWER9 should not use this handler */
> >>  	if (doorbell_try_core_ipi(cpu))
> >>  		return;
> >>  
> >> -	icp_ops->cause_ipi(cpu);
> >> +	ic_cause_ipi(cpu);
> > 
> > Wouldn't it make more sense to change smp_ops->cause_ipi, rather than
> > having a double indirection through smp_ops, then the ic_cause_ipi
> > global?
> 
> we need to retain the original setting of smp_ops->cause_ipi 
> somewhere as it can be set from xive_smp_probe() to : 
> 
> 	icp_ops->cause_ipi 
> 
> or from xics_smp_probe() to :
> 
> 	xive_cause_ipi()
> 
> I am not sure we can do any better ? 

I don't see why not.  You basically have two bits of options xics vs
xive, and doorbell vs no doorbells.  At worst that gives you 4
different versions of ->cause_ipi, and you can work out the right one
in smp_probe().  If the number of combinations got too much larger you
might indeed need some more indirection.  But with 4 there's a little
code duplication, but small enough that I think it's preferable to an
extra global and an extra indirection.

Also, will POWER9 always have doorbells?  In which case you could
reduce it to 3 options.

[snip]
> >> +static void xive_spapr_update_pending(struct xive_cpu *xc)
> >> +{
> >> +	u8 nsr, cppr;
> >> +	u16 ack;
> >> +
> >> +	/* Perform the acknowledge hypervisor to register cycle */
> >> +	ack = be16_to_cpu(__raw_readw(xive_tima + TM_SPC_ACK_OS_REG));
> > 
> > Why do you need the raw_readw() + be16_to_cpu + mb, rather than one of
> > the higher level IO helpers?
> 
> This is one of the many ways to do MMIOs on the TIMA. This memory 
> region defines a set of offsets and sizes for which loads and 
> stores have different effects. 
> 
> See the arch/powerpc/include/asm/xive-regs.h file and 
> arch/powerpc/kvm/book3s_xive_template.c for some more usage.

Sure, much like any IO region.  My point is, why do you want this kind
of complex combo, rather than say an in_be16() or readw_be().
Benjamin Herrenschmidt Aug. 10, 2017, 4:46 a.m. UTC | #4 
On Thu, 2017-08-10 at 14:28 +1000, David Gibson wrote:
> 
> Also, will POWER9 always have doorbells?  In which case you could
> reduce it to 3 options.

The problem with doorbells on POWER9 guests is that they may have
to trap and be emulated by the hypervisor, since the guest threads
on P9 don't have to match the HW threads of the core.

Thus it's quite possible that using XIVE for IPIs is actually faster
than doorbells in that case.

Cheers,
Ben.
David Gibson Aug. 10, 2017, 5:54 a.m. UTC | #5 
On Thu, Aug 10, 2017 at 02:46:00PM +1000, Benjamin Herrenschmidt wrote:
> On Thu, 2017-08-10 at 14:28 +1000, David Gibson wrote:
> > 
> > Also, will POWER9 always have doorbells?  In which case you could
> > reduce it to 3 options.
> 
> The problem with doorbells on POWER9 guests is that they may have
> to trap and be emulated by the hypervisor, since the guest threads
> on P9 don't have to match the HW threads of the core.
> 
> Thus it's quite possible that using XIVE for IPIs is actually faster
> than doorbells in that case.

Ok, well unless there's a compelling reason to use doorbells you can
instead make it 3 cases with POWER9 never using doorbells.
Cédric Le Goater Aug. 10, 2017, 6:45 a.m. UTC | #6 
On 08/10/2017 06:46 AM, Benjamin Herrenschmidt wrote:
> On Thu, 2017-08-10 at 14:28 +1000, David Gibson wrote:
>>
>> Also, will POWER9 always have doorbells?  In which case you could
>> reduce it to 3 options.
> 
> The problem with doorbells on POWER9 guests is that they may have
> to trap and be emulated by the hypervisor, since the guest threads
> on P9 don't have to match the HW threads of the core.

Well, the pseries cause_ipi() handler does :

 	if (doorbell_try_core_ipi(cpu))
 		return;

to limit the doorbells to the same core. So we should be fine ? If not
I suppose we should check CPU_FTR_ARCH_300 and use IPIs only for XIVE.

> Thus it's quite possible that using XIVE for IPIs is actually faster
> than doorbells in that case.

How can we measure that ? ebizzy may be.

C.
Cédric Le Goater Aug. 10, 2017, 7:04 a.m. UTC | #7 
On 08/10/2017 07:54 AM, David Gibson wrote:
> On Thu, Aug 10, 2017 at 02:46:00PM +1000, Benjamin Herrenschmidt wrote:
>> On Thu, 2017-08-10 at 14:28 +1000, David Gibson wrote:
>>>
>>> Also, will POWER9 always have doorbells?  In which case you could
>>> reduce it to 3 options.
>>
>> The problem with doorbells on POWER9 guests is that they may have
>> to trap and be emulated by the hypervisor, since the guest threads
>> on P9 don't have to match the HW threads of the core.
>>
>> Thus it's quite possible that using XIVE for IPIs is actually faster
>> than doorbells in that case.
> 
> Ok, well unless there's a compelling reason to use doorbells you can
> instead make it 3 cases with POWER9 never using doorbells.

yes I suppose we could improve things for XIVE. For XICS, there are 
another 3 backends behind icp_ops->cause_ipi. I need to take a look.

