[PATCH 01/21] powerpc/udbg: Fix Kconfig entry for avoiding 44x early debug with KVM

[PATCH 01/21] powerpc/udbg: Fix Kconfig entry for avoiding 44x early debug with KVM

[Benjamin Herrenschmidt]

It was preventing the global early debug selection whenever KVM was enabled
instead of only preventing the 440 specific one.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/Kconfig.debug |    7 +++----
 1 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/arch/powerpc/Kconfig.debug b/arch/powerpc/Kconfig.debug
index 067cb84..cc01f1d 100644
--- a/arch/powerpc/Kconfig.debug
+++ b/arch/powerpc/Kconfig.debug
@@ -141,9 +141,6 @@ config BOOTX_TEXT
 
 config PPC_EARLY_DEBUG
 	bool "Early debugging (dangerous)"
-	# PPC_EARLY_DEBUG on 440 leaves AS=1 mappings above the TLB high water
-	# mark, which doesn't work with current 440 KVM.
-	depends on !KVM
 	help
 	  Say Y to enable some early debugging facilities that may be available
 	  for your processor/board combination. Those facilities are hacks
@@ -222,7 +219,9 @@ config PPC_EARLY_DEBUG_BEAT
 
 config PPC_EARLY_DEBUG_44x
 	bool "Early serial debugging for IBM/AMCC 44x CPUs"
-	depends on 44x
+	# PPC_EARLY_DEBUG on 440 leaves AS=1 mappings above the TLB high water
+	# mark, which doesn't work with current 440 KVM.
+	depends on 44x && !KVM
 	help
 	  Select this to enable early debugging for IBM 44x chips via the
 	  inbuilt serial port.  If you enable this, ensure you set
-- 
1.7.4.1

[PATCH 02/21] powerpc/smp: More generic support for "soft hotplug"

[Benjamin Herrenschmidt]

This adds more generic support for doing CPU hotplug with a simple
idle loop and no actual reset of the processors. The generic
smp_generic_kick_cpu() does the hotplug bringup trick if the PACA
shows that the CPU has already been started at boot and we provide
an accessor for the CPU state.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/include/asm/smp.h |    1 +
 arch/powerpc/kernel/smp.c      |   30 +++++++++++++++++++++++++-----
 2 files changed, 26 insertions(+), 5 deletions(-)

diff --git a/arch/powerpc/include/asm/smp.h b/arch/powerpc/include/asm/smp.h
index 15a70b7..adba970 100644
--- a/arch/powerpc/include/asm/smp.h
+++ b/arch/powerpc/include/asm/smp.h
@@ -65,6 +65,7 @@ int generic_cpu_disable(void);
 void generic_cpu_die(unsigned int cpu);
 void generic_mach_cpu_die(void);
 void generic_set_cpu_dead(unsigned int cpu);
+int generic_check_cpu_restart(unsigned int cpu);
 #endif
 
 #ifdef CONFIG_PPC64
diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index 7bf2187..af7e772 100644
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -70,6 +70,10 @@
 static DEFINE_PER_CPU(struct task_struct *, idle_thread_array);
 #define get_idle_for_cpu(x)      (per_cpu(idle_thread_array, x))
 #define set_idle_for_cpu(x, p)   (per_cpu(idle_thread_array, x) = (p))
+
+/* State of each CPU during hotplug phases */
+static DEFINE_PER_CPU(int, cpu_state) = { 0 };
+
 #else
 static struct task_struct *idle_thread_array[NR_CPUS] __cpuinitdata ;
 #define get_idle_for_cpu(x)      (idle_thread_array[(x)])
@@ -104,12 +108,25 @@ int __devinit smp_generic_kick_cpu(int nr)
 	 * cpu_start field to become non-zero After we set cpu_start,
 	 * the processor will continue on to secondary_start
 	 */
-	paca[nr].cpu_start = 1;
-	smp_mb();
+	if (!paca[nr].cpu_start) {
+		paca[nr].cpu_start = 1;
+		smp_mb();
+		return 0;
+	}
+
+#ifdef CONFIG_HOTPLUG_CPU
+	/*
+	 * Ok it's not there, so it might be soft-unplugged, let's
+	 * try to bring it back
+	 */
+	per_cpu(cpu_state, nr) = CPU_UP_PREPARE;
+	smp_wmb();
+	smp_send_reschedule(nr);
+#endif /* CONFIG_HOTPLUG_CPU */
 
 	return 0;
 }
-#endif
+#endif /* CONFIG_PPC64 */
 
 static irqreturn_t call_function_action(int irq, void *data)
 {
@@ -357,8 +374,6 @@ void __devinit smp_prepare_boot_cpu(void)
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
-/* State of each CPU during hotplug phases */
-static DEFINE_PER_CPU(int, cpu_state) = { 0 };
 
 int generic_cpu_disable(void)
 {
@@ -406,6 +421,11 @@ void generic_set_cpu_dead(unsigned int cpu)
 {
 	per_cpu(cpu_state, cpu) = CPU_DEAD;
 }
+
+int generic_check_cpu_restart(unsigned int cpu)
+{
+	return per_cpu(cpu_state, cpu) == CPU_UP_PREPARE;
+}
 #endif
 
 struct create_idle {
-- 
1.7.4.1

[PATCH 03/21] powerpc/pci: Call pcie_bus_configure_settings()

[Benjamin Herrenschmidt]

This new function is used to properly setup the PCI Express Max Payload Size
(and in some circumstances Max Read Request Size).

Some systems will not operate properly if these aren't set correctly and
the firmware doesn't always do it.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/pci-common.c |   11 +++++++++++
 1 files changed, 11 insertions(+), 0 deletions(-)

diff --git a/arch/powerpc/kernel/pci-common.c b/arch/powerpc/kernel/pci-common.c
index 32656f1..1bd47f3 100644
--- a/arch/powerpc/kernel/pci-common.c
+++ b/arch/powerpc/kernel/pci-common.c
@@ -1730,6 +1730,17 @@ void __devinit pcibios_scan_phb(struct pci_controller *hose)
 
 	if (mode == PCI_PROBE_NORMAL)
 		hose->last_busno = bus->subordinate = pci_scan_child_bus(bus);
+
+	/* Configure PCI Express settings */
+	if (bus) {
+		struct pci_bus *child;
+		list_for_each_entry(child, &bus->children, node) {
+			struct pci_dev *self = child->self;
+			if (!self)
+				continue;
+			pcie_bus_configure_settings(child, self->pcie_mpss);
+		}
+	}
 }
 
 static void fixup_hide_host_resource_fsl(struct pci_dev *dev)
-- 
1.7.4.1

[PATCH 04/21] tty/hvc/hvsi_lib: Updates for running under OPAL v2

[Benjamin Herrenschmidt]

Experience shows that our hvsi negotiation timeouts were too short
under OPAL v2. This patch bumps them from 200ms to 500ms.

To limit the impact when the console is repeatedly opened/closed,
we also avoid re-negotiating an already opened & established
connection which typically happens when running init scripts
on the tty shared with the kernel console.

Finally, while touching hvsi, I added a comment explaining why we
limit our output packet size to 12 bytes.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/include/asm/hvsi.h |    4 ++++
 drivers/tty/hvc/hvsi_lib.c      |   12 +++++++-----
 2 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/arch/powerpc/include/asm/hvsi.h b/arch/powerpc/include/asm/hvsi.h
index d3f64f3..ca85d9d 100644
--- a/arch/powerpc/include/asm/hvsi.h
+++ b/arch/powerpc/include/asm/hvsi.h
@@ -19,6 +19,10 @@
 #define HVSI_TSDTR 0x01
 #define HVSI_TSCD  0x20
 
+/* Never send more than 12 bytes at once, that is 16 bytes
+ * with header, which is the max pHyp can swallow in one
+ * hcall.
+ */
 #define HVSI_MAX_OUTGOING_DATA 12
 #define HVSI_VERSION 1
 
diff --git a/drivers/tty/hvc/hvsi_lib.c b/drivers/tty/hvc/hvsi_lib.c
index bd9b098..08e3e90 100644
--- a/drivers/tty/hvc/hvsi_lib.c
+++ b/drivers/tty/hvc/hvsi_lib.c
@@ -316,10 +316,10 @@ void hvsilib_establish(struct hvsi_priv *pv)
 
 	pr_devel("HVSI@%x: Establishing...\n", pv->termno);
 
-	/* Try for up to 200ms, there can be a packet to
+	/* Try for up to 500ms, there can be a packet to
 	 * start the process waiting for us...
 	 */
-	for (timeout = 0; timeout < 20; timeout++) {
+	for (timeout = 0; timeout < 50; timeout++) {
 		if (pv->established)
 			goto established;
 		if (!hvsi_get_packet(pv))
@@ -341,8 +341,8 @@ void hvsilib_establish(struct hvsi_priv *pv)
 
 	pr_devel("HVSI@%x:   ... waiting handshake\n", pv->termno);
 
-	/* Try for up to 200s */
-	for (timeout = 0; timeout < 20; timeout++) {
+	/* Try for up to 500ms */
+	for (timeout = 0; timeout < 50; timeout++) {
 		if (pv->established)
 			goto established;
 		if (!hvsi_get_packet(pv))
@@ -379,7 +379,9 @@ int hvsilib_open(struct hvsi_priv *pv, struct hvc_struct *hp)
 	/* Keep track of the tty data structure */
 	pv->tty = tty_kref_get(hp->tty);
 
-	hvsilib_establish(pv);
+	WARN_ON(pv->opened && !pv->is_console);
+	if (!pv->opened)
+		hvsilib_establish(pv);
 
 	return 0;
 }
-- 
1.7.4.1

[PATCH 05/21] powerpc/powernv: Don't clobber r9 in relative_toc()

[Benjamin Herrenschmidt]

With OPAL, r8 and r9 will be used to pass the OPAL base and entry
for debugging purposes (those informations are also in the
device-tree). We don't want to clobber those registers that
early.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/head_64.S |    6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index 3564c49..e708abe 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -674,9 +674,9 @@ _GLOBAL(enable_64b_mode)
 _GLOBAL(relative_toc)
 	mflr	r0
 	bcl	20,31,$+4
-0:	mflr	r9
-	ld	r2,(p_toc - 0b)(r9)
-	add	r2,r2,r9
+0:	mflr	r11
+	ld	r2,(p_toc - 0b)(r11)
+	add	r2,r2,r11
 	mtlr	r0
 	blr
 
-- 
1.7.4.1

[PATCH 06/21] powerpc: Add skeleton PowerNV platform

[Benjamin Herrenschmidt]

This adds a skeletton for the new Power "Non Virtualized"
platform which will be used by machines supporting running
without an hypervisor, for example in order to run KVM.

These machines will be using a new firmware called OPAL
for which the support will be provided by later patches.

The PowerNV platform is intended to be also usable under
the BML environment used internally for early CPU bringup
which is why the code also supports using RTAS instead of
OPAL in various places.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/boot/Makefile               |    1 +
 arch/powerpc/platforms/Kconfig           |    1 +
 arch/powerpc/platforms/Makefile          |    1 +
 arch/powerpc/platforms/powernv/Kconfig   |   12 +++
 arch/powerpc/platforms/powernv/Makefile  |    2 +
 arch/powerpc/platforms/powernv/powernv.h |   10 ++
 arch/powerpc/platforms/powernv/setup.c   |  153 ++++++++++++++++++++++++++++++
 arch/powerpc/platforms/powernv/smp.c     |   83 ++++++++++++++++
 8 files changed, 263 insertions(+), 0 deletions(-)
 create mode 100644 arch/powerpc/platforms/powernv/Kconfig
 create mode 100644 arch/powerpc/platforms/powernv/Makefile
 create mode 100644 arch/powerpc/platforms/powernv/powernv.h
 create mode 100644 arch/powerpc/platforms/powernv/setup.c
 create mode 100644 arch/powerpc/platforms/powernv/smp.c

diff --git a/arch/powerpc/boot/Makefile b/arch/powerpc/boot/Makefile
index c26200b..52cde90 100644
--- a/arch/powerpc/boot/Makefile
+++ b/arch/powerpc/boot/Makefile
@@ -171,6 +171,7 @@ quiet_cmd_wrap	= WRAP    $@
 		$(if $3, -s $3)$(if $4, -d $4)$(if $5, -i $5) vmlinux
 
 image-$(CONFIG_PPC_PSERIES)		+= zImage.pseries
+image-$(CONFIG_PPC_POWERNV)		+= zImage.pseries
 image-$(CONFIG_PPC_MAPLE)		+= zImage.maple
 image-$(CONFIG_PPC_IBM_CELL_BLADE)	+= zImage.pseries
 image-$(CONFIG_PPC_PS3)			+= dtbImage.ps3
diff --git a/arch/powerpc/platforms/Kconfig b/arch/powerpc/platforms/Kconfig
index b9ba861..6de27d2 100644
--- a/arch/powerpc/platforms/Kconfig
+++ b/arch/powerpc/platforms/Kconfig
@@ -1,5 +1,6 @@
 menu "Platform support"
 
+source "arch/powerpc/platforms/powernv/Kconfig"
 source "arch/powerpc/platforms/pseries/Kconfig"
 source "arch/powerpc/platforms/iseries/Kconfig"
 source "arch/powerpc/platforms/chrp/Kconfig"
diff --git a/arch/powerpc/platforms/Makefile b/arch/powerpc/platforms/Makefile
index 73e2116..2635a22 100644
--- a/arch/powerpc/platforms/Makefile
+++ b/arch/powerpc/platforms/Makefile
@@ -14,6 +14,7 @@ obj-$(CONFIG_PPC_82xx)		+= 82xx/
 obj-$(CONFIG_PPC_83xx)		+= 83xx/
 obj-$(CONFIG_FSL_SOC_BOOKE)	+= 85xx/
 obj-$(CONFIG_PPC_86xx)		+= 86xx/
+obj-$(CONFIG_PPC_POWERNV)	+= powernv/
 obj-$(CONFIG_PPC_PSERIES)	+= pseries/
 obj-$(CONFIG_PPC_ISERIES)	+= iseries/
 obj-$(CONFIG_PPC_MAPLE)		+= maple/
diff --git a/arch/powerpc/platforms/powernv/Kconfig b/arch/powerpc/platforms/powernv/Kconfig
new file mode 100644
index 0000000..5cd04f5
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/Kconfig
@@ -0,0 +1,12 @@
+config PPC_POWERNV
+	depends on PPC64 && PPC_BOOK3S
+	bool "IBM PowerNV (Non-Virtualized) platform support"	
+	select PPC_RTAS
+	select PPC_NATIVE
+	select PPC_XICS
+	select PPC_ICP_NATIVE
+	select PPC_ICS_RTAS
+	select PPC_P7_NAP
+	select PPC_PCI_CHOICE if EMBEDDED
+	default y
+
diff --git a/arch/powerpc/platforms/powernv/Makefile b/arch/powerpc/platforms/powernv/Makefile
new file mode 100644
index 0000000..1c43250
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/Makefile
@@ -0,0 +1,2 @@
+obj-y			+= setup.o
+obj-$(CONFIG_SMP)	+= smp.o
diff --git a/arch/powerpc/platforms/powernv/powernv.h b/arch/powerpc/platforms/powernv/powernv.h
new file mode 100644
index 0000000..35b7160
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/powernv.h
@@ -0,0 +1,10 @@
+#ifndef _POWERNV_H
+#define _POWERNV_H
+
+#ifdef CONFIG_SMP
+extern void pnv_smp_init(void);
+#else
+static inline void pnv_smp_init(void) { }
+#endif
+
+#endif /* _POWERNV_H */
diff --git a/arch/powerpc/platforms/powernv/setup.c b/arch/powerpc/platforms/powernv/setup.c
new file mode 100644
index 0000000..569f9cc
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/setup.c
@@ -0,0 +1,153 @@
+/*
+ * PowerNV setup code.
+ *
+ * Copyright 2011 IBM Corp.
+ *
+ * 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.
+ */
+
+#undef DEBUG
+
+#include <linux/cpu.h>
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/tty.h>
+#include <linux/reboot.h>
+#include <linux/init.h>
+#include <linux/console.h>
+#include <linux/delay.h>
+#include <linux/irq.h>
+#include <linux/seq_file.h>
+#include <linux/of.h>
+#include <linux/interrupt.h>
+#include <linux/bug.h>
+
+#include <asm/machdep.h>
+#include <asm/firmware.h>
+#include <asm/xics.h>
+
+#include "powernv.h"
+
+static void __init pnv_setup_arch(void)
+{
+	/* Force console to hvc for now until we have sorted out the
+	 * real console situation for the platform. This will make
+	 * hvc_udbg work at least.
+	 */
+	add_preferred_console("hvc", 0, NULL);
+
+	/* Initialize SMP */
+	pnv_smp_init();
+
+	/* XXX PCI */
+
+	/* XXX NVRAM */
+
+	/* Enable NAP mode */
+	powersave_nap = 1;
+
+	/* XXX PMCS */
+}
+
+static void __init pnv_init_early(void)
+{
+	/* XXX IOMMU */
+}
+
+static void __init pnv_init_IRQ(void)
+{
+	xics_init();
+
+	WARN_ON(!ppc_md.get_irq);
+}
+
+static void pnv_show_cpuinfo(struct seq_file *m)
+{
+	struct device_node *root;
+	const char *model = "";
+
+	root = of_find_node_by_path("/");
+	if (root)
+		model = of_get_property(root, "model", NULL);
+	seq_printf(m, "machine\t\t: PowerNV %s\n", model);
+	of_node_put(root);
+}
+
+static void pnv_restart(char *cmd)
+{
+	for (;;);
+}
+
+static void pnv_power_off(void)
+{
+	for (;;);
+}
+
+static void pnv_halt(void)
+{
+	for (;;);
+}
+
+static unsigned long __init pnv_get_boot_time(void)
+{
+	return 0;
+}
+
+static void pnv_get_rtc_time(struct rtc_time *rtc_tm)
+{
+}
+
+static int pnv_set_rtc_time(struct rtc_time *tm)
+{
+	return 0;
+}
+
+static void pnv_progress(char *s, unsigned short hex)
+{
+}
+
+#ifdef CONFIG_KEXEC
+static void pnv_kexec_cpu_down(int crash_shutdown, int secondary)
+{
+	xics_kexec_teardown_cpu(secondary);
+}
+#endif /* CONFIG_KEXEC */
+
+static int __init pnv_probe(void)
+{
+	unsigned long root = of_get_flat_dt_root();
+
+	if (!of_flat_dt_is_compatible(root, "ibm,powernv"))
+		return 0;
+
+	hpte_init_native();
+
+	pr_debug("PowerNV detected !\n");
+
+	return 1;
+}
+
+define_machine(powernv) {
+	.name			= "PowerNV",
+	.probe			= pnv_probe,
+	.setup_arch		= pnv_setup_arch,
+	.init_early		= pnv_init_early,
+	.init_IRQ		= pnv_init_IRQ,
+	.show_cpuinfo		= pnv_show_cpuinfo,
+	.restart		= pnv_restart,
+	.power_off		= pnv_power_off,
+	.halt			= pnv_halt,
+	.get_boot_time		= pnv_get_boot_time,
+	.get_rtc_time		= pnv_get_rtc_time,
+	.set_rtc_time		= pnv_set_rtc_time,
+	.progress		= pnv_progress,
+	.power_save             = power7_idle,
+	.calibrate_decr		= generic_calibrate_decr,
+#ifdef CONFIG_KEXEC
+	.kexec_cpu_down		= pnv_kexec_cpu_down,
+#endif
+};
diff --git a/arch/powerpc/platforms/powernv/smp.c b/arch/powerpc/platforms/powernv/smp.c
new file mode 100644
index 0000000..36c7151
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/smp.c
@@ -0,0 +1,83 @@
+/*
+ * SMP support for PowerNV machines.
+ *
+ * Copyright 2011 IBM Corp.
+ *
+ * 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.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/smp.h>
+#include <linux/interrupt.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/cpu.h>
+
+#include <asm/irq.h>
+#include <asm/smp.h>
+#include <asm/paca.h>
+#include <asm/machdep.h>
+#include <asm/cputable.h>
+#include <asm/firmware.h>
+#include <asm/system.h>
+#include <asm/rtas.h>
+#include <asm/vdso_datapage.h>
+#include <asm/cputhreads.h>
+#include <asm/xics.h>
+
+#include "powernv.h"
+
+static void __devinit pnv_smp_setup_cpu(int cpu)
+{
+	if (cpu != boot_cpuid)
+		xics_setup_cpu();
+}
+
+static int pnv_smp_cpu_bootable(unsigned int nr)
+{
+	/* Special case - we inhibit secondary thread startup
+	 * during boot if the user requests it.
+	 */
+	if (system_state < SYSTEM_RUNNING && cpu_has_feature(CPU_FTR_SMT)) {
+		if (!smt_enabled_at_boot && cpu_thread_in_core(nr) != 0)
+			return 0;
+		if (smt_enabled_at_boot
+		    && cpu_thread_in_core(nr) >= smt_enabled_at_boot)
+			return 0;
+	}
+
+	return 1;
+}
+
+static struct smp_ops_t pnv_smp_ops = {
+	.message_pass	= smp_muxed_ipi_message_pass,
+	.cause_ipi	= NULL,	/* Filled at runtime by xics_smp_probe() */
+	.probe		= xics_smp_probe,
+	.kick_cpu	= smp_generic_kick_cpu,
+	.setup_cpu	= pnv_smp_setup_cpu,
+	.cpu_bootable	= pnv_smp_cpu_bootable,
+};
+
+/* This is called very early during platform setup_arch */
+void __init pnv_smp_init(void)
+{
+	smp_ops = &pnv_smp_ops;
+
+	/* XXX We don't yet have a proper entry point from HAL, for
+	 * now we rely on kexec-style entry from BML
+	 */
+
+#ifdef CONFIG_PPC_RTAS
+	/* Non-lpar has additional take/give timebase */
+	if (rtas_token("freeze-time-base") != RTAS_UNKNOWN_SERVICE) {
+		smp_ops->give_timebase = rtas_give_timebase;
+		smp_ops->take_timebase = rtas_take_timebase;
+	}
+#endif /* CONFIG_PPC_RTAS */
+}
-- 
1.7.4.1

Re: [PATCH 06/21] powerpc: Add skeleton PowerNV platform

[Michael Neuling]

In message <1315664408-16797-6-git-send-email-benh@kernel.crashing.org> you wro
te:
> This adds a skeletton for the new Power "Non Virtualized"
> platform which will be used by machines supporting running
> without an hypervisor, for example in order to run KVM.
> 
> These machines will be using a new firmware called OPAL
> for which the support will be provided by later patches.
> 
> The PowerNV platform is intended to be also usable under
> the BML environment used internally for early CPU bringup
> which is why the code also supports using RTAS instead of
> OPAL in various places.
> 
> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>

Giggity-Giggity-by: Michael Neuling <mikey@neuling.org>

> ---
>  arch/powerpc/boot/Makefile               |    1 +
>  arch/powerpc/platforms/Kconfig           |    1 +
>  arch/powerpc/platforms/Makefile          |    1 +
>  arch/powerpc/platforms/powernv/Kconfig   |   12 +++
>  arch/powerpc/platforms/powernv/Makefile  |    2 +
>  arch/powerpc/platforms/powernv/powernv.h |   10 ++
>  arch/powerpc/platforms/powernv/setup.c   |  153 ++++++++++++++++++++++++++++
++
>  arch/powerpc/platforms/powernv/smp.c     |   83 ++++++++++++++++
>  8 files changed, 263 insertions(+), 0 deletions(-)
>  create mode 100644 arch/powerpc/platforms/powernv/Kconfig
>  create mode 100644 arch/powerpc/platforms/powernv/Makefile
>  create mode 100644 arch/powerpc/platforms/powernv/powernv.h
>  create mode 100644 arch/powerpc/platforms/powernv/setup.c
>  create mode 100644 arch/powerpc/platforms/powernv/smp.c
> 
> diff --git a/arch/powerpc/boot/Makefile b/arch/powerpc/boot/Makefile
> index c26200b..52cde90 100644
> --- a/arch/powerpc/boot/Makefile
> +++ b/arch/powerpc/boot/Makefile
> @@ -171,6 +171,7 @@ quiet_cmd_wrap	= WRAP    $@
>  		$(if $3, -s $3)$(if $4, -d $4)$(if $5, -i $5) vmlinux
>  
>  image-$(CONFIG_PPC_PSERIES)		+= zImage.pseries
> +image-$(CONFIG_PPC_POWERNV)		+= zImage.pseries
>  image-$(CONFIG_PPC_MAPLE)		+= zImage.maple
>  image-$(CONFIG_PPC_IBM_CELL_BLADE)	+= zImage.pseries
>  image-$(CONFIG_PPC_PS3)			+= dtbImage.ps3
> diff --git a/arch/powerpc/platforms/Kconfig b/arch/powerpc/platforms/Kconfig
> index b9ba861..6de27d2 100644
> --- a/arch/powerpc/platforms/Kconfig
> +++ b/arch/powerpc/platforms/Kconfig
> @@ -1,5 +1,6 @@
>  menu "Platform support"
>  
> +source "arch/powerpc/platforms/powernv/Kconfig"
>  source "arch/powerpc/platforms/pseries/Kconfig"
>  source "arch/powerpc/platforms/iseries/Kconfig"
>  source "arch/powerpc/platforms/chrp/Kconfig"
> diff --git a/arch/powerpc/platforms/Makefile b/arch/powerpc/platforms/Makefil
e
> index 73e2116..2635a22 100644
> --- a/arch/powerpc/platforms/Makefile
> +++ b/arch/powerpc/platforms/Makefile
> @@ -14,6 +14,7 @@ obj-$(CONFIG_PPC_82xx)		+= 82xx/
>  obj-$(CONFIG_PPC_83xx)		+= 83xx/
>  obj-$(CONFIG_FSL_SOC_BOOKE)	+= 85xx/
>  obj-$(CONFIG_PPC_86xx)		+= 86xx/
> +obj-$(CONFIG_PPC_POWERNV)	+= powernv/
>  obj-$(CONFIG_PPC_PSERIES)	+= pseries/
>  obj-$(CONFIG_PPC_ISERIES)	+= iseries/
>  obj-$(CONFIG_PPC_MAPLE)		+= maple/
> diff --git a/arch/powerpc/platforms/powernv/Kconfig b/arch/powerpc/platforms/
powernv/Kconfig
> new file mode 100644
> index 0000000..5cd04f5
> --- /dev/null
> +++ b/arch/powerpc/platforms/powernv/Kconfig
> @@ -0,0 +1,12 @@
> +config PPC_POWERNV
> +	depends on PPC64 && PPC_BOOK3S
> +	bool "IBM PowerNV (Non-Virtualized) platform support"	
> +	select PPC_RTAS
> +	select PPC_NATIVE
> +	select PPC_XICS
> +	select PPC_ICP_NATIVE
> +	select PPC_ICS_RTAS
> +	select PPC_P7_NAP
> +	select PPC_PCI_CHOICE if EMBEDDED
> +	default y
> +
> diff --git a/arch/powerpc/platforms/powernv/Makefile b/arch/powerpc/platforms
/powernv/Makefile
> new file mode 100644
> index 0000000..1c43250
> --- /dev/null
> +++ b/arch/powerpc/platforms/powernv/Makefile
> @@ -0,0 +1,2 @@
> +obj-y			+= setup.o
> +obj-$(CONFIG_SMP)	+= smp.o
> diff --git a/arch/powerpc/platforms/powernv/powernv.h b/arch/powerpc/platform
s/powernv/powernv.h
> new file mode 100644
> index 0000000..35b7160
> --- /dev/null
> +++ b/arch/powerpc/platforms/powernv/powernv.h
> @@ -0,0 +1,10 @@
> +#ifndef _POWERNV_H
> +#define _POWERNV_H
> +
> +#ifdef CONFIG_SMP
> +extern void pnv_smp_init(void);
> +#else
> +static inline void pnv_smp_init(void) { }
> +#endif
> +
> +#endif /* _POWERNV_H */
> diff --git a/arch/powerpc/platforms/powernv/setup.c b/arch/powerpc/platforms/
powernv/setup.c
> new file mode 100644
> index 0000000..569f9cc
> --- /dev/null
> +++ b/arch/powerpc/platforms/powernv/setup.c
> @@ -0,0 +1,153 @@
> +/*
> + * PowerNV setup code.
> + *
> + * Copyright 2011 IBM Corp.
> + *
> + * 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.
> + */
> +
> +#undef DEBUG
> +
> +#include <linux/cpu.h>
> +#include <linux/errno.h>
> +#include <linux/sched.h>
> +#include <linux/kernel.h>
> +#include <linux/tty.h>
> +#include <linux/reboot.h>
> +#include <linux/init.h>
> +#include <linux/console.h>
> +#include <linux/delay.h>
> +#include <linux/irq.h>
> +#include <linux/seq_file.h>
> +#include <linux/of.h>
> +#include <linux/interrupt.h>
> +#include <linux/bug.h>
> +
> +#include <asm/machdep.h>
> +#include <asm/firmware.h>
> +#include <asm/xics.h>
> +
> +#include "powernv.h"
> +
> +static void __init pnv_setup_arch(void)
> +{
> +	/* Force console to hvc for now until we have sorted out the
> +	 * real console situation for the platform. This will make
> +	 * hvc_udbg work at least.
> +	 */
> +	add_preferred_console("hvc", 0, NULL);
> +
> +	/* Initialize SMP */
> +	pnv_smp_init();
> +
> +	/* XXX PCI */
> +
> +	/* XXX NVRAM */
> +
> +	/* Enable NAP mode */
> +	powersave_nap = 1;
> +
> +	/* XXX PMCS */
> +}
> +
> +static void __init pnv_init_early(void)
> +{
> +	/* XXX IOMMU */
> +}
> +
> +static void __init pnv_init_IRQ(void)
> +{
> +	xics_init();
> +
> +	WARN_ON(!ppc_md.get_irq);
> +}
> +
> +static void pnv_show_cpuinfo(struct seq_file *m)
> +{
> +	struct device_node *root;
> +	const char *model = "";
> +
> +	root = of_find_node_by_path("/");
> +	if (root)
> +		model = of_get_property(root, "model", NULL);
> +	seq_printf(m, "machine\t\t: PowerNV %s\n", model);
> +	of_node_put(root);
> +}
> +
> +static void pnv_restart(char *cmd)
> +{
> +	for (;;);
> +}
> +
> +static void pnv_power_off(void)
> +{
> +	for (;;);
> +}
> +
> +static void pnv_halt(void)
> +{
> +	for (;;);
> +}
> +
> +static unsigned long __init pnv_get_boot_time(void)
> +{
> +	return 0;
> +}
> +
> +static void pnv_get_rtc_time(struct rtc_time *rtc_tm)
> +{
> +}
> +
> +static int pnv_set_rtc_time(struct rtc_time *tm)
> +{
> +	return 0;
> +}
> +
> +static void pnv_progress(char *s, unsigned short hex)
> +{
> +}
> +
> +#ifdef CONFIG_KEXEC
> +static void pnv_kexec_cpu_down(int crash_shutdown, int secondary)
> +{
> +	xics_kexec_teardown_cpu(secondary);
> +}
> +#endif /* CONFIG_KEXEC */
> +
> +static int __init pnv_probe(void)
> +{
> +	unsigned long root = of_get_flat_dt_root();
> +
> +	if (!of_flat_dt_is_compatible(root, "ibm,powernv"))
> +		return 0;
> +
> +	hpte_init_native();
> +
> +	pr_debug("PowerNV detected !\n");
> +
> +	return 1;
> +}
> +
> +define_machine(powernv) {
> +	.name			= "PowerNV",
> +	.probe			= pnv_probe,
> +	.setup_arch		= pnv_setup_arch,
> +	.init_early		= pnv_init_early,
> +	.init_IRQ		= pnv_init_IRQ,
> +	.show_cpuinfo		= pnv_show_cpuinfo,
> +	.restart		= pnv_restart,
> +	.power_off		= pnv_power_off,
> +	.halt			= pnv_halt,
> +	.get_boot_time		= pnv_get_boot_time,
> +	.get_rtc_time		= pnv_get_rtc_time,
> +	.set_rtc_time		= pnv_set_rtc_time,
> +	.progress		= pnv_progress,
> +	.power_save             = power7_idle,
> +	.calibrate_decr		= generic_calibrate_decr,
> +#ifdef CONFIG_KEXEC
> +	.kexec_cpu_down		= pnv_kexec_cpu_down,
> +#endif
> +};
> diff --git a/arch/powerpc/platforms/powernv/smp.c b/arch/powerpc/platforms/po
wernv/smp.c
> new file mode 100644
> index 0000000..36c7151
> --- /dev/null
> +++ b/arch/powerpc/platforms/powernv/smp.c
> @@ -0,0 +1,83 @@
> +/*
> + * SMP support for PowerNV machines.
> + *
> + * Copyright 2011 IBM Corp.
> + *
> + * 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.
> + */
> +
> +#include <linux/kernel.h>
> +#include <linux/module.h>
> +#include <linux/sched.h>
> +#include <linux/smp.h>
> +#include <linux/interrupt.h>
> +#include <linux/delay.h>
> +#include <linux/init.h>
> +#include <linux/spinlock.h>
> +#include <linux/cpu.h>
> +
> +#include <asm/irq.h>
> +#include <asm/smp.h>
> +#include <asm/paca.h>
> +#include <asm/machdep.h>
> +#include <asm/cputable.h>
> +#include <asm/firmware.h>
> +#include <asm/system.h>
> +#include <asm/rtas.h>
> +#include <asm/vdso_datapage.h>
> +#include <asm/cputhreads.h>
> +#include <asm/xics.h>
> +
> +#include "powernv.h"
> +
> +static void __devinit pnv_smp_setup_cpu(int cpu)
> +{
> +	if (cpu != boot_cpuid)
> +		xics_setup_cpu();
> +}
> +
> +static int pnv_smp_cpu_bootable(unsigned int nr)
> +{
> +	/* Special case - we inhibit secondary thread startup
> +	 * during boot if the user requests it.
> +	 */
> +	if (system_state < SYSTEM_RUNNING && cpu_has_feature(CPU_FTR_SMT)) {
> +		if (!smt_enabled_at_boot && cpu_thread_in_core(nr) != 0)
> +			return 0;
> +		if (smt_enabled_at_boot
> +		    && cpu_thread_in_core(nr) >= smt_enabled_at_boot)
> +			return 0;
> +	}
> +
> +	return 1;
> +}
> +
> +static struct smp_ops_t pnv_smp_ops = {
> +	.message_pass	= smp_muxed_ipi_message_pass,
> +	.cause_ipi	= NULL,	/* Filled at runtime by xics_smp_probe() */
> +	.probe		= xics_smp_probe,
> +	.kick_cpu	= smp_generic_kick_cpu,
> +	.setup_cpu	= pnv_smp_setup_cpu,
> +	.cpu_bootable	= pnv_smp_cpu_bootable,
> +};
> +
> +/* This is called very early during platform setup_arch */
> +void __init pnv_smp_init(void)
> +{
> +	smp_ops = &pnv_smp_ops;
> +
> +	/* XXX We don't yet have a proper entry point from HAL, for
> +	 * now we rely on kexec-style entry from BML
> +	 */
> +
> +#ifdef CONFIG_PPC_RTAS
> +	/* Non-lpar has additional take/give timebase */
> +	if (rtas_token("freeze-time-base") != RTAS_UNKNOWN_SERVICE) {
> +		smp_ops->give_timebase = rtas_give_timebase;
> +		smp_ops->take_timebase = rtas_take_timebase;
> +	}
> +#endif /* CONFIG_PPC_RTAS */
> +}
> -- 
> 1.7.4.1
> 
> _______________________________________________
> Linuxppc-dev mailing list
> Linuxppc-dev@lists.ozlabs.org
> https://lists.ozlabs.org/listinfo/linuxppc-dev
> 

Re: [PATCH 06/21] powerpc: Add skeleton PowerNV platform

[Geert Uytterhoeven]

Hi Mikey,

On Mon, Sep 12, 2011 at 03:17, Michael Neuling <mikey@neuling.org> wrote:
>> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
>
> Giggity-Giggity-by: Michael Neuling <mikey@neuling.org>

Burps?