Thanks,

C.
Cédric Le Goater Aug. 10, 2017, 7:19 a.m. UTC | #8 
>>>> +static void xive_spapr_update_pending(struct xive_cpu *xc)
>>>> +{
>>>> +	u8 nsr, cppr;
>>>> +	u16 ack;
>>>> +
>>>> +	/* Perform the acknowledge hypervisor to register cycle */
>>>> +	ack = be16_to_cpu(__raw_readw(xive_tima + TM_SPC_ACK_OS_REG));
>>>
>>> Why do you need the raw_readw() + be16_to_cpu + mb, rather than one of
>>> the higher level IO helpers?
>>
>> This is one of the many ways to do MMIOs on the TIMA. This memory 
>> region defines a set of offsets and sizes for which loads and 
>> stores have different effects. 
>>
>> See the arch/powerpc/include/asm/xive-regs.h file and 
>> arch/powerpc/kvm/book3s_xive_template.c for some more usage.
> 
> Sure, much like any IO region.  My point is, why do you want this kind
> of complex combo, rather than say an in_be16() or readw_be().
> 

The code is inherited from the native backend. 

I think this is because we know what we are doing and we skip 
the synchronization routines of the higher level IO helpers. 
That might not be necessary for sPAPR. Ben ? 

Thanks,

C.
Benjamin Herrenschmidt Aug. 10, 2017, 11:33 a.m. UTC | #9 
On Thu, 2017-08-10 at 08:45 +0200, Cédric Le Goater wrote:
> > The problem with doorbells on POWER9 guests is that they may have
> > to trap and be emulated by the hypervisor, since the guest threads
> > on P9 don't have to match the HW threads of the core.
> 
> Well, the pseries cause_ipi() handler does :
> 
>         if (doorbell_try_core_ipi(cpu))
>                 return;
> 
> to limit the doorbells to the same core. So we should be fine ?

No. It's theorically possible to create a guest that think it has 4
threads on P9 but those threads run on different cores of the host.

The doorbells are useful if KVM uses a "P8 style" whole-core dispatch
model or with PowerVM. We should probably invent some kind of DT
property to tell the guest I suppoes.

>  If not
> I suppose we should check CPU_FTR_ARCH_300 and use IPIs only for XIVE.
> 
> > Thus it's quite possible that using XIVE for IPIs is actually faster
> > than doorbells in that case.
> 
> How can we measure that ? ebizzy may be.

Or a simple socket ping pong with processes pinned to different
threads.

However the current KVM for P9 doesn't do threads yet afaik.

Cheers,
Ben.
Benjamin Herrenschmidt Aug. 10, 2017, 11:36 a.m. UTC | #10 
On Thu, 2017-08-10 at 09:19 +0200, Cédric Le Goater wrote:
> > > > > +  /* Perform the acknowledge hypervisor to register cycle */
> > > > > +  ack = be16_to_cpu(__raw_readw(xive_tima + TM_SPC_ACK_OS_REG));
> > > > 
> > > > Why do you need the raw_readw() + be16_to_cpu + mb, rather than one of
> > > > the higher level IO helpers?
> > > 
> > > This is one of the many ways to do MMIOs on the TIMA. This memory 
> > > region defines a set of offsets and sizes for which loads and 
> > > stores have different effects. 
> > > 
> > > See the arch/powerpc/include/asm/xive-regs.h file and 
> > > arch/powerpc/kvm/book3s_xive_template.c for some more usage.
> > 
> > Sure, much like any IO region.  My point is, why do you want this kind
> > of complex combo, rather than say an in_be16() or readw_be().
> > 
> 
> The code is inherited from the native backend. 
> 
> I think this is because we know what we are doing and we skip 
> the synchronization routines of the higher level IO helpers. 
> That might not be necessary for sPAPR. Ben ? 

It's a performance optimisation, we know we don't need the
sync,twi,isync crap of the higher level accessor here.

Cheers,
Ben.
David Gibson Aug. 11, 2017, 3:55 a.m. UTC | #11 
On Thu, Aug 10, 2017 at 09:36:04PM +1000, Benjamin Herrenschmidt wrote:
> On Thu, 2017-08-10 at 09:19 +0200, Cédric Le Goater wrote:
> > > > > > +  /* Perform the acknowledge hypervisor to register cycle */
> > > > > > +  ack = be16_to_cpu(__raw_readw(xive_tima + TM_SPC_ACK_OS_REG));
> > > > > 
> > > > > Why do you need the raw_readw() + be16_to_cpu + mb, rather than one of
> > > > > the higher level IO helpers?
> > > > 
> > > > This is one of the many ways to do MMIOs on the TIMA. This memory 
> > > > region defines a set of offsets and sizes for which loads and 
> > > > stores have different effects. 
> > > > 
> > > > See the arch/powerpc/include/asm/xive-regs.h file and 
> > > > arch/powerpc/kvm/book3s_xive_template.c for some more usage.
> > > 
> > > Sure, much like any IO region.  My point is, why do you want this kind
> > > of complex combo, rather than say an in_be16() or readw_be().
> > > 
> > 
> > The code is inherited from the native backend. 
> > 
> > I think this is because we know what we are doing and we skip 
> > the synchronization routines of the higher level IO helpers. 
> > That might not be necessary for sPAPR. Ben ? 
> 
> It's a performance optimisation, we know we don't need the
> sync,twi,isync crap of the higher level accessor here.

Ok.  A comment mentioning this would be good - unless you're very
familiar with the code and hardware it's not going to be obvious if
the nonstandard IO accessors are for legitimate performance reasons,
or just cargo-culting.
diff mbox

Patch

diff --git a/arch/powerpc/include/asm/hvcall.h b/arch/powerpc/include/asm/hvcall.h
index 57d38b504ff7..3d34dc0869f6 100644
--- a/arch/powerpc/include/asm/hvcall.h
+++ b/arch/powerpc/include/asm/hvcall.h
@@ -280,7 +280,18 @@ 
 #define H_RESIZE_HPT_COMMIT	0x370
 #define H_REGISTER_PROC_TBL	0x37C
 #define H_SIGNAL_SYS_RESET	0x380
-#define MAX_HCALL_OPCODE	H_SIGNAL_SYS_RESET
+#define H_INT_GET_SOURCE_INFO   0x3A8
+#define H_INT_SET_SOURCE_CONFIG 0x3AC
+#define H_INT_GET_SOURCE_CONFIG 0x3B0
+#define H_INT_GET_QUEUE_INFO    0x3B4
+#define H_INT_SET_QUEUE_CONFIG  0x3B8
+#define H_INT_GET_QUEUE_CONFIG  0x3BC
+#define H_INT_SET_OS_REPORTING_LINE 0x3C0
+#define H_INT_GET_OS_REPORTING_LINE 0x3C4
+#define H_INT_ESB               0x3C8
+#define H_INT_SYNC              0x3CC
+#define H_INT_RESET             0x3D0
+#define MAX_HCALL_OPCODE	H_INT_RESET
 
 /* H_VIOCTL functions */
 #define H_GET_VIOA_DUMP_SIZE	0x01
diff --git a/arch/powerpc/include/asm/xive.h b/arch/powerpc/include/asm/xive.h
index c23ff4389ca2..1deb10032d61 100644
--- a/arch/powerpc/include/asm/xive.h
+++ b/arch/powerpc/include/asm/xive.h
@@ -110,6 +110,7 @@  extern bool __xive_enabled;
 
 static inline bool xive_enabled(void) { return __xive_enabled; }
 
+extern bool xive_spapr_init(void);
 extern bool xive_native_init(void);
 extern void xive_smp_probe(void);
 extern int  xive_smp_prepare_cpu(unsigned int cpu);
@@ -147,6 +148,7 @@  extern int xive_native_get_vp_info(u32 vp_id, u32 *out_cam_id, u32 *out_chip_id)
 
 static inline bool xive_enabled(void) { return false; }
 
+static inline bool xive_spapr_init(void) { return false; }
 static inline bool xive_native_init(void) { return false; }
 static inline void xive_smp_probe(void) { }
 extern inline int  xive_smp_prepare_cpu(unsigned int cpu) { return -EINVAL; }
diff --git a/arch/powerpc/platforms/pseries/Kconfig b/arch/powerpc/platforms/pseries/Kconfig
index 3a6dfd14f64b..71dd69d9ec64 100644
--- a/arch/powerpc/platforms/pseries/Kconfig
+++ b/arch/powerpc/platforms/pseries/Kconfig
@@ -7,6 +7,7 @@  config PPC_PSERIES
 	select PCI
 	select PCI_MSI
 	select PPC_XICS
+	select PPC_XIVE_SPAPR
 	select PPC_ICP_NATIVE
 	select PPC_ICP_HV
 	select PPC_ICS_RTAS
diff --git a/arch/powerpc/platforms/pseries/hotplug-cpu.c b/arch/powerpc/platforms/pseries/hotplug-cpu.c
index 6afd1efd3633..de0a5c1d7b29 100644
--- a/arch/powerpc/platforms/pseries/hotplug-cpu.c
+++ b/arch/powerpc/platforms/pseries/hotplug-cpu.c
@@ -34,6 +34,7 @@ 
 #include <asm/machdep.h>
 #include <asm/vdso_datapage.h>
 #include <asm/xics.h>
+#include <asm/xive.h>
 #include <asm/plpar_wrappers.h>
 
 #include "pseries.h"
@@ -109,7 +110,9 @@  static void pseries_mach_cpu_die(void)
 
 	local_irq_disable();
 	idle_task_exit();
-	xics_teardown_cpu();
+	/* Nothing TODO for xive ? */;
+	if (!xive_enabled())
+		xics_teardown_cpu();
 
 	if (get_preferred_offline_state(cpu) == CPU_STATE_INACTIVE) {
 		set_cpu_current_state(cpu, CPU_STATE_INACTIVE);
@@ -174,7 +177,10 @@  static int pseries_cpu_disable(void)
 		boot_cpuid = cpumask_any(cpu_online_mask);
 