Gr{oetje,eeting}s,

=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k=
.org

In personal conversations with technical people, I call myself a hacker. Bu=
t
when I'm talking to journalists I just say "programmer" or something like t=
hat.
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 =C2=A0=C2=A0 -- Linus Torvalds

[PATCH 07/21] of: Change logic to overwrite cmd_line with CONFIG_CMDLINE

[Benjamin Herrenschmidt]

We used to overwrite with CONFIG_CMDLINE if we found a chosen
node but failed to get bootargs out of it or they were empty,
unless CONFIG_CMDLINE_FORCE is set.

Instead change that to overwrite if cmd_line is non empty after
the bootargs check. It allows arch code to have other mechanisms
to retrieve the command line prior to parsing the device-tree.

Note: CONFIG_CMDLINE_FORCE case should ideally be handled elsewhere
as it won't work as it-is if the device-tree has no /chosen node

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 drivers/of/fdt.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c
index 65200af..d382163 100644
--- a/drivers/of/fdt.c
+++ b/drivers/of/fdt.c
@@ -683,7 +683,7 @@ int __init early_init_dt_scan_chosen(unsigned long node, const char *uname,
 
 #ifdef CONFIG_CMDLINE
 #ifndef CONFIG_CMDLINE_FORCE
-	if (p == NULL || l == 0 || (l == 1 && (*p) == 0))
+	if (!cmd_line[0])
 #endif
 		strlcpy(data, CONFIG_CMDLINE, COMMAND_LINE_SIZE);
 #endif /* CONFIG_CMDLINE */
-- 
1.7.4.1

[PATCH 08/21] powerpc/powernv: Add CPU hotplug support

[Benjamin Herrenschmidt]

Unplugged CPU go into NAP mode in a loop until woken up

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/Kconfig                 |    2 +-
 arch/powerpc/platforms/powernv/smp.c |   78 +++++++++++++++++++++++++++++++++-
 2 files changed, 78 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 6926b61..85baae5 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -323,7 +323,7 @@ config SWIOTLB
 
 config HOTPLUG_CPU
 	bool "Support for enabling/disabling CPUs"
-	depends on SMP && HOTPLUG && EXPERIMENTAL && (PPC_PSERIES || PPC_PMAC)
+	depends on SMP && HOTPLUG && EXPERIMENTAL && (PPC_PSERIES || PPC_PMAC || PPC_POWERNV)
 	---help---
 	  Say Y here to be able to disable and re-enable individual
 	  CPUs at runtime on SMP machines.
diff --git a/arch/powerpc/platforms/powernv/smp.c b/arch/powerpc/platforms/powernv/smp.c
index 36c7151..4f4ec37 100644
--- a/arch/powerpc/platforms/powernv/smp.c
+++ b/arch/powerpc/platforms/powernv/smp.c
@@ -33,7 +33,14 @@
 
 #include "powernv.h"
 
-static void __devinit pnv_smp_setup_cpu(int cpu)
+#ifdef DEBUG
+#include <asm/udbg.h>
+#define DBG(fmt...) udbg_printf(fmt)
+#else
+#define DBG(fmt...)
+#endif
+
+static void __cpuinit pnv_smp_setup_cpu(int cpu)
 {
 	if (cpu != boot_cpuid)
 		xics_setup_cpu();
@@ -55,6 +62,67 @@ static int pnv_smp_cpu_bootable(unsigned int nr)
 	return 1;
 }
 
+#ifdef CONFIG_HOTPLUG_CPU
+
+static int pnv_smp_cpu_disable(void)
+{
+	int cpu = smp_processor_id();
+
+	/* This is identical to pSeries... might consolidate by
+	 * moving migrate_irqs_away to a ppc_md with default to
+	 * the generic fixup_irqs. --BenH.
+	 */
+	set_cpu_online(cpu, false);
+	vdso_data->processorCount--;
+	if (cpu == boot_cpuid)
+		boot_cpuid = cpumask_any(cpu_online_mask);
+	xics_migrate_irqs_away();
+	return 0;
+}
+
+static void pnv_smp_cpu_kill_self(void)
+{
+	unsigned int cpu;
+
+	/* If powersave_nap is enabled, use NAP mode, else just
+	 * spin aimlessly
+	 */
+	if (!powersave_nap) {
+		generic_mach_cpu_die();
+		return;
+	}
+
+	/* Standard hot unplug procedure */
+	local_irq_disable();
+	idle_task_exit();
+	current->active_mm = NULL; /* for sanity */
+	cpu = smp_processor_id();
+	DBG("CPU%d offline\n", cpu);
+	generic_set_cpu_dead(cpu);
+	smp_wmb();
+
+	/* We don't want to take decrementer interrupts while we are offline,
+	 * so clear LPCR:PECE1. We keep PECE2 enabled.
+	 */
+	mtspr(SPRN_LPCR, mfspr(SPRN_LPCR) & ~(u64)LPCR_PECE1);
+	while (!generic_check_cpu_restart(cpu)) {
+		power7_idle();
+		if (!generic_check_cpu_restart(cpu)) {
+			DBG("CPU%d Unexpected exit while offline !\n", cpu);
+			/* We may be getting an IPI, so we re-enable
+			 * interrupts to process it, it will be ignored
+			 * since we aren't online (hopefully)
+			 */
+			local_irq_enable();
+			local_irq_disable();
+		}
+	}
+	mtspr(SPRN_LPCR, mfspr(SPRN_LPCR) | LPCR_PECE1);
+	DBG("CPU%d coming online...\n", cpu);
+}
+
+#endif /* CONFIG_HOTPLUG_CPU */
+
 static struct smp_ops_t pnv_smp_ops = {
 	.message_pass	= smp_muxed_ipi_message_pass,
 	.cause_ipi	= NULL,	/* Filled at runtime by xics_smp_probe() */
@@ -62,6 +130,10 @@ static struct smp_ops_t pnv_smp_ops = {
 	.kick_cpu	= smp_generic_kick_cpu,
 	.setup_cpu	= pnv_smp_setup_cpu,
 	.cpu_bootable	= pnv_smp_cpu_bootable,
+#ifdef CONFIG_HOTPLUG_CPU
+	.cpu_disable	= pnv_smp_cpu_disable,
+	.cpu_die	= generic_cpu_die,
+#endif /* CONFIG_HOTPLUG_CPU */
 };
 
 /* This is called very early during platform setup_arch */
@@ -80,4 +152,8 @@ void __init pnv_smp_init(void)
 		smp_ops->take_timebase = rtas_take_timebase;
 	}
 #endif /* CONFIG_PPC_RTAS */
+
+#ifdef CONFIG_HOTPLUG_CPU
+	ppc_md.cpu_die	= pnv_smp_cpu_kill_self;
+#endif
 }
-- 
1.7.4.1

[PATCH 09/21] powerpc/powernv: Add OPAL takeover from PowerVM

[Benjamin Herrenschmidt]

On machines supporting the OPAL firmware version 1, the system
is initially booted under pHyp. We then use a special hypercall
to verify if OPAL is available and if it is, we then trigger
a "takeover" which disables pHyp and loads the OPAL runtime
firmware, giving control to the kernel in hypervisor mode.

This patch add the necessary code to detect that the OPAL takeover
capability is present when running under PowerVM (aka pHyp) and
perform said takeover to get hypervisor control of the processor.

To perform the takeover, we must first use RTAS (within Open
Firmware runtime environment) to start all processors & threads,
in order to give control to OPAL on all of them. We then call
the takeover hypercall on everybody, OPAL will re-enter the kernel
main entry point passing it a flat device-tree.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/include/asm/opal.h                |   50 +++++
 arch/powerpc/kernel/head_64.S                  |    4 +
 arch/powerpc/kernel/prom_init.c                |  239 ++++++++++++++++++++++--
 arch/powerpc/kernel/prom_init_check.sh         |    3 +-
 arch/powerpc/platforms/powernv/Makefile        |    2 +-
 arch/powerpc/platforms/powernv/opal-takeover.S |  140 ++++++++++++++
 6 files changed, 419 insertions(+), 19 deletions(-)
 create mode 100644 arch/powerpc/include/asm/opal.h
 create mode 100644 arch/powerpc/platforms/powernv/opal-takeover.S

diff --git a/arch/powerpc/include/asm/opal.h b/arch/powerpc/include/asm/opal.h
new file mode 100644
index 0000000..ecdb283
--- /dev/null
+++ b/arch/powerpc/include/asm/opal.h
@@ -0,0 +1,50 @@
+/*
+ * PowerNV OPAL definitions.
+ *
+ * Copyright 2011 IBM Corp.
+ *
+ * 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.
+ */
+
+#ifndef __OPAL_H
+#define __OPAL_H
+
+/****** Takeover interface ********/
+
+/* PAPR H-Call used to querty the HAL existence and/or instanciate
+ * it from within pHyp (tech preview only).
+ *
+ * This is exclusively used in prom_init.c
+ */
+
+#ifndef __ASSEMBLY__
+
+struct opal_takeover_args {
+	u64	k_image;		/* r4 */
+	u64	k_size;			/* r5 */
+	u64	k_entry;		/* r6 */
+	u64	k_entry2;		/* r7 */
+	u64	hal_addr;		/* r8 */
+	u64	rd_image;		/* r9 */
+	u64	rd_size;		/* r10 */
+	u64	rd_loc;			/* r11 */
+};
+
+extern long opal_query_takeover(u64 *hal_size, u64 *hal_align);
+
+extern long opal_do_takeover(struct opal_takeover_args *args);
+
+extern int opal_enter_rtas(struct rtas_args *args,
+			   unsigned long data,
+			   unsigned long entry);
+
+
+#endif /* __ASSEMBLY__ */
+
+/****** OPAL APIs ******/
+
+
+#endif /* __OPAL_H */
diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index e708abe..dea8191 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -51,6 +51,10 @@
  *  For pSeries or server processors:
  *   1. The MMU is off & open firmware is running in real mode.
  *   2. The kernel is entered at __start
+ * -or- For OPAL entry:
+ *   1. The MMU is off, processor in HV mode, primary CPU enters at 0
+ *      with device-tree in gpr3
+ *   2. Secondary processors enter at 0x60 with PIR in gpr3
  *
  *  For iSeries:
  *   1. The MMU is on (as it always is for iSeries)
diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c
index a909f4e..9369287 100644
--- a/arch/powerpc/kernel/prom_init.c
+++ b/arch/powerpc/kernel/prom_init.c
@@ -43,6 +43,7 @@
 #include <asm/btext.h>
 #include <asm/sections.h>
 #include <asm/machdep.h>
+#include <asm/opal.h>
 
 #include <linux/linux_logo.h>
 
@@ -185,6 +186,7 @@ static unsigned long __initdata prom_tce_alloc_end;
 #define PLATFORM_LPAR		0x0001
 #define PLATFORM_POWERMAC	0x0400
 #define PLATFORM_GENERIC	0x0500
+#define PLATFORM_OPAL		0x0600
 
 static int __initdata of_platform;
 
@@ -644,7 +646,7 @@ static void __init early_cmdline_parse(void)
 	}
 }
 
-#ifdef CONFIG_PPC_PSERIES
+#if defined(CONFIG_PPC_PSERIES) || defined(CONFIG_PPC_POWERNV)
 /*
  * There are two methods for telling firmware what our capabilities are.
  * Newer machines have an "ibm,client-architecture-support" method on the
@@ -1274,6 +1276,195 @@ static void __init prom_init_mem(void)
 	prom_printf("  ram_top      : %x\n", RELOC(ram_top));
 }
 
+static void __init prom_close_stdin(void)
+{
+	struct prom_t *_prom = &RELOC(prom);
+	ihandle val;
+
+	if (prom_getprop(_prom->chosen, "stdin", &val, sizeof(val)) > 0)
+		call_prom("close", 1, 0, val);
+}
+
+#ifdef CONFIG_PPC_POWERNV
+
+static u64 __initdata prom_opal_size;
+static u64 __initdata prom_opal_align;
+static int __initdata prom_rtas_start_cpu;
+static u64 __initdata prom_rtas_data;
+static u64 __initdata prom_rtas_entry;
+
+/* XXX Don't change this structure without updating opal-takeover.S */
+static struct opal_secondary_data {
+	s64				ack;	/*  0 */
+	u64				go;	/*  8 */
+	struct opal_takeover_args	args;	/* 16 */
+} opal_secondary_data;
+
+extern char opal_secondary_entry;
+
+static void prom_query_opal(void)
+{
+	long rc;
+
+	prom_printf("Querying for OPAL presence... ");
+	rc = opal_query_takeover(&RELOC(prom_opal_size),
+				 &RELOC(prom_opal_align));
+	prom_debug("(rc = %ld) ", rc);
+	if (rc != 0) {
+		prom_printf("not there.\n");
+		return;
+	}
+	RELOC(of_platform) = PLATFORM_OPAL;
+	prom_printf(" there !\n");
+	prom_debug("  opal_size  = 0x%lx\n", RELOC(prom_opal_size));
+	prom_debug("  opal_align = 0x%lx\n", RELOC(prom_opal_align));
+	if (RELOC(prom_opal_align) < 0x10000)
+		RELOC(prom_opal_align) = 0x10000;
+}
+
+static int prom_rtas_call(int token, int nargs, int nret, int *outputs, ...)
+{
+	struct rtas_args rtas_args;
+	va_list list;
+	int i;
+
+	rtas_args.token = token;
+	rtas_args.nargs = nargs;
+	rtas_args.nret  = nret;
+	rtas_args.rets  = (rtas_arg_t *)&(rtas_args.args[nargs]);
+	va_start(list, outputs);
+	for (i = 0; i < nargs; ++i)
+		rtas_args.args[i] = va_arg(list, rtas_arg_t);
+	va_end(list);
+
+	for (i = 0; i < nret; ++i)
+		rtas_args.rets[i] = 0;
+
+	opal_enter_rtas(&rtas_args, RELOC(prom_rtas_data),
+			RELOC(prom_rtas_entry));
+
+	if (nret > 1 && outputs != NULL)
+		for (i = 0; i < nret-1; ++i)
+			outputs[i] = rtas_args.rets[i+1];
+	return (nret > 0)? rtas_args.rets[0]: 0;
+}
+
+static void __init prom_opal_hold_cpus(void)
+{
+	int i, cnt, cpu, rc;
+	long j;
+	phandle node;
+	char type[64];
+	u32 servers[8];
+	struct prom_t *_prom = &RELOC(prom);
+	void *entry = (unsigned long *)&RELOC(opal_secondary_entry);
+	struct opal_secondary_data *data = &RELOC(opal_secondary_data);
+
+	prom_debug("prom_opal_hold_cpus: start...\n");
+	prom_debug("    - entry       = 0x%x\n", entry);
+	prom_debug("    - data        = 0x%x\n", data);
+
+	data->ack = -1;
+	data->go = 0;
+
+	/* look for cpus */
+	for (node = 0; prom_next_node(&node); ) {
+		type[0] = 0;
+		prom_getprop(node, "device_type", type, sizeof(type));
+		if (strcmp(type, RELOC("cpu")) != 0)
+			continue;
+
+		/* Skip non-configured cpus. */
+		if (prom_getprop(node, "status", type, sizeof(type)) > 0)
+			if (strcmp(type, RELOC("okay")) != 0)
+				continue;
+
+		cnt = prom_getprop(node, "ibm,ppc-interrupt-server#s", servers,
+			     sizeof(servers));
+		if (cnt == PROM_ERROR)
+			break;
+		cnt >>= 2;
+		for (i = 0; i < cnt; i++) {
+			cpu = servers[i];
+			prom_debug("CPU %d ... ", cpu);
+			if (cpu == _prom->cpu) {
+				prom_debug("booted !\n");
+				continue;
+			}
+			prom_debug("starting ... ");
+
+			/* Init the acknowledge var which will be reset by
+			 * the secondary cpu when it awakens from its OF
+			 * spinloop.
+			 */
+			data->ack = -1;
+			rc = prom_rtas_call(RELOC(prom_rtas_start_cpu), 3, 1,
+					    NULL, cpu, entry, data);
+			prom_debug("rtas rc=%d ...", rc);
+
+			for (j = 0; j < 100000000 && data->ack == -1; j++) {
+				HMT_low();
+				mb();
+			}
+			HMT_medium();
+			if (data->ack != -1)
+				prom_debug("done, PIR=0x%x\n", data->ack);
+			else
+				prom_debug("timeout !\n");
+		}
+	}
+	prom_debug("prom_opal_hold_cpus: end...\n");
+}
+
+static void prom_opal_takeover(void)
+{
+	struct opal_secondary_data *data = &RELOC(opal_secondary_data);
+	struct opal_takeover_args *args = &data->args;
+	u64 align = RELOC(prom_opal_align);
+	u64 top_addr, opal_addr;
+
+	args->k_image	= (u64)RELOC(_stext);
+	args->k_size	= _end - _stext;
+	args->k_entry	= 0;
+	args->k_entry2	= 0x60;
+
+	top_addr = _ALIGN_UP(args->k_size, align);
+
+	if (RELOC(prom_initrd_start) != 0) {
+		args->rd_image = RELOC(prom_initrd_start);
+		args->rd_size = RELOC(prom_initrd_end) - args->rd_image;
+		args->rd_loc = top_addr;
+		top_addr = _ALIGN_UP(args->rd_loc + args->rd_size, align);
+	}
+
+	/* Pickup an address for the HAL. We want to go really high
+	 * up to avoid problem with future kexecs. On the other hand
+	 * we don't want to be all over the TCEs on P5IOC2 machines
+	 * which are going to be up there too. We assume the machine
+	 * has plenty of memory, and we ask for the HAL for now to
+	 * be just below the 1G point, or above the initrd
+	 */
+	opal_addr = _ALIGN_DOWN(0x40000000 - RELOC(prom_opal_size), align);
+	if (opal_addr < top_addr)
+		opal_addr = top_addr;
+	args->hal_addr = opal_addr;
+
+	prom_debug("  k_image    = 0x%lx\n", args->k_image);
+	prom_debug("  k_size     = 0x%lx\n", args->k_size);
+	prom_debug("  k_entry    = 0x%lx\n", args->k_entry);
+	prom_debug("  k_entry2   = 0x%lx\n", args->k_entry2);
+	prom_debug("  hal_addr   = 0x%lx\n", args->hal_addr);
+	prom_debug("  rd_image   = 0x%lx\n", args->rd_image);
+	prom_debug("  rd_size    = 0x%lx\n", args->rd_size);
+	prom_debug("  rd_loc     = 0x%lx\n", args->rd_loc);
+	prom_printf("Performing OPAL takeover,this can take a few minutes..\n");
+	prom_close_stdin();
+	mb();
+	data->go = 1;
+	for (;;)
+		opal_do_takeover(args);
+}
+#endif /* CONFIG_PPC_POWERNV */
 