 	/* FIXME: abstract this to not be platform specific later on */
-	xics_migrate_irqs_away();
+	if (xive_enabled())
+		xive_smp_disable_cpu();
+	else
+		xics_migrate_irqs_away();
 	return 0;
 }
 
diff --git a/arch/powerpc/platforms/pseries/kexec.c b/arch/powerpc/platforms/pseries/kexec.c
index 6681ac97fb18..eeb13429d685 100644
--- a/arch/powerpc/platforms/pseries/kexec.c
+++ b/arch/powerpc/platforms/pseries/kexec.c
@@ -15,6 +15,7 @@ 
 #include <asm/firmware.h>
 #include <asm/kexec.h>
 #include <asm/xics.h>
+#include <asm/xive.h>
 #include <asm/smp.h>
 #include <asm/plpar_wrappers.h>
 
@@ -51,5 +52,8 @@  void pseries_kexec_cpu_down(int crash_shutdown, int secondary)
 		}
 	}
 
-	xics_kexec_teardown_cpu(secondary);
+	if (xive_enabled())
+		xive_kexec_teardown_cpu(secondary);
+	else
+		xics_kexec_teardown_cpu(secondary);
 }
diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c
index b5d86426e97b..a8531e012658 100644
--- a/arch/powerpc/platforms/pseries/setup.c
+++ b/arch/powerpc/platforms/pseries/setup.c
@@ -57,6 +57,7 @@ 
 #include <asm/nvram.h>
 #include <asm/pmc.h>
 #include <asm/xics.h>
+#include <asm/xive.h>
 #include <asm/ppc-pci.h>
 #include <asm/i8259.h>
 #include <asm/udbg.h>
@@ -176,8 +177,11 @@  static void __init pseries_setup_i8259_cascade(void)
 
 static void __init pseries_init_irq(void)
 {
-	xics_init();
-	pseries_setup_i8259_cascade();
+	/* Try using a XIVE if available, otherwise use a XICS */
+	if (!xive_spapr_init()) {
+		xics_init();
+		pseries_setup_i8259_cascade();
+	}
 }
 
 static void pseries_lpar_enable_pmcs(void)
diff --git a/arch/powerpc/platforms/pseries/smp.c b/arch/powerpc/platforms/pseries/smp.c
index 24785f63fb40..244bb9974963 100644
--- a/arch/powerpc/platforms/pseries/smp.c
+++ b/arch/powerpc/platforms/pseries/smp.c
@@ -41,6 +41,7 @@ 
 #include <asm/vdso_datapage.h>
 #include <asm/cputhreads.h>
 #include <asm/xics.h>
+#include <asm/xive.h>
 #include <asm/dbell.h>
 #include <asm/plpar_wrappers.h>
 #include <asm/code-patching.h>
@@ -136,7 +137,9 @@  static inline int smp_startup_cpu(unsigned int lcpu)
 
 static void smp_setup_cpu(int cpu)
 {
-	if (cpu != boot_cpuid)
+	if (xive_enabled())
+		xive_smp_setup_cpu();
+	else if (cpu != boot_cpuid)
 		xics_setup_cpu();
 
 	if (firmware_has_feature(FW_FEATURE_SPLPAR))
@@ -181,13 +184,22 @@  static int smp_pSeries_kick_cpu(int nr)
 	return 0;
 }
 
+static int pseries_smp_prepare_cpu(int cpu)
+{
+	if (xive_enabled())
+		return xive_smp_prepare_cpu(cpu);
+	return 0;
+}
+
+/* Cause IPI as setup by the interrupt controller (xics or xive) */
+static void (*ic_cause_ipi)(int cpu);
+
 static void smp_pseries_cause_ipi(int cpu)
 {
-	/* POWER9 should not use this handler */
 	if (doorbell_try_core_ipi(cpu))
 		return;
 
-	icp_ops->cause_ipi(cpu);
+	ic_cause_ipi(cpu);
 }
 
 static int pseries_cause_nmi_ipi(int cpu)
@@ -213,12 +225,17 @@  static int pseries_cause_nmi_ipi(int cpu)
 
 static __init void pSeries_smp_probe(void)
 {
-	xics_smp_probe();
+	if (xive_enabled())
+		xive_smp_probe();
+	else
+		xics_smp_probe();
+
+	if (cpu_has_feature(CPU_FTR_DBELL)) {
+		ic_cause_ipi = smp_ops->cause_ipi;
+		WARN_ON(!ic_cause_ipi);
 
-	if (cpu_has_feature(CPU_FTR_DBELL))
 		smp_ops->cause_ipi = smp_pseries_cause_ipi;
-	else
-		smp_ops->cause_ipi = icp_ops->cause_ipi;
+	}
 }
 