 /*
  * Allocate room for and instantiate RTAS
@@ -1326,6 +1517,12 @@ static void __init prom_instantiate_rtas(void)
 	prom_setprop(rtas_node, "/rtas", "linux,rtas-entry",
 		     &entry, sizeof(entry));
 
+#ifdef CONFIG_PPC_POWERNV
+	/* PowerVN takeover hack */
+	RELOC(prom_rtas_data) = base;
+	RELOC(prom_rtas_entry) = entry;
+	prom_getprop(rtas_node, "start-cpu", &RELOC(prom_rtas_start_cpu), 4);
+#endif
 	prom_debug("rtas base     = 0x%x\n", base);
 	prom_debug("rtas entry    = 0x%x\n", entry);
 	prom_debug("rtas size     = 0x%x\n", (long)size);
@@ -1543,7 +1740,7 @@ static void __init prom_hold_cpus(void)
 		*acknowledge = (unsigned long)-1;
 
 		if (reg != _prom->cpu) {
-			/* Primary Thread of non-boot cpu */
+			/* Primary Thread of non-boot cpu or any thread */
 			prom_printf("starting cpu hw idx %lu... ", reg);
 			call_prom("start-cpu", 3, 0, node,
 				  secondary_hold, reg);
@@ -1652,15 +1849,6 @@ static void __init prom_init_stdout(void)
 		prom_setprop(val, path, "linux,boot-display", NULL, 0);
 }
 
-static void __init prom_close_stdin(void)
-{
-	struct prom_t *_prom = &RELOC(prom);
-	ihandle val;
-
-	if (prom_getprop(_prom->chosen, "stdin", &val, sizeof(val)) > 0)
-		call_prom("close", 1, 0, val);
-}
-
 static int __init prom_find_machine_type(void)
 {
 	struct prom_t *_prom = &RELOC(prom);
@@ -2504,6 +2692,7 @@ static void __init prom_check_initrd(unsigned long r3, unsigned long r4)
 #endif /* CONFIG_BLK_DEV_INITRD */
 }
 
+
 /*
  * We enter here early on, when the Open Firmware prom is still
  * handling exceptions and the MMU hash table for us.
@@ -2565,7 +2754,7 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,
 	 */
 	prom_check_initrd(r3, r4);
 
-#ifdef CONFIG_PPC_PSERIES
+#if defined(CONFIG_PPC_PSERIES) || defined(CONFIG_PPC_POWERNV)
 	/*
 	 * On pSeries, inform the firmware about our capabilities
 	 */
@@ -2611,14 +2800,30 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,
 #endif
 
 	/*
-	 * On non-powermacs, try to instantiate RTAS and puts all CPUs
-	 * in spin-loops. PowerMacs don't have a working RTAS and use
-	 * a different way to spin CPUs
+	 * On non-powermacs, try to instantiate RTAS. PowerMacs don't
+	 * have a usable RTAS implementation.
 	 */
-	if (RELOC(of_platform) != PLATFORM_POWERMAC) {
+	if (RELOC(of_platform) != PLATFORM_POWERMAC)
 		prom_instantiate_rtas();
-		prom_hold_cpus();
+
+#ifdef CONFIG_PPC_POWERNV
+	/* Detect HAL and try instanciating it & doing takeover */
+	if (RELOC(of_platform) == PLATFORM_PSERIES_LPAR) {
+		prom_query_opal();
+		if (RELOC(of_platform) == PLATFORM_OPAL) {
+			prom_opal_hold_cpus();
+			prom_opal_takeover();
+		}
 	}
+#endif
+
+	/*
+	 * On non-powermacs, put all CPUs in spin-loops.
+	 *
+	 * PowerMacs use a different mechanism to spin CPUs
+	 */
+	if (RELOC(of_platform) != PLATFORM_POWERMAC)
+		prom_hold_cpus();
 
 	/*
 	 * Fill in some infos for use by the kernel later on
diff --git a/arch/powerpc/kernel/prom_init_check.sh b/arch/powerpc/kernel/prom_init_check.sh
index 9f82f49..20af6aa 100644
--- a/arch/powerpc/kernel/prom_init_check.sh
+++ b/arch/powerpc/kernel/prom_init_check.sh
@@ -20,7 +20,8 @@ WHITELIST="add_reloc_offset __bss_start __bss_stop copy_and_flush
 _end enter_prom memcpy memset reloc_offset __secondary_hold
 __secondary_hold_acknowledge __secondary_hold_spinloop __start
 strcmp strcpy strlcpy strlen strncmp strstr logo_linux_clut224
-reloc_got2 kernstart_addr memstart_addr linux_banner"
+reloc_got2 kernstart_addr memstart_addr linux_banner _stext
+opal_query_takeover opal_do_takeover opal_enter_rtas opal_secondary_entry"
 
 NM="$1"
 OBJ="$2"
diff --git a/arch/powerpc/platforms/powernv/Makefile b/arch/powerpc/platforms/powernv/Makefile
index 1c43250..4971330 100644
--- a/arch/powerpc/platforms/powernv/Makefile
+++ b/arch/powerpc/platforms/powernv/Makefile
@@ -1,2 +1,2 @@
-obj-y			+= setup.o
+obj-y			+= setup.o opal-takeover.o
 obj-$(CONFIG_SMP)	+= smp.o
diff --git a/arch/powerpc/platforms/powernv/opal-takeover.S b/arch/powerpc/platforms/powernv/opal-takeover.S
new file mode 100644
index 0000000..77b48b2
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/opal-takeover.S
@@ -0,0 +1,140 @@
+/*
+ * PowerNV OPAL takeover assembly code, for use by prom_init.c
+ *
+ * Copyright 2011 IBM Corp.
+ *
+ * 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.
+ */
+
+#include <asm/ppc_asm.h>
+#include <asm/hvcall.h>
+#include <asm/asm-offsets.h>
+#include <asm/opal.h>
+
+#define STK_PARAM(i)	(48 + ((i)-3)*8)
+
+#define H_HAL_TAKEOVER			0x5124
+#define H_HAL_TAKEOVER_QUERY_MAGIC	-1
+
+	.text
+_GLOBAL(opal_query_takeover)
+	mfcr	r0
+	stw	r0,8(r1)
+	std	r3,STK_PARAM(r3)(r1)
+	std	r4,STK_PARAM(r4)(r1)
+	li	r3,H_HAL_TAKEOVER
+	li	r4,H_HAL_TAKEOVER_QUERY_MAGIC
+	HVSC
+	ld	r10,STK_PARAM(r3)(r1)
+	std	r4,0(r10)
+	ld	r10,STK_PARAM(r4)(r1)
+	std	r5,0(r10)
+	lwz	r0,8(r1)
+	mtcrf	0xff,r0
+	blr
+
+_GLOBAL(opal_do_takeover)
+	mfcr	r0
+	stw	r0,8(r1)
+	mflr	r0
+	std	r0,16(r1)
+	bl	__opal_do_takeover
+	ld	r0,16(r1)
+	mtlr	r0
+	lwz	r0,8(r1)
+	mtcrf	0xff,r0
+	blr
+
+__opal_do_takeover:
+	ld	r4,0(r3)
+	ld	r5,0x8(r3)
+	ld	r6,0x10(r3)
+	ld	r7,0x18(r3)
+	ld	r8,0x20(r3)
+	ld	r9,0x28(r3)
+	ld	r10,0x30(r3)
+	ld	r11,0x38(r3)
+	li	r3,H_HAL_TAKEOVER
+	HVSC
+	blr
+
+	.globl opal_secondary_entry
+opal_secondary_entry:
+	mr	r31,r3
+	mfmsr	r11
+	li	r12,(MSR_SF | MSR_ISF)@highest
+	sldi	r12,r12,48
+	or	r11,r11,r12
+	mtmsrd	r11
+	isync
+	mfspr	r4,SPRN_PIR
+	std	r4,0(r3)
+1:	HMT_LOW
+	ld	r4,8(r3)
+	cmpli	cr0,r4,0
+	beq	1b
+	HMT_MEDIUM
+1:	addi	r3,r31,16
+	bl	__opal_do_takeover
+	b	1b
+
+_GLOBAL(opal_enter_rtas)
+	mflr	r0
+	std	r0,16(r1)
+        stdu	r1,-PROM_FRAME_SIZE(r1)	/* Save SP and create stack space */
+
+	/* Because PROM is running in 32b mode, it clobbers the high order half
+	 * of all registers that it saves.  We therefore save those registers
+	 * PROM might touch to the stack.  (r0, r3-r13 are caller saved)
+	*/
+	SAVE_GPR(2, r1)
+	SAVE_GPR(13, r1)
+	SAVE_8GPRS(14, r1)
+	SAVE_10GPRS(22, r1)
+	mfcr	r10
+	mfmsr	r11
+	std	r10,_CCR(r1)
+	std	r11,_MSR(r1)
+
+	/* Get the PROM entrypoint */
+	mtlr	r5
+
+	/* Switch MSR to 32 bits mode
+	 */
+        li      r12,1
+        rldicr  r12,r12,MSR_SF_LG,(63-MSR_SF_LG)
+        andc    r11,r11,r12
+        li      r12,1
+        rldicr  r12,r12,MSR_ISF_LG,(63-MSR_ISF_LG)
+        andc    r11,r11,r12
+        mtmsrd  r11
+        isync
+
+	/* Enter RTAS here... */
+	blrl
+
+	/* Just make sure that r1 top 32 bits didn't get
+	 * corrupt by OF
+	 */
+	rldicl	r1,r1,0,32
+
+	/* Restore the MSR (back to 64 bits) */
+	ld	r0,_MSR(r1)
+	MTMSRD(r0)
+        isync
+
+	/* Restore other registers */
+	REST_GPR(2, r1)
+	REST_GPR(13, r1)
+	REST_8GPRS(14, r1)
+	REST_10GPRS(22, r1)
+	ld	r4,_CCR(r1)
+	mtcr	r4
+
+        addi	r1,r1,PROM_FRAME_SIZE
+	ld	r0,16(r1)
+	mtlr    r0
+	blr
-- 
1.7.4.1

[PATCH 10/21] powerpc/powernv: Get kernel command line accross OPAL takeover

[Benjamin Herrenschmidt]

We stash it in boot_command_line which isn't in BSS and so won't
be overwritten. We then use that as a default cmd_line before
we walk the device-tree.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/prom.c             |    7 +++++++
 arch/powerpc/kernel/prom_init.c        |    4 ++++
 arch/powerpc/kernel/prom_init_check.sh |    3 ++-
 3 files changed, 13 insertions(+), 1 deletions(-)

diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
index 174e1e9..7b90c56 100644
--- a/arch/powerpc/kernel/prom.c
+++ b/arch/powerpc/kernel/prom.c
@@ -712,6 +712,13 @@ void __init early_init_devtree(void *params)
 	of_scan_flat_dt(early_init_dt_scan_phyp_dump, NULL);
 #endif
 
+	/* Pre-initialize the cmd_line with the content of boot_commmand_line,
+	 * which will be empty except when the content of the variable has
+	 * been overriden by a bootloading mechanism. This happens typically
+	 * with HAL takeover
+	 */
+	strlcpy(cmd_line, boot_command_line, COMMAND_LINE_SIZE);
+
 	/* Retrieve various informations from the /chosen node of the
 	 * device-tree, including the platform type, initrd location and
 	 * size, TCE reserve, and more ...
diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c
index 9369287..e3f3904 100644
--- a/arch/powerpc/kernel/prom_init.c
+++ b/arch/powerpc/kernel/prom_init.c
@@ -1449,6 +1449,10 @@ static void prom_opal_takeover(void)
 		opal_addr = top_addr;
 	args->hal_addr = opal_addr;
 
+	/* Copy the command line to the kernel image */
+	strlcpy(RELOC(boot_command_line), RELOC(prom_cmd_line),
+		COMMAND_LINE_SIZE);
+
 	prom_debug("  k_image    = 0x%lx\n", args->k_image);
 	prom_debug("  k_size     = 0x%lx\n", args->k_size);
 	prom_debug("  k_entry    = 0x%lx\n", args->k_entry);
diff --git a/arch/powerpc/kernel/prom_init_check.sh b/arch/powerpc/kernel/prom_init_check.sh
index 20af6aa..70f4286 100644
--- a/arch/powerpc/kernel/prom_init_check.sh
+++ b/arch/powerpc/kernel/prom_init_check.sh
@@ -21,7 +21,8 @@ _end enter_prom memcpy memset reloc_offset __secondary_hold
 __secondary_hold_acknowledge __secondary_hold_spinloop __start
 strcmp strcpy strlcpy strlen strncmp strstr logo_linux_clut224
 reloc_got2 kernstart_addr memstart_addr linux_banner _stext
-opal_query_takeover opal_do_takeover opal_enter_rtas opal_secondary_entry"
+opal_query_takeover opal_do_takeover opal_enter_rtas opal_secondary_entry
+boot_command_line"
 
 NM="$1"
 OBJ="$2"
-- 
1.7.4.1

[PATCH 11/21] powerpc/powernv: Basic support for OPAL

[Benjamin Herrenschmidt]

Add definition of OPAL interfaces along with  the wrappers to call
into OPAL runtime and the early device-tree parsing hook to locate
the OPAL runtime firmware.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/include/asm/firmware.h            |   10 +
 arch/powerpc/include/asm/opal.h                |  382 +++++++++++++++++++++++-
 arch/powerpc/kernel/prom.c                     |    7 +
 arch/powerpc/platforms/powernv/Kconfig         |    8 +-
 arch/powerpc/platforms/powernv/Makefile        |    2 +-
 arch/powerpc/platforms/powernv/opal-wrappers.S |  101 +++++++
 arch/powerpc/platforms/powernv/opal.c          |  154 ++++++++++
 arch/powerpc/platforms/powernv/setup.c         |    6 +
 arch/powerpc/platforms/powernv/smp.c           |   25 ++-
 9 files changed, 690 insertions(+), 5 deletions(-)
 create mode 100644 arch/powerpc/platforms/powernv/opal-wrappers.S
 create mode 100644 arch/powerpc/platforms/powernv/opal.c

diff --git a/arch/powerpc/include/asm/firmware.h b/arch/powerpc/include/asm/firmware.h
index 3a6c586..14db29b 100644
--- a/arch/powerpc/include/asm/firmware.h
+++ b/arch/powerpc/include/asm/firmware.h
@@ -48,6 +48,8 @@
 #define FW_FEATURE_CMO		ASM_CONST(0x0000000002000000)
 #define FW_FEATURE_VPHN		ASM_CONST(0x0000000004000000)
 #define FW_FEATURE_XCMO		ASM_CONST(0x0000000008000000)
+#define FW_FEATURE_OPAL		ASM_CONST(0x0000000010000000)
+#define FW_FEATURE_OPALv2	ASM_CONST(0x0000000020000000)
 
 #ifndef __ASSEMBLY__
 
@@ -65,6 +67,8 @@ enum {
 	FW_FEATURE_PSERIES_ALWAYS = 0,
 	FW_FEATURE_ISERIES_POSSIBLE = FW_FEATURE_ISERIES | FW_FEATURE_LPAR,
 	FW_FEATURE_ISERIES_ALWAYS = FW_FEATURE_ISERIES | FW_FEATURE_LPAR,
+	FW_FEATURE_POWERNV_POSSIBLE = FW_FEATURE_OPAL | FW_FEATURE_OPALv2,
+	FW_FEATURE_POWERNV_ALWAYS = 0,
 	FW_FEATURE_PS3_POSSIBLE = FW_FEATURE_LPAR | FW_FEATURE_PS3_LV1,
 	FW_FEATURE_PS3_ALWAYS = FW_FEATURE_LPAR | FW_FEATURE_PS3_LV1,
 	FW_FEATURE_CELLEB_POSSIBLE = FW_FEATURE_LPAR | FW_FEATURE_BEAT,
@@ -78,6 +82,9 @@ enum {
 #ifdef CONFIG_PPC_ISERIES
 		FW_FEATURE_ISERIES_POSSIBLE |
 #endif
+#ifdef CONFIG_PPC_POWERNV
+		FW_FEATURE_POWERNV_POSSIBLE |
+#endif
 #ifdef CONFIG_PPC_PS3
 		FW_FEATURE_PS3_POSSIBLE |
 #endif
@@ -95,6 +102,9 @@ enum {
 #ifdef CONFIG_PPC_ISERIES
 		FW_FEATURE_ISERIES_ALWAYS &
 #endif
+#ifdef CONFIG_PPC_POWERNV
+		FW_FEATURE_POWERNV_ALWAYS &
+#endif
 #ifdef CONFIG_PPC_PS3
 		FW_FEATURE_PS3_ALWAYS &
 #endif
diff --git a/arch/powerpc/include/asm/opal.h b/arch/powerpc/include/asm/opal.h
index ecdb283..53cda41 100644
--- a/arch/powerpc/include/asm/opal.h
+++ b/arch/powerpc/include/asm/opal.h
@@ -37,14 +37,394 @@ extern long opal_query_takeover(u64 *hal_size, u64 *hal_align);
 
 extern long opal_do_takeover(struct opal_takeover_args *args);
 
+struct rtas_args;
 extern int opal_enter_rtas(struct rtas_args *args,
 			   unsigned long data,
 			   unsigned long entry);
 