 static struct smp_ops_t pseries_smp_ops = {
@@ -226,6 +243,7 @@  static struct smp_ops_t pseries_smp_ops = {
 	.cause_ipi	= NULL,	/* Filled at runtime by pSeries_smp_probe() */
 	.cause_nmi_ipi	= pseries_cause_nmi_ipi,
 	.probe		= pSeries_smp_probe,
+	.prepare_cpu	= pseries_smp_prepare_cpu,
 	.kick_cpu	= smp_pSeries_kick_cpu,
 	.setup_cpu	= smp_setup_cpu,
 	.cpu_bootable	= smp_generic_cpu_bootable,
diff --git a/arch/powerpc/sysdev/xive/Kconfig b/arch/powerpc/sysdev/xive/Kconfig
index 12ccd7373d2f..3e3e25b5e30d 100644
--- a/arch/powerpc/sysdev/xive/Kconfig
+++ b/arch/powerpc/sysdev/xive/Kconfig
@@ -9,3 +9,8 @@  config PPC_XIVE_NATIVE
 	default n
 	select PPC_XIVE
 	depends on PPC_POWERNV
+
+config PPC_XIVE_SPAPR
+	bool
+	default n
+	select PPC_XIVE
diff --git a/arch/powerpc/sysdev/xive/Makefile b/arch/powerpc/sysdev/xive/Makefile
index 3fab303fc169..536d6e5706e3 100644
--- a/arch/powerpc/sysdev/xive/Makefile
+++ b/arch/powerpc/sysdev/xive/Makefile
@@ -2,3 +2,4 @@  subdir-ccflags-$(CONFIG_PPC_WERROR) := -Werror
 