-
 #endif /* __ASSEMBLY__ */
 
 /****** OPAL APIs ******/
 
+/* Return codes */
+#define OPAL_SUCCESS 		0
+#define OPAL_PARAMETER		-1
+#define OPAL_BUSY		-2
+#define OPAL_PARTIAL		-3
+#define OPAL_CONSTRAINED	-4
+#define OPAL_CLOSED		-5
+#define OPAL_HARDWARE		-6
+#define OPAL_UNSUPPORTED	-7
+#define OPAL_PERMISSION		-8
+#define OPAL_NO_MEM		-9
+#define OPAL_RESOURCE		-10
+#define OPAL_INTERNAL_ERROR	-11
+#define OPAL_BUSY_EVENT		-12
+#define OPAL_HARDWARE_FROZEN	-13
+
+/* API Tokens (in r0) */
+#define OPAL_CONSOLE_WRITE			1
+#define OPAL_CONSOLE_READ			2
+#define OPAL_RTC_READ				3
+#define OPAL_RTC_WRITE				4
+#define OPAL_CEC_POWER_DOWN			5
+#define OPAL_CEC_REBOOT				6
+#define OPAL_READ_NVRAM				7
+#define OPAL_WRITE_NVRAM			8
+#define OPAL_HANDLE_INTERRUPT			9
+#define OPAL_POLL_EVENTS			10
+#define OPAL_PCI_SET_HUB_TCE_MEMORY		11
+#define OPAL_PCI_SET_PHB_TCE_MEMORY		12
+#define OPAL_PCI_CONFIG_READ_BYTE		13
+#define OPAL_PCI_CONFIG_READ_HALF_WORD  	14
+#define OPAL_PCI_CONFIG_READ_WORD		15
+#define OPAL_PCI_CONFIG_WRITE_BYTE		16
+#define OPAL_PCI_CONFIG_WRITE_HALF_WORD		17
+#define OPAL_PCI_CONFIG_WRITE_WORD		18
+#define OPAL_SET_XIVE				19
+#define OPAL_GET_XIVE				20
+#define OPAL_GET_COMPLETION_TOKEN_STATUS	21 /* obsolete */
+#define OPAL_REGISTER_OPAL_EXCEPTION_HANDLER	22
+#define OPAL_PCI_EEH_FREEZE_STATUS		23
+#define OPAL_PCI_SHPC				24
+#define OPAL_CONSOLE_WRITE_BUFFER_SPACE		25
+#define OPAL_PCI_EEH_FREEZE_CLEAR		26
+#define OPAL_PCI_PHB_MMIO_ENABLE		27
+#define OPAL_PCI_SET_PHB_MEM_WINDOW		28
+#define OPAL_PCI_MAP_PE_MMIO_WINDOW		29
+#define OPAL_PCI_SET_PHB_TABLE_MEMORY		30
+#define OPAL_PCI_SET_PE				31
+#define OPAL_PCI_SET_PELTV			32
+#define OPAL_PCI_SET_MVE			33
+#define OPAL_PCI_SET_MVE_ENABLE			34
+#define OPAL_PCI_GET_XIVE_REISSUE		35
+#define OPAL_PCI_SET_XIVE_REISSUE		36
+#define OPAL_PCI_SET_XIVE_PE			37
+#define OPAL_GET_XIVE_SOURCE			38
+#define OPAL_GET_MSI_32				39
+#define OPAL_GET_MSI_64				40
+#define OPAL_START_CPU				41
+#define OPAL_QUERY_CPU_STATUS			42
+#define OPAL_WRITE_OPPANEL			43
+#define OPAL_PCI_MAP_PE_DMA_WINDOW		44
+#define OPAL_PCI_MAP_PE_DMA_WINDOW_REAL		45
+#define OPAL_PCI_RESET				49
+
+#ifndef __ASSEMBLY__
+
+/* Other enums */
+enum OpalVendorApiTokens {
+	OPAL_START_VENDOR_API_RANGE = 1000, OPAL_END_VENDOR_API_RANGE = 1999
+};
+enum OpalFreezeState {
+	OPAL_EEH_STOPPED_NOT_FROZEN = 0,
+	OPAL_EEH_STOPPED_MMIO_FREEZE = 1,
+	OPAL_EEH_STOPPED_DMA_FREEZE = 2,
+	OPAL_EEH_STOPPED_MMIO_DMA_FREEZE = 3,
+	OPAL_EEH_STOPPED_RESET = 4,
+	OPAL_EEH_STOPPED_TEMP_UNAVAIL = 5,
+	OPAL_EEH_STOPPED_PERM_UNAVAIL = 6
+};
+enum OpalEehFreezeActionToken {
+	OPAL_EEH_ACTION_CLEAR_FREEZE_MMIO = 1,
+	OPAL_EEH_ACTION_CLEAR_FREEZE_DMA = 2,
+	OPAL_EEH_ACTION_CLEAR_FREEZE_ALL = 3
+};
+enum OpalPciStatusToken {
+	OPAL_EEH_PHB_NO_ERROR = 0,
+	OPAL_EEH_PHB_FATAL = 1,
+	OPAL_EEH_PHB_RECOVERABLE = 2,
+	OPAL_EEH_PHB_BUS_ERROR = 3,
+	OPAL_EEH_PCI_NO_DEVSEL = 4,
+	OPAL_EEH_PCI_TA = 5,
+	OPAL_EEH_PCIEX_UR = 6,
+	OPAL_EEH_PCIEX_CA = 7,
+	OPAL_EEH_PCI_MMIO_ERROR = 8,
+	OPAL_EEH_PCI_DMA_ERROR = 9
+};
+enum OpalShpcAction {
+	OPAL_SHPC_GET_LINK_STATE = 0,
+	OPAL_SHPC_GET_SLOT_STATE = 1
+};
+enum OpalShpcLinkState {
+	OPAL_SHPC_LINK_DOWN = 0,
+	OPAL_SHPC_LINK_UP = 1
+};
+enum OpalMmioWindowType {
+	OPAL_M32_WINDOW_TYPE = 1,
+	OPAL_M64_WINDOW_TYPE = 2,
+	OPAL_IO_WINDOW_TYPE = 3
+};
+enum OpalShpcSlotState {
+	OPAL_SHPC_DEV_NOT_PRESENT = 0,
+	OPAL_SHPC_DEV_PRESENT = 1
+};
+enum OpalExceptionHandler {
+	OPAL_MACHINE_CHECK_HANDLER = 1,
+	OPAL_HYPERVISOR_MAINTENANCE_HANDLER = 2,
+	OPAL_SOFTPATCH_HANDLER = 3
+};
+enum OpalPendingState {
+	OPAL_EVENT_OPAL_INTERNAL = 0x1,
+	OPAL_EVENT_NVRAM = 0x2,
+	OPAL_EVENT_RTC = 0x4,
+	OPAL_EVENT_CONSOLE_OUTPUT = 0x8,
+	OPAL_EVENT_CONSOLE_INPUT = 0x10
+};
+
+/* Machine check related definitions */
+enum OpalMCE_Version {
+	OpalMCE_V1 = 1,
+};
+
+enum OpalMCE_Severity {
+	OpalMCE_SEV_NO_ERROR = 0,
+	OpalMCE_SEV_WARNING = 1,
+	OpalMCE_SEV_ERROR_SYNC = 2,
+	OpalMCE_SEV_FATAL = 3,
+};
+
+enum OpalMCE_Disposition {
+	OpalMCE_DISPOSITION_RECOVERED = 0,
+	OpalMCE_DISPOSITION_NOT_RECOVERED = 1,
+};
+
+enum OpalMCE_Initiator {
+	OpalMCE_INITIATOR_UNKNOWN = 0,
+	OpalMCE_INITIATOR_CPU = 1,
+};
+
+enum OpalMCE_ErrorType {
+	OpalMCE_ERROR_TYPE_UNKNOWN = 0,
+	OpalMCE_ERROR_TYPE_UE = 1,
+	OpalMCE_ERROR_TYPE_SLB = 2,
+	OpalMCE_ERROR_TYPE_ERAT = 3,
+	OpalMCE_ERROR_TYPE_TLB = 4,
+};
+
+enum OpalMCE_UeErrorType {
+	OpalMCE_UE_ERROR_INDETERMINATE = 0,
+	OpalMCE_UE_ERROR_IFETCH = 1,
+	OpalMCE_UE_ERROR_PAGE_TABLE_WALK_IFETCH = 2,
+	OpalMCE_UE_ERROR_LOAD_STORE = 3,
+	OpalMCE_UE_ERROR_PAGE_TABLE_WALK_LOAD_STORE = 4,
+};
+
+enum OpalMCE_SlbErrorType {
+	OpalMCE_SLB_ERROR_INDETERMINATE = 0,
+	OpalMCE_SLB_ERROR_PARITY = 1,
+	OpalMCE_SLB_ERROR_MULTIHIT = 2,
+};
+
+enum OpalMCE_EratErrorType {
+	OpalMCE_ERAT_ERROR_INDETERMINATE = 0,
+	OpalMCE_ERAT_ERROR_PARITY = 1,
+	OpalMCE_ERAT_ERROR_MULTIHIT = 2,
+};
+
+enum OpalMCE_TlbErrorType {
+	OpalMCE_TLB_ERROR_INDETERMINATE = 0,
+	OpalMCE_TLB_ERROR_PARITY = 1,
+	OpalMCE_TLB_ERROR_MULTIHIT = 2,
+};
+
+enum OpalThreadStatus {
+	OPAL_THREAD_INACTIVE = 0x0,
+	OPAL_THREAD_STARTED = 0x1
+};
+
+enum OpalPciBusCompare {
+	OpalPciBusAny	= 0,	/* Any bus number match */
+	OpalPciBus3Bits	= 2,	/* Match top 3 bits of bus number */
+	OpalPciBus4Bits	= 3,	/* Match top 4 bits of bus number */
+	OpalPciBus5Bits	= 4,	/* Match top 5 bits of bus number */
+	OpalPciBus6Bits	= 5,	/* Match top 6 bits of bus number */
+	OpalPciBus7Bits	= 6,	/* Match top 7 bits of bus number */
+	OpalPciBusAll	= 7,	/* Match bus number exactly */
+};
+
+enum OpalDeviceCompare {
+	OPAL_IGNORE_RID_DEVICE_NUMBER = 0,
+	OPAL_COMPARE_RID_DEVICE_NUMBER = 1
+};
+
+enum OpalFuncCompare {
+	OPAL_IGNORE_RID_FUNCTION_NUMBER = 0,
+	OPAL_COMPARE_RID_FUNCTION_NUMBER = 1
+};
+
+enum OpalPeAction {
+	OPAL_UNMAP_PE = 0,
+	OPAL_MAP_PE = 1
+};
+
+enum OpalPciResetAndReinitScope { 
+	OPAL_PHB_COMPLETE = 1, OPAL_PCI_LINK = 2, OPAL_PHB_ERROR = 3,
+	OPAL_PCI_HOT_RESET = 4, OPAL_PCI_FUNDAMENTAL_RESET = 5,
+	OPAL_PCI_IODA_RESET = 6,
+};
+
+enum OpalPciResetState { OPAL_DEASSERT_RESET = 0, OPAL_ASSERT_RESET = 1 };
+
+struct opal_machine_check_event {
+	enum OpalMCE_Version	version:8;	/* 0x00 */
+	uint8_t			in_use;		/* 0x01 */
+	enum OpalMCE_Severity	severity:8;	/* 0x02 */
+	enum OpalMCE_Initiator	initiator:8;	/* 0x03 */
+	enum OpalMCE_ErrorType	error_type:8;	/* 0x04 */
+	enum OpalMCE_Disposition disposition:8; /* 0x05 */
+	uint8_t			reserved_1[2];	/* 0x06 */
+	uint64_t		gpr3;		/* 0x08 */
+	uint64_t		srr0;		/* 0x10 */
+	uint64_t		srr1;		/* 0x18 */
+	union {					/* 0x20 */
+		struct {
+			enum OpalMCE_UeErrorType ue_error_type:8;
+			uint8_t		effective_address_provided;
+			uint8_t		physical_address_provided;
+			uint8_t		reserved_1[5];
+			uint64_t	effective_address;
+			uint64_t	physical_address;
+			uint8_t		reserved_2[8];
+		} ue_error;
+
+		struct {
+			enum OpalMCE_SlbErrorType slb_error_type:8;
+			uint8_t		effective_address_provided;
+			uint8_t		reserved_1[6];
+			uint64_t	effective_address;
+			uint8_t		reserved_2[16];
+		} slb_error;
+
+		struct {
+			enum OpalMCE_EratErrorType erat_error_type:8;
+			uint8_t		effective_address_provided;
+			uint8_t		reserved_1[6];
+			uint64_t	effective_address;
+			uint8_t		reserved_2[16];
+		} erat_error;
+
+		struct {
+			enum OpalMCE_TlbErrorType tlb_error_type:8;
+			uint8_t		effective_address_provided;
+			uint8_t		reserved_1[6];
+			uint64_t	effective_address;
+			uint8_t		reserved_2[16];
+		} tlb_error;
+	} u;
+};
+
+typedef struct oppanel_line {
+	/* XXX */
+} oppanel_line_t;
+
+/* API functions */
+int64_t opal_console_write(int64_t term_number, int64_t *length,
+			   const uint8_t *buffer);
+int64_t opal_console_read(int64_t term_number, int64_t *length,
+			  uint8_t *buffer);
+int64_t opal_console_write_buffer_space(int64_t term_number,
+					int64_t *length);
+int64_t opal_rtc_read(uint32_t *year_month_day,
+		      uint64_t *hour_minute_second_millisecond);
+int64_t opal_rtc_write(uint32_t year_month_day,
+		       uint64_t hour_minute_second_millisecond);
+int64_t opal_cec_power_down(uint64_t request);
+int64_t opal_cec_reboot(void);
+int64_t opal_read_nvram(uint64_t buffer, uint64_t size, uint64_t offset);
+int64_t opal_write_nvram(uint64_t buffer, uint64_t size, uint64_t offset);
+int64_t opal_handle_interrupt(uint64_t isn, uint64_t *outstanding_event_mask);
+int64_t opal_poll_events(uint64_t *outstanding_event_mask);
+int64_t opal_pci_set_hub_tce_memory(uint64_t hub_id, uint64_t tce_mem_addr,
+				    uint64_t tce_mem_size);
+int64_t opal_pci_set_phb_tce_memory(uint64_t phb_id, uint64_t tce_mem_addr,
+				    uint64_t tce_mem_size);
+int64_t opal_pci_config_read_byte(uint64_t phb_id, uint64_t bus_dev_func,
+				  uint64_t offset, uint8_t *data);
+int64_t opal_pci_config_read_half_word(uint64_t phb_id, uint64_t bus_dev_func,
+				       uint64_t offset, uint16_t *data);
+int64_t opal_pci_config_read_word(uint64_t phb_id, uint64_t bus_dev_func,
+				  uint64_t offset, uint32_t *data);
+int64_t opal_pci_config_write_byte(uint64_t phb_id, uint64_t bus_dev_func,
+				   uint64_t offset, uint8_t data);
+int64_t opal_pci_config_write_half_word(uint64_t phb_id, uint64_t bus_dev_func,
+					uint64_t offset, uint16_t data);
+int64_t opal_pci_config_write_word(uint64_t phb_id, uint64_t bus_dev_func,
+				   uint64_t offset, uint32_t data);
+int64_t opal_set_xive(uint32_t isn, uint16_t server, uint8_t priority);
+int64_t opal_get_xive(uint32_t isn, uint16_t *server, uint8_t *priority);
+int64_t opal_register_exception_handler(uint64_t opal_exception,
+					uint64_t handler_address,
+					uint64_t glue_cache_line);
+int64_t opal_pci_eeh_freeze_status(uint64_t phb_id, uint64_t pe_number,
+				   uint8_t *freeze_state,
+				   uint16_t *pci_error_type,
+				   uint64_t *phb_status);
+int64_t opal_pci_eeh_freeze_clear(uint64_t phb_id, uint64_t pe_number,
+				  uint64_t eeh_action_token);
+int64_t opal_pci_shpc(uint64_t phb_id, uint64_t shpc_action, uint8_t *state);
+
+
+
+int64_t opal_pci_phb_mmio_enable(uint64_t phb_id, uint16_t window_type,
+				 uint16_t window_num, uint16_t enable);
+int64_t opal_pci_set_phb_mem_window(uint64_t phb_id, uint16_t window_type,
+				    uint16_t window_num,
+				    uint64_t starting_real_address,
+				    uint64_t starting_pci_address,
+				    uint16_t segment_size);
+int64_t opal_pci_map_pe_mmio_window(uint64_t phb_id, uint16_t pe_number,
+				    uint16_t window_type, uint16_t window_num,
+				    uint16_t segment_num);
+int64_t opal_pci_set_phb_table_memory(uint64_t phb_id, uint64_t rtt_addr,
+				      uint64_t ivt_addr, uint64_t ivt_len,
+				      uint64_t reject_array_addr,
+				      uint64_t peltv_addr);
+int64_t opal_pci_set_pe(uint64_t phb_id, uint64_t pe_number, uint64_t bus_dev_func,
+			uint8_t bus_compare, uint8_t dev_compare, uint8_t func_compare,
+			uint8_t pe_action);
+int64_t opal_pci_set_peltv(uint64_t phb_id, uint32_t parent_pe, uint32_t child_pe,
+			   uint8_t state);
+int64_t opal_pci_set_mve(uint64_t phb_id, uint32_t mve_number, uint32_t pe_number);
+int64_t opal_pci_set_mve_enable(uint64_t phb_id, uint32_t mve_number,
+				uint32_t state);
+int64_t opal_pci_get_xive_reissue(uint64_t phb_id, uint32_t xive_number,
+				  uint8_t *p_bit, uint8_t *q_bit);
+int64_t opal_pci_set_xive_reissue(uint64_t phb_id, uint32_t xive_number,
+				  uint8_t p_bit, uint8_t q_bit);
+int64_t opal_pci_set_xive_pe(uint64_t phb_id, uint32_t pe_number,
+			     uint32_t xive_num);
+int64_t opal_get_xive_source(uint64_t phb_id, uint32_t xive_num,
+			     int32_t *interrupt_source_number);
+int64_t opal_get_msi_32(uint64_t phb_id, uint32_t mve_number, uint32_t xive_num,
+			uint8_t msi_range, uint32_t *msi_address,
+			uint32_t *message_data);
+int64_t opal_get_msi_64(uint64_t phb_id, uint32_t mve_number,
+			uint32_t xive_num, uint8_t msi_range,
+			uint64_t *msi_address, uint32_t *message_data);
+int64_t opal_start_cpu(uint64_t thread_number, uint64_t start_address);
+int64_t opal_query_cpu_status(uint64_t thread_number, uint8_t *thread_status);
+int64_t opal_write_oppanel(oppanel_line_t *lines, uint64_t num_lines);
+int64_t opal_pci_map_pe_dma_window(uint64_t phb_id, uint16_t pe_number, uint16_t window_id,
+				   uint16_t tce_levels, uint64_t tce_table_addr,
+				   uint64_t tce_table_size, uint64_t tce_page_size);
+int64_t opal_pci_map_pe_dma_window_real(uint64_t phb_id, uint16_t pe_number,
+					uint16_t dma_window_number, uint64_t pci_start_addr,
+					uint64_t pci_mem_size);
+int64_t opal_pci_reset(uint64_t phb_id, uint8_t reset_scope, uint8_t assert_state);
+
+/* Internal functions */
+extern int early_init_dt_scan_opal(unsigned long node, const char *uname, int depth, void *data);
+
+extern int opal_get_chars(uint32_t vtermno, char *buf, int count);
+extern int opal_put_chars(uint32_t vtermno, const char *buf, int total_len);
+
+extern void hvc_opal_init_early(void);
+
+/* Internal functions */
+extern int early_init_dt_scan_opal(unsigned long node, const char *uname,
+				   int depth, void *data);
+
+#endif /* __ASSEMBLY__ */
 