 obj-y				+= common.o
 obj-$(CONFIG_PPC_XIVE_NATIVE)	+= native.o
+obj-$(CONFIG_PPC_XIVE_SPAPR)	+= spapr.o
diff --git a/arch/powerpc/sysdev/xive/spapr.c b/arch/powerpc/sysdev/xive/spapr.c
new file mode 100644
index 000000000000..a3668815d5c1
--- /dev/null
+++ b/arch/powerpc/sysdev/xive/spapr.c
@@ -0,0 +1,554 @@ 
+/*
+ * Copyright 2016,2017 IBM Corporation.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#define pr_fmt(fmt) "xive: " fmt
+
+#include <linux/types.h>
+#include <linux/irq.h>
+#include <linux/smp.h>
+#include <linux/interrupt.h>
+#include <linux/init.h>
+#include <linux/of.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/cpumask.h>
+#include <linux/mm.h>
+
+#include <asm/prom.h>
+#include <asm/io.h>
+#include <asm/smp.h>
+#include <asm/irq.h>
+#include <asm/errno.h>
+#include <asm/xive.h>
+#include <asm/xive-regs.h>
+#include <asm/hvcall.h>
+
+#include "xive-internal.h"
+
+static u32 xive_queue_shift;
+
+struct xive_irq_bitmap {
+	unsigned long		*bitmap;
+	unsigned int		base;
+	unsigned int		count;
+	spinlock_t		lock;
+	struct list_head	list;
+};
+
+static LIST_HEAD(xive_irq_bitmaps);
+
+static int xive_irq_bitmap_add(int base, int count)
+{
+	struct xive_irq_bitmap *xibm;
+
+	xibm = kzalloc(sizeof(*xibm), GFP_ATOMIC);
+	if (!xibm)
+		return -ENOMEM;
+
+	spin_lock_init(&xibm->lock);
+	xibm->base = base;
+	xibm->count = count;
+	xibm->bitmap = kzalloc(xibm->count, GFP_KERNEL);
+	list_add(&xibm->list, &xive_irq_bitmaps);
+
+	pr_info("Using IRQ range [%x-%x]", xibm->base,
+		xibm->base + xibm->count - 1);
+	return 0;
+}
+
+static int __xive_irq_bitmap_alloc(struct xive_irq_bitmap *xibm)
+{
+	int irq;
+
+	irq = find_first_zero_bit(xibm->bitmap, xibm->count);
+	if (irq != xibm->count) {
+		set_bit(irq, xibm->bitmap);
+		irq += xibm->base;
+	} else {
+		irq = -ENOMEM;
+	}
+
+	return irq;
+}
+
+static int xive_irq_bitmap_alloc(void)
+{
+	struct xive_irq_bitmap *xibm;
+	unsigned long flags;
+	int irq = -ENOENT;
+
+	list_for_each_entry(xibm, &xive_irq_bitmaps, list) {
+		spin_lock_irqsave(&xibm->lock, flags);
+		irq = __xive_irq_bitmap_alloc(xibm);
+		spin_unlock_irqrestore(&xibm->lock, flags);
+		if (irq >= 0)
+			break;
+	}
+	return irq;
+}
+
+static void xive_irq_bitmap_free(int irq)
+{
+	unsigned long flags;
+	struct xive_irq_bitmap *xibm;
+
+	list_for_each_entry(xibm, &xive_irq_bitmaps, list) {
+		if ((irq >= xibm->base) && (irq < xibm->base + xibm->count)) {
+			spin_lock_irqsave(&xibm->lock, flags);
+			clear_bit(irq - xibm->base, xibm->bitmap);
+			spin_unlock_irqrestore(&xibm->lock, flags);
+			break;
+		}
+	}
+}
+
+static long plpar_int_get_source_info(unsigned long flags,
+				      unsigned long lisn,
+				      unsigned long *src_flags,
+				      unsigned long *eoi_page,
+				      unsigned long *trig_page,
+				      unsigned long *esb_shift)
+{
+	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
+	long rc;
+
+	rc = plpar_hcall(H_INT_GET_SOURCE_INFO, retbuf, flags, lisn);
+	if (rc) {
+		pr_err("H_INT_GET_SOURCE_INFO lisn=%ld failed %ld\n", lisn, rc);
+		return rc;
+	}
+
+	*src_flags = retbuf[0];
+	*eoi_page  = retbuf[1];
+	*trig_page = retbuf[2];
+	*esb_shift = retbuf[3];
+
+	pr_devel("H_INT_GET_SOURCE_INFO flags=%lx eoi=%lx trig=%lx shift=%lx\n",
+		retbuf[0], retbuf[1], retbuf[2], retbuf[3]);
+
+	return 0;
+}
+
+#define XIVE_SRC_SET_EISN (1ull << (63 - 62))
+#define XIVE_SRC_MASK     (1ull << (63 - 63)) /* unused */
+
+static long plpar_int_set_source_config(unsigned long flags,
+					unsigned long lisn,
+					unsigned long target,
+					unsigned long prio,
+					unsigned long sw_irq)
+{
+	long rc;
+
+
+	pr_devel("H_INT_SET_SOURCE_CONFIG flags=%lx lisn=%lx target=%lx prio=%lx sw_irq=%lx\n",
+		flags, lisn, target, prio, sw_irq);
+
+
+	rc = plpar_hcall_norets(H_INT_SET_SOURCE_CONFIG, flags, lisn,
+				target, prio, sw_irq);
+	if (rc) {
+		pr_err("H_INT_SET_SOURCE_CONFIG lisn=%ld failed %ld\n",
+		       lisn, rc);
+		return rc;
+	}
+
+	return 0;
+}
+
+static long plpar_int_get_queue_info(unsigned long flags,
+				     unsigned long target,
+				     unsigned long priority,
+				     unsigned long *esn_page,
+				     unsigned long *esn_size)
+{
+	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
+	long rc;
+
+	rc = plpar_hcall(H_INT_GET_QUEUE_INFO, retbuf, flags, target, priority);
+	if (rc) {
+		pr_err("H_INT_GET_QUEUE_INFO cpu=%ld prio=%ld failed %ld\n",
+		       target, priority, rc);
+		return rc;
+	}
+
+	*esn_page = retbuf[0];
+	*esn_size = retbuf[1];
+
+	pr_devel("H_INT_GET_QUEUE_INFO page=%lx size=%lx\n",
+		retbuf[0], retbuf[1]);
+
+	return 0;
+}
+
+#define XIVE_EQ_ALWAYS_NOTIFY (1ull << (63 - 63))
+
+static long plpar_int_set_queue_config(unsigned long flags,
+				       unsigned long target,
+				       unsigned long priority,
+				       unsigned long qpage,
+				       unsigned long qsize)
+{
+	long rc;
+
+	pr_devel("H_INT_SET_QUEUE_CONFIG flags=%lx target=%lx priority=%lx qpage=%lx qsize=%lx\n",
+		flags,  target, priority, qpage, qsize);
+
+	rc = plpar_hcall_norets(H_INT_SET_QUEUE_CONFIG, flags, target,
+				priority, qpage, qsize);
+	if (rc) {
+		pr_err("H_INT_SET_QUEUE_CONFIG cpu=%ld prio=%ld qpage=%lx returned %ld\n",
+		       target, priority, qpage, rc);
+		return  rc;
+	}
+
+	return 0;
+}
+
+static long plpar_int_sync(unsigned long flags)
+{
+	long rc;
+
+	rc = plpar_hcall_norets(H_INT_SYNC, flags);
+	if (rc) {
+		pr_err("H_INT_SYNC returned %ld\n", rc);
+		return  rc;
+	}
+
+	return 0;
+}
+
+#define XIVE_SRC_H_INT_ESB     (1ull << (63 - 60)) /* TODO */
+#define XIVE_SRC_LSI           (1ull << (63 - 61))
+#define XIVE_SRC_TRIGGER       (1ull << (63 - 62))
+#define XIVE_SRC_STORE_EOI     (1ull << (63 - 63))
+
+static int xive_spapr_populate_irq_data(u32 hw_irq, struct xive_irq_data *data)
+{
+	long rc;
+	unsigned long flags;
+	unsigned long eoi_page;
+	unsigned long trig_page;
+	unsigned long esb_shift;
+
+	memset(data, 0, sizeof(*data));
+
+	rc = plpar_int_get_source_info(0, hw_irq, &flags, &eoi_page, &trig_page,
+				       &esb_shift);
+	if (rc)
+		return  -EINVAL;
+
+	if (flags & XIVE_SRC_STORE_EOI)
+		data->flags  |= XIVE_IRQ_FLAG_STORE_EOI;
+	if (flags & XIVE_SRC_LSI)
+		data->flags  |= XIVE_IRQ_FLAG_LSI;
+	data->eoi_page  = eoi_page;
+	data->esb_shift = esb_shift;
+	data->trig_page = trig_page;
+
+	/*
+	 * No chip-id for the sPAPR backend. This has an impact how we
+	 * pick a target. See xive_pick_irq_target().
+	 */
+	data->src_chip = XIVE_INVALID_CHIP_ID;
+
+	data->eoi_mmio = ioremap(data->eoi_page, 1u << data->esb_shift);
+	if (!data->eoi_mmio) {
+		pr_err("Failed to map EOI page for irq 0x%x\n", hw_irq);
+		return -ENOMEM;
+	}
+
+	/* Full function page supports trigger */
+	if (flags & XIVE_SRC_TRIGGER) {
+		data->trig_mmio = data->eoi_mmio;
+		return 0;
+	}
+
+	data->trig_mmio = ioremap(data->trig_page, 1u << data->esb_shift);
+	if (!data->trig_mmio) {
+		pr_err("Failed to map trigger page for irq 0x%x\n", hw_irq);
+		return -ENOMEM;
+	}
+	return 0;
+}
+
+static int xive_spapr_configure_irq(u32 hw_irq, u32 target, u8 prio, u32 sw_irq)
+{
+	long rc;
+
+	rc = plpar_int_set_source_config(XIVE_SRC_SET_EISN, hw_irq, target,
+					 prio, sw_irq);
+
+	return rc == 0 ? 0 : -ENXIO;
+}
+
+/* This can be called multiple time to change a queue configuration */
+static int xive_spapr_configure_queue(u32 target, struct xive_q *q, u8 prio,
+				   __be32 *qpage, u32 order)
+{
+	s64 rc = 0;
+	unsigned long esn_page;
+	unsigned long esn_size;
+	u64 flags, qpage_phys;
+
+	/* If there's an actual queue page, clean it */
+	if (order) {
+		if (WARN_ON(!qpage))
+			return -EINVAL;
+		qpage_phys = __pa(qpage);
+	} else {
+		qpage_phys = 0;
+	}
+
+	/* Initialize the rest of the fields */
+	q->msk = order ? ((1u << (order - 2)) - 1) : 0;
+	q->idx = 0;
+	q->toggle = 0;
+
+	rc = plpar_int_get_queue_info(0, target, prio, &esn_page, &esn_size);
+	if (rc) {
+		pr_err("Error %lld getting queue info prio %d\n", rc, prio);
+		rc = -EIO;
+		goto fail;
+	}
+
+	/* TODO: add support for the notification page */
+	q->eoi_phys = esn_page;
+
+	/* Default is to always notify */
+	flags = XIVE_EQ_ALWAYS_NOTIFY;
+
+	/* Configure and enable the queue in HW */
+	rc = plpar_int_set_queue_config(flags, target, prio, qpage_phys, order);
+	if (rc) {
+		pr_err("Error %lld setting queue for prio %d\n", rc, prio);
+		rc = -EIO;
+	} else {
+		q->qpage = qpage;
+	}
+fail:
+	return rc;
+}
+
+static int xive_spapr_setup_queue(unsigned int cpu, struct xive_cpu *xc,
+				  u8 prio)
+{
+	struct xive_q *q = &xc->queue[prio];
+	unsigned int alloc_order;
+	struct page *pages;
+	__be32 *qpage;
+
+	alloc_order = (xive_queue_shift > PAGE_SHIFT) ?
+		(xive_queue_shift - PAGE_SHIFT) : 0;
+	pages = alloc_pages_node(cpu_to_node(cpu), GFP_KERNEL, alloc_order);
+	if (!pages)
+		return -ENOMEM;
+	qpage = (__be32 *)page_address(pages);
+	memset(qpage, 0, 1 << xive_queue_shift);
+
+	return xive_spapr_configure_queue(cpu, q, prio, qpage,
+					  xive_queue_shift);
+}
+
+static void xive_spapr_cleanup_queue(unsigned int cpu, struct xive_cpu *xc,
+				  u8 prio)
+{
+	struct xive_q *q = &xc->queue[prio];
+	unsigned int alloc_order;
+	long rc;
+
+	rc = plpar_int_set_queue_config(0, cpu, prio, 0, 0);
+	if (rc)