 #endif /* __OPAL_H */
diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
index 7b90c56..831a201 100644
--- a/arch/powerpc/kernel/prom.c
+++ b/arch/powerpc/kernel/prom.c
@@ -54,6 +54,8 @@
 #include <asm/pci-bridge.h>
 #include <asm/phyp_dump.h>
 #include <asm/kexec.h>
+#include <asm/opal.h>
+
 #include <mm/mmu_decl.h>
 
 #ifdef DEBUG
@@ -707,6 +709,11 @@ void __init early_init_devtree(void *params)
 	of_scan_flat_dt(early_init_dt_scan_rtas, NULL);
 #endif
 
+#ifdef CONFIG_PPC_POWERNV
+	/* Some machines might need OPAL info for debugging, grab it now. */
+	of_scan_flat_dt(early_init_dt_scan_opal, NULL);
+#endif
+
 #ifdef CONFIG_PHYP_DUMP
 	/* scan tree to see if dump occurred during last boot */
 	of_scan_flat_dt(early_init_dt_scan_phyp_dump, NULL);
diff --git a/arch/powerpc/platforms/powernv/Kconfig b/arch/powerpc/platforms/powernv/Kconfig
index 5cd04f5..268cadb 100644
--- a/arch/powerpc/platforms/powernv/Kconfig
+++ b/arch/powerpc/platforms/powernv/Kconfig
@@ -1,12 +1,16 @@
 config PPC_POWERNV
 	depends on PPC64 && PPC_BOOK3S
 	bool "IBM PowerNV (Non-Virtualized) platform support"	
-	select PPC_RTAS
 	select PPC_NATIVE
 	select PPC_XICS
 	select PPC_ICP_NATIVE
-	select PPC_ICS_RTAS
 	select PPC_P7_NAP
 	select PPC_PCI_CHOICE if EMBEDDED
 	default y
 
+config PPC_POWERNV_RTAS
+	depends on PPC_POWERNV
+	bool "Support for RTAS based PowerNV platforms such as BML"
+	default y
+	select PPC_ICS_RTAS
+	select PPC_RTAS
diff --git a/arch/powerpc/platforms/powernv/Makefile b/arch/powerpc/platforms/powernv/Makefile
index 4971330..8f69c0d 100644
--- a/arch/powerpc/platforms/powernv/Makefile
+++ b/arch/powerpc/platforms/powernv/Makefile
@@ -1,2 +1,2 @@
-obj-y			+= setup.o opal-takeover.o
+obj-y			+= setup.o opal-takeover.o opal-wrappers.o opal.o
 obj-$(CONFIG_SMP)	+= smp.o
diff --git a/arch/powerpc/platforms/powernv/opal-wrappers.S b/arch/powerpc/platforms/powernv/opal-wrappers.S
new file mode 100644
index 0000000..4a3f46d
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/opal-wrappers.S
@@ -0,0 +1,101 @@
+/*
+ * PowerNV OPAL API wrappers
+ *
+ * Copyright 2011 IBM Corp.
+ *
+ * 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.
+ */
+
+#include <asm/ppc_asm.h>
+#include <asm/hvcall.h>
+#include <asm/asm-offsets.h>
+#include <asm/opal.h>
+
+/* TODO:
+ *
+ * - Trace irqs in/off (needs saving/restoring all args, argh...)
+ * - Get r11 feed up by Dave so I can have better register usage
+ */
+#define OPAL_CALL(name, token)		\
+ _GLOBAL(name);				\
+	mflr	r0;			\
+	mfcr	r12;			\
+	std	r0,16(r1);		\
+	std	r12,8(r1);		\
+	std	r1,PACAR1(r13);		\
+	li	r0,0;			\
+	mfmsr	r12;			\
+	ori	r0,r0,MSR_EE;		\
+	std	r12,PACASAVEDMSR(r13);	\
+	andc	r12,r12,r0;		\
+	mtmsrd	r12,1;			\
+	LOAD_REG_ADDR(r0,.opal_return);	\
+	mtlr	r0;			\
+	li	r0,MSR_DR|MSR_IR;	\
+	andc	r12,r12,r0;		\
+	li	r0,token;		\
+	mtspr	SPRN_HSRR1,r12;		\
+	LOAD_REG_ADDR(r11,opal);	\
+	ld	r12,8(r11);		\
+	ld	r2,0(r11);		\
+	mtspr	SPRN_HSRR0,r12;		\
+	hrfid
+
+_STATIC(opal_return)
+	ld	r2,PACATOC(r13);
+	ld	r4,8(r1);
+	ld	r5,16(r1);
+	ld	r6,PACASAVEDMSR(r13);
+	mtspr	SPRN_SRR0,r5;
+	mtspr	SPRN_SRR1,r6;
+	mtcr	r4;
+	rfid
+
+OPAL_CALL(opal_console_write,			OPAL_CONSOLE_WRITE);
+OPAL_CALL(opal_console_read,			OPAL_CONSOLE_READ);
+OPAL_CALL(opal_console_write_buffer_space,	OPAL_CONSOLE_WRITE_BUFFER_SPACE);
+OPAL_CALL(opal_rtc_read,			OPAL_RTC_READ);
+OPAL_CALL(opal_rtc_write,			OPAL_RTC_WRITE);
+OPAL_CALL(opal_cec_power_down,			OPAL_CEC_POWER_DOWN);
+OPAL_CALL(opal_cec_reboot,			OPAL_CEC_REBOOT);
+OPAL_CALL(opal_read_nvram,			OPAL_READ_NVRAM);
+OPAL_CALL(opal_write_nvram,			OPAL_WRITE_NVRAM);
+OPAL_CALL(opal_handle_interrupt,		OPAL_HANDLE_INTERRUPT);
+OPAL_CALL(opal_poll_events,			OPAL_POLL_EVENTS);
+OPAL_CALL(opal_pci_set_hub_tce_memory,		OPAL_PCI_SET_HUB_TCE_MEMORY);
+OPAL_CALL(opal_pci_set_phb_tce_memory,		OPAL_PCI_SET_PHB_TCE_MEMORY);
+OPAL_CALL(opal_pci_config_read_byte,		OPAL_PCI_CONFIG_READ_BYTE);
+OPAL_CALL(opal_pci_config_read_half_word,	OPAL_PCI_CONFIG_READ_HALF_WORD);
+OPAL_CALL(opal_pci_config_read_word,		OPAL_PCI_CONFIG_READ_WORD);
+OPAL_CALL(opal_pci_config_write_byte,		OPAL_PCI_CONFIG_WRITE_BYTE);
+OPAL_CALL(opal_pci_config_write_half_word,	OPAL_PCI_CONFIG_WRITE_HALF_WORD);
+OPAL_CALL(opal_pci_config_write_word,		OPAL_PCI_CONFIG_WRITE_WORD);
+OPAL_CALL(opal_set_xive,			OPAL_SET_XIVE);
+OPAL_CALL(opal_get_xive,			OPAL_GET_XIVE);
+OPAL_CALL(opal_register_exception_handler,	OPAL_REGISTER_OPAL_EXCEPTION_HANDLER);
+OPAL_CALL(opal_pci_eeh_freeze_status,		OPAL_PCI_EEH_FREEZE_STATUS);
+OPAL_CALL(opal_pci_eeh_freeze_clear,		OPAL_PCI_EEH_FREEZE_CLEAR);
+OPAL_CALL(opal_pci_shpc,			OPAL_PCI_SHPC);
+OPAL_CALL(opal_pci_phb_mmio_enable,		OPAL_PCI_PHB_MMIO_ENABLE);
+OPAL_CALL(opal_pci_set_phb_mem_window,		OPAL_PCI_SET_PHB_MEM_WINDOW);
+OPAL_CALL(opal_pci_map_pe_mmio_window,		OPAL_PCI_MAP_PE_MMIO_WINDOW);
+OPAL_CALL(opal_pci_set_phb_table_memory,	OPAL_PCI_SET_PHB_TABLE_MEMORY);
+OPAL_CALL(opal_pci_set_pe,			OPAL_PCI_SET_PE);
+OPAL_CALL(opal_pci_set_peltv,			OPAL_PCI_SET_PELTV);
+OPAL_CALL(opal_pci_set_mve,			OPAL_PCI_SET_MVE);
+OPAL_CALL(opal_pci_set_mve_enable,		OPAL_PCI_SET_MVE_ENABLE);
+OPAL_CALL(opal_pci_get_xive_reissue,		OPAL_PCI_GET_XIVE_REISSUE);
+OPAL_CALL(opal_pci_set_xive_reissue,		OPAL_PCI_SET_XIVE_REISSUE);
+OPAL_CALL(opal_pci_set_xive_pe,			OPAL_PCI_SET_XIVE_PE);
+OPAL_CALL(opal_get_xive_source,			OPAL_GET_XIVE_SOURCE);
+OPAL_CALL(opal_get_msi_32,			OPAL_GET_MSI_32);
+OPAL_CALL(opal_get_msi_64,			OPAL_GET_MSI_64);
+OPAL_CALL(opal_start_cpu,			OPAL_START_CPU);
+OPAL_CALL(opal_query_cpu_status,		OPAL_QUERY_CPU_STATUS);
+OPAL_CALL(opal_write_oppanel,			OPAL_WRITE_OPPANEL);
+OPAL_CALL(opal_pci_map_pe_dma_window,		OPAL_PCI_MAP_PE_DMA_WINDOW);
+OPAL_CALL(opal_pci_map_pe_dma_window_real,	OPAL_PCI_MAP_PE_DMA_WINDOW_REAL);
+OPAL_CALL(opal_pci_reset,			OPAL_PCI_RESET);
diff --git a/arch/powerpc/platforms/powernv/opal.c b/arch/powerpc/platforms/powernv/opal.c
new file mode 100644
index 0000000..8d55107
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/opal.c
@@ -0,0 +1,154 @@
+/*
+ * PowerNV OPAL high level interfaces
+ *
+ * Copyright 2011 IBM Corp.
+ *
+ * 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.
+ */
+
+#undef DEBUG
+
+#include <linux/types.h>
+#include <linux/of.h>
+#include <linux/of_platform.h>
+#include <asm/opal.h>
+#include <asm/firmware.h>
+
+#include "powernv.h"
+
+struct opal {
+	u64 base;
+	u64 entry;
+} opal;
+
+static struct device_node *opal_node;
+static DEFINE_SPINLOCK(opal_write_lock);
+
+int __init early_init_dt_scan_opal(unsigned long node,
+				   const char *uname, int depth, void *data)
+{
+	const void *basep, *entryp;
+	unsigned long basesz, entrysz;
+
+	if (depth != 1 || strcmp(uname, "ibm,opal") != 0)
+		return 0;
+
+	basep  = of_get_flat_dt_prop(node, "opal-base-address", &basesz);
+	entryp = of_get_flat_dt_prop(node, "opal-entry-address", &entrysz);
+
+	if (!basep || !entryp)
+		return 1;
+
+	opal.base = of_read_number(basep, basesz/4);
+	opal.entry = of_read_number(entryp, entrysz/4);
+
+	pr_debug("OPAL Base  = 0x%llx (basep=%p basesz=%ld)\n",
+		 opal.base, basep, basesz);
+	pr_debug("OPAL Entry = 0x%llx (entryp=%p basesz=%ld)\n",
+		 opal.entry, entryp, entrysz);
+
+	powerpc_firmware_features |= FW_FEATURE_OPAL;
+	if (of_flat_dt_is_compatible(node, "ibm,opal-v2")) {
+		powerpc_firmware_features |= FW_FEATURE_OPALv2;
+		printk("OPAL V2 detected !\n");
+	} else {
+		printk("OPAL V1 detected !\n");
+	}
+
+	return 1;
+}
+
+int opal_get_chars(uint32_t vtermno, char *buf, int count)
+{
+	s64 len, rc;
+	u64 evt;
+
+	if (!opal.entry)
+		return 0;
+	opal_poll_events(&evt);
+	if ((evt & OPAL_EVENT_CONSOLE_INPUT) == 0)
+		return 0;
+	len = count;
+	rc = opal_console_read(vtermno, &len, buf);
+	if (rc == OPAL_SUCCESS)
+		return len;
+	return 0;
+}
+
+int opal_put_chars(uint32_t vtermno, const char *data, int total_len)
+{
+	int written = 0;
+	s64 len, rc = OPAL_BUSY;
+	unsigned long flags;
+	u64 evt;
+
+	if (!opal.entry)
+		return 0;
+
+	/* We want put_chars to be atomic to avoid mangling of hvsi
+	 * packets. To do that, we first test for room and return
+	 * -EAGAIN if there isn't enough
+	 */
+	spin_lock_irqsave(&opal_write_lock, flags);
+	rc = opal_console_write_buffer_space(vtermno, &len);
+	if (rc || len < total_len) {
+		spin_unlock_irqrestore(&opal_write_lock, flags);
+		/* Closed -> drop characters */
+		if (rc)
+			return total_len;
+		opal_poll_events(&evt);
+		return -EAGAIN;
+	}
+
+	/* We still try to handle partial completions, though they
+	 * should no longer happen.
+	 */
+	while(total_len > 0 && (rc == OPAL_BUSY ||
+				rc == OPAL_BUSY_EVENT || rc == OPAL_SUCCESS)) {
+		len = total_len;
+		rc = opal_console_write(vtermno, &len, data);
+		if (rc == OPAL_SUCCESS) {
+			total_len -= len;
+			data += len;
+			written += len;
+		}
+		/* This is a bit nasty but we need that for the console to
+		 * flush when there aren't any interrupts. We will clean
+		 * things a bit later to limit that to synchronous path
+		 * such as the kernel console and xmon/udbg
+		 */
+		do
+			opal_poll_events(&evt);
+		while(rc == OPAL_SUCCESS && (evt & OPAL_EVENT_CONSOLE_OUTPUT));
+	}
+	spin_unlock_irqrestore(&opal_write_lock, flags);
+	return written;
+}
+
+static int __init opal_init(void)
+{
+	struct device_node *np, *consoles;
+
+	opal_node = of_find_node_by_path("/ibm,opal");
+	if (!opal_node) {
+		pr_warn("opal: Node not found\n");
+		return -ENODEV;
+	}
+	if (firmware_has_feature(FW_FEATURE_OPALv2))
+		consoles = of_find_node_by_path("/ibm,opal/consoles");
+	else
+		consoles = of_node_get(opal_node);
+
+	/* Register serial ports */
+	for_each_child_of_node(consoles, np) {
+		if (strcmp(np->name, "serial"))
+			continue;
+		of_platform_device_create(np, NULL, NULL);
+	}
+	of_node_put(consoles);
+	return 0;
+}
+subsys_initcall(opal_init);
diff --git a/arch/powerpc/platforms/powernv/setup.c b/arch/powerpc/platforms/powernv/setup.c
index 569f9cc..b6e5ff8 100644
--- a/arch/powerpc/platforms/powernv/setup.c
+++ b/arch/powerpc/platforms/powernv/setup.c
@@ -74,6 +74,12 @@ static void pnv_show_cpuinfo(struct seq_file *m)
 	if (root)
 		model = of_get_property(root, "model", NULL);
 	seq_printf(m, "machine\t\t: PowerNV %s\n", model);
+	if (firmware_has_feature(FW_FEATURE_OPALv2))
+		seq_printf(m, "firmware\t: OPAL v2\n");
+	else if (firmware_has_feature(FW_FEATURE_OPAL))
+		seq_printf(m, "firmware\t: OPAL v1\n");
+	else
+		seq_printf(m, "firmware\t: BML\n");
 	of_node_put(root);
 }
 
diff --git a/arch/powerpc/platforms/powernv/smp.c b/arch/powerpc/platforms/powernv/smp.c
index 4f4ec37..e877366 100644
--- a/arch/powerpc/platforms/powernv/smp.c
+++ b/arch/powerpc/platforms/powernv/smp.c
@@ -30,6 +30,7 @@
 #include <asm/vdso_datapage.h>
 #include <asm/cputhreads.h>
 #include <asm/xics.h>
+#include <asm/opal.h>
 
 #include "powernv.h"
 
@@ -62,6 +63,28 @@ static int pnv_smp_cpu_bootable(unsigned int nr)
 	return 1;
 }
 
+int __devinit pnv_smp_kick_cpu(int nr)
+{
+	unsigned int pcpu = get_hard_smp_processor_id(nr);
+	unsigned long start_here = __pa(*((unsigned long *)
+					  generic_secondary_smp_init));
+	long rc;
+
+	BUG_ON(nr < 0 || nr >= NR_CPUS);
+
+	/* On OPAL v2 the CPU are still spinning inside OPAL itself,
+	 * get them back now
+	 */
+	if (firmware_has_feature(FW_FEATURE_OPALv2)) {
+		pr_devel("OPAL: Starting CPU %d (HW 0x%x)...\n", nr, pcpu);
+		rc = opal_start_cpu(pcpu, start_here);
+		if (rc != OPAL_SUCCESS)
+			pr_warn("OPAL Error %ld starting CPU %d\n",
+				rc, nr);
+	}
+	return smp_generic_kick_cpu(nr);
+}
+
 #ifdef CONFIG_HOTPLUG_CPU
 
 static int pnv_smp_cpu_disable(void)
@@ -127,7 +150,7 @@ static struct smp_ops_t pnv_smp_ops = {
 	.message_pass	= smp_muxed_ipi_message_pass,
 	.cause_ipi	= NULL,	/* Filled at runtime by xics_smp_probe() */
 	.probe		= xics_smp_probe,
-	.kick_cpu	= smp_generic_kick_cpu,
+	.kick_cpu	= pnv_smp_kick_cpu,
 	.setup_cpu	= pnv_smp_setup_cpu,
 	.cpu_bootable	= pnv_smp_cpu_bootable,
 #ifdef CONFIG_HOTPLUG_CPU
-- 
1.7.4.1