+		pr_err("Error %ld setting queue for prio %d\n", rc, prio);
+
+	alloc_order = (xive_queue_shift > PAGE_SHIFT) ?
+		(xive_queue_shift - PAGE_SHIFT) : 0;
+	free_pages((unsigned long)q->qpage, alloc_order);
+	q->qpage = NULL;
+}
+
+static bool xive_spapr_match(struct device_node *node)
+{
+	return 1;
+}
+
+#ifdef CONFIG_SMP
+static int xive_spapr_get_ipi(unsigned int cpu, struct xive_cpu *xc)
+{
+	int irq = xive_irq_bitmap_alloc();
+
+	if (irq < 0) {
+		pr_err("Failed to allocate IPI on CPU %d\n", cpu);
+		return -ENXIO;
+	}
+
+	xc->hw_ipi = irq;
+	return 0;
+}
+
+static void xive_spapr_put_ipi(unsigned int cpu, struct xive_cpu *xc)
+{
+	xive_irq_bitmap_free(xc->hw_ipi);
+}
+#endif /* CONFIG_SMP */
+
+static void xive_spapr_shutdown(void)
+{
+	long rc;
+
+	rc = plpar_hcall_norets(H_INT_RESET, 0);
+	if (rc)
+		pr_err("H_INT_RESET failed %ld\n", rc);
+}
+
+static void xive_spapr_update_pending(struct xive_cpu *xc)
+{
+	u8 nsr, cppr;
+	u16 ack;
+
+	/* Perform the acknowledge hypervisor to register cycle */
+	ack = be16_to_cpu(__raw_readw(xive_tima + TM_SPC_ACK_OS_REG));
+
+	/* Synchronize subsequent queue accesses */
+	mb();
+
+	/*
+	 * Grab the CPPR and the "NSR" field which indicates the source
+	 * of the hypervisor interrupt (if any)
+	 */
+	cppr = ack & 0xff;
+	nsr = ack >> 8;
+
+	if (nsr & TM_QW1_NSR_EO) {
+		if (cppr == 0xff)
+			return;
+		/* Mark the priority pending */
+		xc->pending_prio |= 1 << cppr;
+
+		/*
+		 * A new interrupt should never have a CPPR less favored
+		 * than our current one.
+		 */
+		if (cppr >= xc->cppr)
+			pr_err("CPU %d odd ack CPPR, got %d at %d\n",
+			       smp_processor_id(), cppr, xc->cppr);
+
+		/* Update our idea of what the CPPR is */
+		xc->cppr = cppr;
+	}
+}
+
+static void xive_spapr_eoi(u32 hw_irq)
+{
+	/* Not used */;
+}
+
+static void xive_spapr_setup_cpu(unsigned int cpu, struct xive_cpu *xc)
+{
+	/* Only some debug on the TIMA settings */
+	pr_debug("(HW value: %08x %08x %08x)\n",
+		 in_be32(xive_tima + TM_QW1_OS + TM_WORD0),
+		 in_be32(xive_tima + TM_QW1_OS + TM_WORD1),
+		 in_be32(xive_tima + TM_QW1_OS + TM_WORD2));
+}
+
+static void xive_spapr_teardown_cpu(unsigned int cpu, struct xive_cpu *xc)
+{
+	/* Nothing to do */;
+}
+
+static void xive_spapr_sync_source(u32 hw_irq)
+{
+	/* Specs are unclear on what this is doing */
+	plpar_int_sync(0);
+}
+
+static const struct xive_ops xive_spapr_ops = {
+	.populate_irq_data	= xive_spapr_populate_irq_data,
+	.configure_irq		= xive_spapr_configure_irq,
+	.setup_queue		= xive_spapr_setup_queue,
+	.cleanup_queue		= xive_spapr_cleanup_queue,
+	.match			= xive_spapr_match,
+	.shutdown		= xive_spapr_shutdown,
+	.update_pending		= xive_spapr_update_pending,
+	.eoi			= xive_spapr_eoi,
+	.setup_cpu		= xive_spapr_setup_cpu,
+	.teardown_cpu		= xive_spapr_teardown_cpu,
+	.sync_source		= xive_spapr_sync_source,
+#ifdef CONFIG_SMP
+	.get_ipi		= xive_spapr_get_ipi,
+	.put_ipi		= xive_spapr_put_ipi,
+#endif /* CONFIG_SMP */
+	.name			= "spapr",
+};
+
+bool xive_spapr_init(void)
+{
+	struct device_node *np;
+	struct resource r;
+	void __iomem *tima;
+	struct property *prop;
+	u8 max_prio = 7;
+	u32 val;
+	u32 len;
+	const __be32 *reg;
+	int i;
+
+	if (xive_cmdline_disabled)
+		return false;
+
+	pr_devel("%s()\n", __func__);
+	np = of_find_compatible_node(NULL, NULL, "ibm,power-ivpe");
+	if (!np) {
+		pr_devel("not found !\n");
+		return false;
+	}
+	pr_devel("Found %s\n", np->full_name);
+
+	/* Resource 1 is the OS ring TIMA */
+	if (of_address_to_resource(np, 1, &r)) {
+		pr_err("Failed to get thread mgmnt area resource\n");
+		return false;
+	}
+	tima = ioremap(r.start, resource_size(&r));
+	if (!tima) {
+		pr_err("Failed to map thread mgmnt area\n");
+		return false;
+	}
+
+	/* Feed the IRQ number allocator with the ranges given in the DT */
+	reg = of_get_property(np, "ibm,xive-lisn-ranges", &len);
+	if (!reg) {
+		pr_err("Failed to read 'ibm,xive-lisn-ranges' property\n");
+		return false;
+	}
+
+	if (len % (2 * sizeof(u32)) != 0) {
+		pr_err("invalid 'ibm,xive-lisn-ranges' property\n");
+		return false;
+	}
+
+	for (i = 0; i < len / (2 * sizeof(u32)); i++, reg += 2)
+		xive_irq_bitmap_add(be32_to_cpu(reg[0]),
+				    be32_to_cpu(reg[1]));
+
+	/* Iterate the EQ sizes and pick one */
+	of_property_for_each_u32(np, "ibm,xive-eq-sizes", prop, reg, val) {
+		xive_queue_shift = val;
+		if (val == PAGE_SHIFT)
+			break;
+	}
+
+	/* Initialize XIVE core with our backend */
+	if (!xive_core_init(&xive_spapr_ops, tima, TM_QW1_OS, max_prio))
+		return false;
+
+	pr_info("Using %dkB queues\n", 1 << (xive_queue_shift - 10));
+	return true;
+}