[PATCH 12/21] powerpc/powernv: Add support for instanciating OPAL v2 from Open Firmware

[Benjamin Herrenschmidt]

OPAL v2 is instantiated in a way similar to RTAS using Open Firmware
client interface calls, and the resulting address and entry point are
put in the device-tree

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/prom_init.c |  136 +++++++++++++++++++++++++++++++++------
 1 files changed, 117 insertions(+), 19 deletions(-)

diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c
index e3f3904..4ef74ea 100644
--- a/arch/powerpc/kernel/prom_init.c
+++ b/arch/powerpc/kernel/prom_init.c
@@ -140,7 +140,9 @@ struct mem_map_entry {
 
 typedef u32 cell_t;
 
-extern void __start(unsigned long r3, unsigned long r4, unsigned long r5);
+extern void __start(unsigned long r3, unsigned long r4, unsigned long r5,
+		    unsigned long r6, unsigned long r7, unsigned long r8,
+		    unsigned long r9);
 
 #ifdef CONFIG_PPC64
 extern int enter_prom(struct prom_args *args, unsigned long entry);
@@ -1293,6 +1295,11 @@ static int __initdata prom_rtas_start_cpu;
 static u64 __initdata prom_rtas_data;
 static u64 __initdata prom_rtas_entry;
 
+#ifdef CONFIG_PPC_EARLY_DEBUG_OPAL
+static u64 __initdata prom_opal_base;
+static u64 __initdata prom_opal_entry;
+#endif
+
 /* XXX Don't change this structure without updating opal-takeover.S */
 static struct opal_secondary_data {
 	s64				ack;	/*  0 */
@@ -1468,6 +1475,76 @@ static void prom_opal_takeover(void)
 	for (;;)
 		opal_do_takeover(args);
 }
+
+/*
+ * Allocate room for and instantiate OPAL
+ */
+static void __init prom_instantiate_opal(void)
+{
+	phandle opal_node;
+	ihandle opal_inst;
+	u64 base, entry;
+	u64 size = 0, align = 0x10000;
+	u32 rets[2];
+
+	prom_debug("prom_instantiate_opal: start...\n");
+
+	opal_node = call_prom("finddevice", 1, 1, ADDR("/ibm,opal"));
+	prom_debug("opal_node: %x\n", opal_node);
+	if (!PHANDLE_VALID(opal_node))
+		return;
+
+	prom_getprop(opal_node, "opal-runtime-size", &size, sizeof(size));
+	if (size == 0)
+		return;
+	prom_getprop(opal_node, "opal-runtime-alignment", &align,
+		     sizeof(align));
+
+	base = alloc_down(size, align, 0);
+	if (base == 0) {
+		prom_printf("OPAL allocation failed !\n");
+		return;
+	}
+
+	opal_inst = call_prom("open", 1, 1, ADDR("/ibm,opal"));
+	if (!IHANDLE_VALID(opal_inst)) {
+		prom_printf("opening opal package failed (%x)\n", opal_inst);
+		return;
+	}
+
+	prom_printf("instantiating opal at 0x%x...", base);
+
+	if (call_prom_ret("call-method", 4, 3, rets,
+			  ADDR("load-opal-runtime"),
+			  opal_inst,
+			  base >> 32, base & 0xffffffff) != 0
+	    || (rets[0] == 0 && rets[1] == 0)) {
+		prom_printf(" failed\n");
+		return;
+	}
+	entry = (((u64)rets[0]) << 32) | rets[1];
+
+	prom_printf(" done\n");
+
+	reserve_mem(base, size);
+
+	prom_debug("opal base     = 0x%x\n", base);
+	prom_debug("opal align    = 0x%x\n", align);
+	prom_debug("opal entry    = 0x%x\n", entry);
+	prom_debug("opal size     = 0x%x\n", (long)size);
+
+	prom_setprop(opal_node, "/ibm,opal", "opal-base-address",
+		     &base, sizeof(base));
+	prom_setprop(opal_node, "/ibm,opal", "opal-entry-address",
+		     &entry, sizeof(entry));
+
+#ifdef CONFIG_PPC_EARLY_DEBUG_OPAL
+	RELOC(prom_opal_base) = base;
+	RELOC(prom_opal_entry) = entry;
+#endif
+	prom_debug("prom_instantiate_opal: end...\n");
+}
+
 #endif /* CONFIG_PPC_POWERNV */
 
 /*
@@ -1863,7 +1940,7 @@ static int __init prom_find_machine_type(void)
 	int x;
 #endif
 
-	/* Look for a PowerMac */
+	/* Look for a PowerMac or a Cell */
 	len = prom_getprop(_prom->root, "compatible",
 			   compat, sizeof(compat)-1);
 	if (len > 0) {
@@ -1889,7 +1966,11 @@ static int __init prom_find_machine_type(void)
 		}
 	}
 #ifdef CONFIG_PPC64
-	/* If not a mac, try to figure out if it's an IBM pSeries or any other
+	/* Try to detect OPAL */
+	if (PHANDLE_VALID(call_prom("finddevice", 1, 1, ADDR("/ibm,opal"))))
+		return PLATFORM_OPAL;
+
+	/* Try to figure out if it's an IBM pSeries or any other
 	 * PAPR compliant platform. We assume it is if :
 	 *  - /device_type is "chrp" (please, do NOT use that for future
 	 *    non-IBM designs !
@@ -2116,7 +2197,7 @@ static void __init scan_dt_build_struct(phandle node, unsigned long *mem_start,
 	unsigned long soff;
 	unsigned char *valp;
 	static char pname[MAX_PROPERTY_NAME];
-	int l, room;
+	int l, room, has_phandle;
 
 	dt_push_token(OF_DT_BEGIN_NODE, mem_start, mem_end);
 
@@ -2200,19 +2281,26 @@ static void __init scan_dt_build_struct(phandle node, unsigned long *mem_start,
 		valp = make_room(mem_start, mem_end, l, 4);
 		call_prom("getprop", 4, 1, node, RELOC(pname), valp, l);
 		*mem_start = _ALIGN(*mem_start, 4);
+
+		if (!strcmp(RELOC(pname), RELOC("phandle")))
+			has_phandle = 1;
 	}
 
-	/* Add a "linux,phandle" property. */
-	soff = dt_find_string(RELOC("linux,phandle"));
-	if (soff == 0)
-		prom_printf("WARNING: Can't find string index for"
-			    " <linux-phandle> node %s\n", path);
-	else {
-		dt_push_token(OF_DT_PROP, mem_start, mem_end);
-		dt_push_token(4, mem_start, mem_end);
-		dt_push_token(soff, mem_start, mem_end);
-		valp = make_room(mem_start, mem_end, 4, 4);
-		*(u32 *)valp = node;
+	/* Add a "linux,phandle" property if no "phandle" property already
+	 * existed (can happen with OPAL)
+	 */
+	if (!has_phandle) {
+		soff = dt_find_string(RELOC("linux,phandle"));
+		if (soff == 0)
+			prom_printf("WARNING: Can't find string index for"
+				    " <linux-phandle> node %s\n", path);
+		else {
+			dt_push_token(OF_DT_PROP, mem_start, mem_end);
+			dt_push_token(4, mem_start, mem_end);
+			dt_push_token(soff, mem_start, mem_end);
+			valp = make_room(mem_start, mem_end, 4, 4);
+			*(u32 *)valp = node;
+		}
 	}
 
 	/* do all our children */
@@ -2746,6 +2834,7 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,
 	 * between pSeries SMP and pSeries LPAR
 	 */
 	RELOC(of_platform) = prom_find_machine_type();
+	prom_printf("Detected machine type: %x\n", RELOC(of_platform));
 
 #ifndef CONFIG_RELOCATABLE
 	/* Bail if this is a kdump kernel. */
@@ -2807,7 +2896,8 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,
 	 * On non-powermacs, try to instantiate RTAS. PowerMacs don't
 	 * have a usable RTAS implementation.
 	 */
-	if (RELOC(of_platform) != PLATFORM_POWERMAC)
+	if (RELOC(of_platform) != PLATFORM_POWERMAC &&
+	    RELOC(of_platform) != PLATFORM_OPAL)
 		prom_instantiate_rtas();
 
 #ifdef CONFIG_PPC_POWERNV
@@ -2818,7 +2908,8 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,
 			prom_opal_hold_cpus();
 			prom_opal_takeover();
 		}
-	}
+	} else if (RELOC(of_platform) == PLATFORM_OPAL)
+		prom_instantiate_opal();
 #endif
 
 	/*
@@ -2826,7 +2917,8 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,
 	 *
 	 * PowerMacs use a different mechanism to spin CPUs
 	 */
-	if (RELOC(of_platform) != PLATFORM_POWERMAC)
+	if (RELOC(of_platform) != PLATFORM_POWERMAC &&
+	    RELOC(of_platform) != PLATFORM_OPAL)
 		prom_hold_cpus();
 
 	/*
@@ -2894,7 +2986,13 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,
 	reloc_got2(-offset);
 #endif
 
-	__start(hdr, kbase, 0);
+#ifdef CONFIG_PPC_EARLY_DEBUG_OPAL
+	/* OPAL early debug gets the OPAL base & entry in r8 and r9 */
+	__start(hdr, kbase, 0, 0, 0,
+		RELOC(prom_opal_base), RELOC(prom_opal_entry));
+#else
+	__start(hdr, kbase, 0, 0, 0, 0, 0);
+#endif
 
 	return 0;
 }
-- 
1.7.4.1

[PATCH 13/21] powerpc/powernv: Support for OPAL console

[Benjamin Herrenschmidt]

This adds a udbg and an hvc console backend for supporting a console
using the OPAL console interfaces.

On OPAL v1 we have hvc0 mapped to whatever console the system was
configured for (network or hvsi serial port) via the service
processor.

On OPAL v2 we have hvcN mapped to the Nth console provided by OPAL
which generally corresponds to:

	hvc0 : network console (raw protocol)
	hvc1 : serial port S1 (hvsi)
	hvc2 : serial port S2 (hvsi)

Note: At this point, early debug console only works with OPAL v1
and shouldn't be enabled in a normal kernel.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/Kconfig.debug             |   31 +++
 arch/powerpc/include/asm/opal.h        |    5 +
 arch/powerpc/include/asm/udbg.h        |    2 +
 arch/powerpc/kernel/head_64.S          |   14 +-
 arch/powerpc/kernel/udbg.c             |    4 +
 arch/powerpc/platforms/powernv/opal.c  |   31 ++-
 arch/powerpc/platforms/powernv/setup.c |   14 +-
 drivers/tty/hvc/Kconfig                |    9 +
 drivers/tty/hvc/Makefile               |    1 +
 drivers/tty/hvc/hvc_opal.c             |  424 ++++++++++++++++++++++++++++++++
 drivers/tty/hvc/hvsi_lib.c             |    4 +-
 11 files changed, 517 insertions(+), 22 deletions(-)
 create mode 100644 drivers/tty/hvc/hvc_opal.c

diff --git a/arch/powerpc/Kconfig.debug b/arch/powerpc/Kconfig.debug
index cc01f1d..e61830f 100644
--- a/arch/powerpc/Kconfig.debug
+++ b/arch/powerpc/Kconfig.debug
@@ -257,8 +257,27 @@ config PPC_EARLY_DEBUG_WSP
 	depends on PPC_WSP
 	select PPC_UDBG_16550
 
+config PPC_EARLY_DEBUG_OPAL_RAW
+	bool "OPAL raw console"
+	depends on HVC_OPAL
+	help
+	  Select this to enable early debugging for the PowerNV platform
+	  using a "raw" console
+
+config PPC_EARLY_DEBUG_OPAL_HVSI
+	bool "OPAL hvsi console"
+	depends on HVC_OPAL
+	help
+	  Select this to enable early debugging for the PowerNV platform
+	  using an "hvsi" console
+
 endchoice
 
+config PPC_EARLY_DEBUG_OPAL
+	def_bool y
+	depends on PPC_EARLY_DEBUG_OPAL_RAW || PPC_EARLY_DEBUG_OPAL_HVSI
+
+
 config PPC_EARLY_DEBUG_HVSI_VTERMNO
 	hex "vterm number to use with early debug HVSI"
 	depends on PPC_EARLY_DEBUG_LPAR_HVSI
@@ -267,6 +286,18 @@ config PPC_EARLY_DEBUG_HVSI_VTERMNO
 	  You probably want 0x30000000 for your first serial port and
 	  0x30000001 for your second one
 
+config PPC_EARLY_DEBUG_OPAL_VTERMNO
+	hex "vterm number to use with OPAL early debug"
+	depends on PPC_EARLY_DEBUG_OPAL
+	default "0"
+	help
+	  This correspond to which /dev/hvcN you want to use for early
+	  debug.
+
+	  On OPAL v1 (takeover) this should always be 0
+	  On OPAL v2, this will be 0 for network console and 1 or 2 for
+	  the machine built-in serial ports.
+
 config PPC_EARLY_DEBUG_44x_PHYSLOW
 	hex "Low 32 bits of early debug UART physical address"
 	depends on PPC_EARLY_DEBUG_44x
diff --git a/arch/powerpc/include/asm/opal.h b/arch/powerpc/include/asm/opal.h
index 53cda41..dbeabea 100644
--- a/arch/powerpc/include/asm/opal.h
+++ b/arch/powerpc/include/asm/opal.h
@@ -425,6 +425,11 @@ extern void hvc_opal_init_early(void);
 extern int early_init_dt_scan_opal(unsigned long node, const char *uname,
 				   int depth, void *data);
 
+extern int opal_get_chars(uint32_t vtermno, char *buf, int count);
+extern int opal_put_chars(uint32_t vtermno, const char *buf, int total_len);
+
+extern void hvc_opal_init_early(void);
+
 #endif /* __ASSEMBLY__ */
 
 #endif /* __OPAL_H */
diff --git a/arch/powerpc/include/asm/udbg.h b/arch/powerpc/include/asm/udbg.h
index 93e05d1..2ac1753 100644
--- a/arch/powerpc/include/asm/udbg.h
+++ b/arch/powerpc/include/asm/udbg.h
@@ -54,6 +54,8 @@ extern void __init udbg_init_40x_realmode(void);
 extern void __init udbg_init_cpm(void);
 extern void __init udbg_init_usbgecko(void);
 extern void __init udbg_init_wsp(void);
+extern void __init udbg_init_debug_opal_raw(void);
+extern void __init udbg_init_debug_opal_hvsi(void);
 
 #endif /* __KERNEL__ */
 #endif /* _ASM_POWERPC_UDBG_H */
diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index dea8191..06c7251 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -53,7 +53,8 @@
  *   2. The kernel is entered at __start
  * -or- For OPAL entry:
  *   1. The MMU is off, processor in HV mode, primary CPU enters at 0
- *      with device-tree in gpr3
+ *      with device-tree in gpr3. We also get OPAL base in r8 and
+ *	entry in r9 for debugging purposes
  *   2. Secondary processors enter at 0x60 with PIR in gpr3
  *
  *  For iSeries:
@@ -335,6 +336,11 @@ _GLOBAL(__start_initialization_multiplatform)
 	/* Save parameters */
 	mr	r31,r3
 	mr	r30,r4
+#ifdef CONFIG_PPC_EARLY_DEBUG_OPAL
+	/* Save OPAL entry */
+	mr	r28,r8
+	mr	r29,r9
+#endif
 
 #ifdef CONFIG_PPC_BOOK3E
 	bl	.start_initialization_book3e
@@ -711,6 +717,12 @@ _INIT_STATIC(start_here_multiplatform)
 	bdnz	3b
 4:
 
+#ifdef CONFIG_PPC_EARLY_DEBUG_OPAL
+	/* Setup OPAL entry */
+	std	r28,0(r11);
+	std	r29,8(r11);
+#endif
+
 #ifndef CONFIG_PPC_BOOK3E
 	mfmsr	r6
 	ori	r6,r6,MSR_RI
diff --git a/arch/powerpc/kernel/udbg.c b/arch/powerpc/kernel/udbg.c
index faa82c1..ea82faf 100644
--- a/arch/powerpc/kernel/udbg.c
+++ b/arch/powerpc/kernel/udbg.c
@@ -67,6 +67,10 @@ void __init udbg_early_init(void)
 	udbg_init_usbgecko();
 #elif defined(CONFIG_PPC_EARLY_DEBUG_WSP)
 	udbg_init_wsp();
+#elif defined(CONFIG_PPC_EARLY_DEBUG_OPAL_RAW)
+	udbg_init_debug_opal_raw();
+#elif defined(CONFIG_PPC_EARLY_DEBUG_OPAL_HVSI)
+	udbg_init_debug_opal_hvsi();
 #endif
 
 #ifdef CONFIG_PPC_EARLY_DEBUG
diff --git a/arch/powerpc/platforms/powernv/opal.c b/arch/powerpc/platforms/powernv/opal.c
index 8d55107..7887733 100644
--- a/arch/powerpc/platforms/powernv/opal.c
+++ b/arch/powerpc/platforms/powernv/opal.c
@@ -67,7 +67,7 @@ int opal_get_chars(uint32_t vtermno, char *buf, int count)
 	u64 evt;
 
 	if (!opal.entry)
-		return 0;
+		return -ENODEV;
 	opal_poll_events(&evt);
 	if ((evt & OPAL_EVENT_CONSOLE_INPUT) == 0)
 		return 0;
@@ -81,31 +81,38 @@ int opal_get_chars(uint32_t vtermno, char *buf, int count)
 int opal_put_chars(uint32_t vtermno, const char *data, int total_len)
 {
 	int written = 0;
-	s64 len, rc = OPAL_BUSY;
+	s64 len, rc;
 	unsigned long flags;
 	u64 evt;
 
 	if (!opal.entry)
-		return 0;
+		return -ENODEV;
 
 	/* We want put_chars to be atomic to avoid mangling of hvsi
 	 * packets. To do that, we first test for room and return
-	 * -EAGAIN if there isn't enough
+	 * -EAGAIN if there isn't enough.
+	 *
+	 * Unfortunately, opal_console_write_buffer_space() doesn't
+	 * appear to work on opal v1, so we just assume there is
+	 * enough room and be done with it
 	 */
 	spin_lock_irqsave(&opal_write_lock, flags);
-	rc = opal_console_write_buffer_space(vtermno, &len);
-	if (rc || len < total_len) {
-		spin_unlock_irqrestore(&opal_write_lock, flags);
-		/* Closed -> drop characters */
-		if (rc)
-			return total_len;
-		opal_poll_events(&evt);
-		return -EAGAIN;
+	if (firmware_has_feature(FW_FEATURE_OPALv2)) {
+		rc = opal_console_write_buffer_space(vtermno, &len);
+		if (rc || len < total_len) {
+			spin_unlock_irqrestore(&opal_write_lock, flags);
+			/* Closed -> drop characters */
+			if (rc)
+				return total_len;
+			opal_poll_events(&evt);
+			return -EAGAIN;
+		}
 	}
 
 	/* We still try to handle partial completions, though they
 	 * should no longer happen.
 	 */
+	rc = OPAL_BUSY;
 	while(total_len > 0 && (rc == OPAL_BUSY ||
 				rc == OPAL_BUSY_EVENT || rc == OPAL_SUCCESS)) {
 		len = total_len;
diff --git a/arch/powerpc/platforms/powernv/setup.c b/arch/powerpc/platforms/powernv/setup.c
index b6e5ff8..07ba1ec 100644
--- a/arch/powerpc/platforms/powernv/setup.c
+++ b/arch/powerpc/platforms/powernv/setup.c
@@ -29,17 +29,12 @@
 #include <asm/machdep.h>
 #include <asm/firmware.h>
 #include <asm/xics.h>
+#include <asm/opal.h>
 
 #include "powernv.h"
 
 static void __init pnv_setup_arch(void)
 {
-	/* Force console to hvc for now until we have sorted out the
-	 * real console situation for the platform. This will make
-	 * hvc_udbg work at least.
-	 */
-	add_preferred_console("hvc", 0, NULL);
-
 	/* Initialize SMP */
 	pnv_smp_init();
 
@@ -55,7 +50,12 @@ static void __init pnv_setup_arch(void)
 
 static void __init pnv_init_early(void)
 {
-	/* XXX IOMMU */
+#ifdef CONFIG_HVC_OPAL
+	if (firmware_has_feature(FW_FEATURE_OPAL))
+		hvc_opal_init_early();
+	else
+#endif
+		add_preferred_console("hvc", 0, NULL);
 }
 
 static void __init pnv_init_IRQ(void)
diff --git a/drivers/tty/hvc/Kconfig b/drivers/tty/hvc/Kconfig
index e371753..4222035 100644
--- a/drivers/tty/hvc/Kconfig
+++ b/drivers/tty/hvc/Kconfig
@@ -34,6 +34,15 @@ config HVC_ISERIES
 	help
 	  iSeries machines support a hypervisor virtual console.
 
+config HVC_OPAL
+	bool "OPAL Console support"
+	depends on PPC_POWERNV
+	select HVC_DRIVER
+	select HVC_IRQ
+	default y
+	help
+	  PowerNV machines running under OPAL need that driver to get a console
+
 config HVC_RTAS
 	bool "IBM RTAS Console support"
 	depends on PPC_RTAS
diff --git a/drivers/tty/hvc/Makefile b/drivers/tty/hvc/Makefile
index e292053..89abf40b 100644
--- a/drivers/tty/hvc/Makefile
+++ b/drivers/tty/hvc/Makefile
@@ -1,4 +1,5 @@
 obj-$(CONFIG_HVC_CONSOLE)	+= hvc_vio.o hvsi_lib.o
+obj-$(CONFIG_HVC_OPAL)		+= hvc_opal.o hvsi_lib.o
 obj-$(CONFIG_HVC_OLD_HVSI)	+= hvsi.o
 obj-$(CONFIG_HVC_ISERIES)	+= hvc_iseries.o
 obj-$(CONFIG_HVC_RTAS)		+= hvc_rtas.o
diff --git a/drivers/tty/hvc/hvc_opal.c b/drivers/tty/hvc/hvc_opal.c
new file mode 100644
index 0000000..7b38512
--- /dev/null
+++ b/drivers/tty/hvc/hvc_opal.c
@@ -0,0 +1,424 @@
+/*
+ * opal driver interface to hvc_console.c
+ *
+ * Copyright 2011 Benjamin Herrenschmidt <benh@kernel.crashing.org>, IBM Corp.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ *
+ */
+
+#undef DEBUG
+
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/slab.h>
+#include <linux/console.h>
+#include <linux/of.h>
+#include <linux/of_platform.h>
+
+#include <asm/hvconsole.h>
+#include <asm/prom.h>
+#include <asm/firmware.h>
+#include <asm/hvsi.h>
+#include <asm/udbg.h>
+#include <asm/opal.h>
+
+#include "hvc_console.h"
+
+static const char hvc_opal_name[] = "hvc_opal";
+
+static struct of_device_id hvc_opal_match[] __devinitdata = {
+	{ .name = "serial", .compatible = "ibm,opal-console-raw" },
+	{ .name = "serial", .compatible = "ibm,opal-console-hvsi" },
+	{ },
+};
+
+typedef enum hv_protocol {
+	HV_PROTOCOL_RAW,
+	HV_PROTOCOL_HVSI
+} hv_protocol_t;
+
+struct hvc_opal_priv {
+	hv_protocol_t		proto;	/* Raw data or HVSI packets */
+	struct hvsi_priv	hvsi;	/* HVSI specific data */
+};
+static struct hvc_opal_priv *hvc_opal_privs[MAX_NR_HVC_CONSOLES];
+
+/* For early boot console */
+static struct hvc_opal_priv hvc_opal_boot_priv;
+static u32 hvc_opal_boot_termno;
+
+static const struct hv_ops hvc_opal_raw_ops = {
+	.get_chars = opal_get_chars,
+	.put_chars = opal_put_chars,
+	.notifier_add = notifier_add_irq,
+	.notifier_del = notifier_del_irq,
+	.notifier_hangup = notifier_hangup_irq,
+};
+
+static int hvc_opal_hvsi_get_chars(uint32_t vtermno, char *buf, int count)
+{
+	struct hvc_opal_priv *pv = hvc_opal_privs[vtermno];
+
+	if (WARN_ON(!pv))
+		return -ENODEV;
+
+	return hvsilib_get_chars(&pv->hvsi, buf, count);
+}
+
+static int hvc_opal_hvsi_put_chars(uint32_t vtermno, const char *buf, int count)
+{
+	struct hvc_opal_priv *pv = hvc_opal_privs[vtermno];
+
+	if (WARN_ON(!pv))
+		return -ENODEV;
+
+	return hvsilib_put_chars(&pv->hvsi, buf, count);
+}
+
+static int hvc_opal_hvsi_open(struct hvc_struct *hp, int data)
+{
+	struct hvc_opal_priv *pv = hvc_opal_privs[hp->vtermno];
+	int rc;
+
+	pr_devel("HVSI@%x: do open !\n", hp->vtermno);
+
+	rc = notifier_add_irq(hp, data);
+	if (rc)
+		return rc;
+
+	return hvsilib_open(&pv->hvsi, hp);
+}
+
+static void hvc_opal_hvsi_close(struct hvc_struct *hp, int data)
+{
+	struct hvc_opal_priv *pv = hvc_opal_privs[hp->vtermno];
+
+	pr_devel("HVSI@%x: do close !\n", hp->vtermno);
+
+	hvsilib_close(&pv->hvsi, hp);
+
+	notifier_del_irq(hp, data);
+}
+
+void hvc_opal_hvsi_hangup(struct hvc_struct *hp, int data)
+{
+	struct hvc_opal_priv *pv = hvc_opal_privs[hp->vtermno];
+
+	pr_devel("HVSI@%x: do hangup !\n", hp->vtermno);
+
+	hvsilib_close(&pv->hvsi, hp);
+
+	notifier_hangup_irq(hp, data);
+}
+
+static int hvc_opal_hvsi_tiocmget(struct hvc_struct *hp)
+{
+	struct hvc_opal_priv *pv = hvc_opal_privs[hp->vtermno];
+
+	if (!pv)
+		return -EINVAL;
+	return pv->hvsi.mctrl;
+}
+
+static int hvc_opal_hvsi_tiocmset(struct hvc_struct *hp, unsigned int set,
+				unsigned int clear)
+{
+	struct hvc_opal_priv *pv = hvc_opal_privs[hp->vtermno];
+
+	pr_devel("HVSI@%x: Set modem control, set=%x,clr=%x\n",
+		 hp->vtermno, set, clear);
+
+	if (set & TIOCM_DTR)
+		hvsilib_write_mctrl(&pv->hvsi, 1);
+	else if (clear & TIOCM_DTR)
+		hvsilib_write_mctrl(&pv->hvsi, 0);
+
+	return 0;
+}
+
+static const struct hv_ops hvc_opal_hvsi_ops = {
+	.get_chars = hvc_opal_hvsi_get_chars,
+	.put_chars = hvc_opal_hvsi_put_chars,
+	.notifier_add = hvc_opal_hvsi_open,
+	.notifier_del = hvc_opal_hvsi_close,
+	.notifier_hangup = hvc_opal_hvsi_hangup,
+	.tiocmget = hvc_opal_hvsi_tiocmget,
+	.tiocmset = hvc_opal_hvsi_tiocmset,
+};
+
+static int __devinit hvc_opal_probe(struct platform_device *dev)
+{
+	const struct hv_ops *ops;
+	struct hvc_struct *hp;
+	struct hvc_opal_priv *pv;
+	hv_protocol_t proto;
+	unsigned int termno, boot = 0;
+	const __be32 *reg;
+
+	if (of_device_is_compatible(dev->dev.of_node, "ibm,opal-console-raw")) {
+		proto = HV_PROTOCOL_RAW;
+		ops = &hvc_opal_raw_ops;
+	} else if (of_device_is_compatible(dev->dev.of_node,
+					   "ibm,opal-console-hvsi")) {
+		proto = HV_PROTOCOL_HVSI;
+		ops = &hvc_opal_hvsi_ops;
+	} else {
+		pr_err("hvc_opal: Unkown protocol for %s\n",
+		       dev->dev.of_node->full_name);
+		return -ENXIO;
+	}
+
+	reg = of_get_property(dev->dev.of_node, "reg", NULL);
+	termno = reg ? be32_to_cpup(reg) : 0;
+
+	/* Is it our boot one ? */
+	if (hvc_opal_privs[termno] == &hvc_opal_boot_priv) {
+		pv = hvc_opal_privs[termno];
+		boot = 1;
+	} else if (hvc_opal_privs[termno] == NULL) {
+		pv = kzalloc(sizeof(struct hvc_opal_priv), GFP_KERNEL);
+		if (!pv)
+			return -ENOMEM;
+		pv->proto = proto;
+		hvc_opal_privs[termno] = pv;
+		if (proto == HV_PROTOCOL_HVSI)
+			hvsilib_init(&pv->hvsi, opal_get_chars, opal_put_chars,
+				     termno, 0);
+
+		/* Instanciate now to establish a mapping index==vtermno */
+		hvc_instantiate(termno, termno, ops);
+	} else {
+		pr_err("hvc_opal: Device %s has duplicate terminal number #%d\n",
+		       dev->dev.of_node->full_name, termno);
+		return -ENXIO;
+	}
+
+	pr_info("hvc%d: %s protocol on %s%s\n", termno,
+		proto == HV_PROTOCOL_RAW ? "raw" : "hvsi",
+		dev->dev.of_node->full_name,
+		boot ? " (boot console)" : "");
+
+	/* We don't do IRQ yet */
+	hp = hvc_alloc(termno, 0, ops, MAX_VIO_PUT_CHARS);
+	if (IS_ERR(hp))
+		return PTR_ERR(hp);
+	dev_set_drvdata(&dev->dev, hp);
+
+	return 0;
+}
+
+static int __devexit hvc_opal_remove(struct platform_device *dev)
+{
+	struct hvc_struct *hp = dev_get_drvdata(&dev->dev);
+	int rc, termno;
+
+	termno = hp->vtermno;
+	rc = hvc_remove(hp);
+	if (rc == 0) {
+		if (hvc_opal_privs[termno] != &hvc_opal_boot_priv)
+			kfree(hvc_opal_privs[termno]);
+		hvc_opal_privs[termno] = NULL;
+	}
+	return rc;
+}
+
+static struct platform_driver hvc_opal_driver = {
+	.probe		= hvc_opal_probe,
+	.remove		= __devexit_p(hvc_opal_remove),
+	.driver		= {
+		.name	= hvc_opal_name,
+		.owner	= THIS_MODULE,
+		.of_match_table	= hvc_opal_match,
+	}
+};
+
+static int __init hvc_opal_init(void)
+{
+	if (!firmware_has_feature(FW_FEATURE_OPAL))
+		return -ENODEV;
+
+	/* Register as a vio device to receive callbacks */
+	return platform_driver_register(&hvc_opal_driver);
+}
+module_init(hvc_opal_init);
+
+static void __exit hvc_opal_exit(void)
+{
+	platform_driver_unregister(&hvc_opal_driver);
+}
+module_exit(hvc_opal_exit);
+
+static void udbg_opal_putc(char c)
+{
+	unsigned int termno = hvc_opal_boot_termno;
+	int count = -1;
+
+	if (c == '\n')
+		udbg_opal_putc('\r');
+
+	do {
+		switch(hvc_opal_boot_priv.proto) {
+		case HV_PROTOCOL_RAW:
+			count = opal_put_chars(termno, &c, 1);
+			break;
+		case HV_PROTOCOL_HVSI:
+			count = hvc_opal_hvsi_put_chars(termno, &c, 1);
+			break;
+		}
+	} while(count == 0 || count == -EAGAIN);
+}
+
+static int udbg_opal_getc_poll(void)
+{
+	unsigned int termno = hvc_opal_boot_termno;
+	int rc = 0;
+	char c;
+
+	switch(hvc_opal_boot_priv.proto) {
+	case HV_PROTOCOL_RAW:
+		rc = opal_get_chars(termno, &c, 1);
+		break;
+	case HV_PROTOCOL_HVSI:
+		rc = hvc_opal_hvsi_get_chars(termno, &c, 1);
+		break;
+	}
+	if (!rc)
+		return -1;
+	return c;
+}
+
+static int udbg_opal_getc(void)
+{
+	int ch;
+	for (;;) {
+		ch = udbg_opal_getc_poll();
+		if (ch == -1) {
+			/* This shouldn't be needed...but... */
+			volatile unsigned long delay;
+			for (delay=0; delay < 2000000; delay++)
+				;
+		} else {
+			return ch;
+		}
+	}
+}
+
+static void udbg_init_opal_common(void)
+{
+	udbg_putc = udbg_opal_putc;
+	udbg_getc = udbg_opal_getc;
+	udbg_getc_poll = udbg_opal_getc_poll;
+	tb_ticks_per_usec = 0x200; /* Make udelay not suck */
+}
+
+void __init hvc_opal_init_early(void)
+{
+	struct device_node *stdout_node = NULL;
+	const u32 *termno;
+	const char *name = NULL;
+	const struct hv_ops *ops;
+	u32 index;
+
+	/* find the boot console from /chosen/stdout */
+	if (of_chosen)
+		name = of_get_property(of_chosen, "linux,stdout-path", NULL);
+	if (name) {
+		stdout_node = of_find_node_by_path(name);
+		if (!stdout_node) {
+			pr_err("hvc_opal: Failed to locate default console!\n");
+			return;
+		}
+	} else {
+		struct device_node *opal, *np;
+
+		/* Current OPAL takeover doesn't provide the stdout
+		 * path, so we hard wire it
+		 */
+		opal = of_find_node_by_path("/ibm,opal/consoles");
+		if (opal)
+			pr_devel("hvc_opal: Found consoles in new location\n");
+		if (!opal) {
+			opal = of_find_node_by_path("/ibm,opal");
+			if (opal)
+				pr_devel("hvc_opal: "
+					 "Found consoles in old location\n");
+		}
+		if (!opal)
+			return;
+		for_each_child_of_node(opal, np) {
+			if (!strcmp(np->name, "serial")) {
+				stdout_node = np;
+				break;
+			}
+		}
+		of_node_put(opal);
+	}
+	if (!stdout_node)
+		return;
+	termno = of_get_property(stdout_node, "reg", NULL);
+	index = termno ? *termno : 0;
+	if (index >= MAX_NR_HVC_CONSOLES)
+		return;
+	hvc_opal_privs[index] = &hvc_opal_boot_priv;
+
+	/* Check the protocol */
+	if (of_device_is_compatible(stdout_node, "ibm,opal-console-raw")) {
+		hvc_opal_boot_priv.proto = HV_PROTOCOL_RAW;
+		ops = &hvc_opal_raw_ops;
+		pr_devel("hvc_opal: Found RAW console\n");
+	}
+	else if (of_device_is_compatible(stdout_node,"ibm,opal-console-hvsi")) {
+		hvc_opal_boot_priv.proto = HV_PROTOCOL_HVSI;
+		ops = &hvc_opal_hvsi_ops;
+		hvsilib_init(&hvc_opal_boot_priv.hvsi, opal_get_chars,
+			     opal_put_chars, index, 1);
+		/* HVSI, perform the handshake now */
+		hvsilib_establish(&hvc_opal_boot_priv.hvsi);
+		pr_devel("hvc_opal: Found HVSI console\n");
+	} else
+		goto out;
+	hvc_opal_boot_termno = index;
+	udbg_init_opal_common();
+	add_preferred_console("hvc", index, NULL);
+	hvc_instantiate(index, index, ops);
+out:
+	of_node_put(stdout_node);
+}
+
+#ifdef CONFIG_PPC_EARLY_DEBUG_OPAL_RAW
+void __init udbg_init_debug_opal(void)
+{
+	u32 index = CONFIG_PPC_EARLY_DEBUG_OPAL_VTERMNO;
+	hvc_opal_privs[index] = &hvc_opal_boot_priv;
+	hvc_opal_boot_priv.proto = HV_PROTOCOL_RAW;
+	hvc_opal_boot_termno = index;
+	udbg_init_opal_common();
+}
+#endif /* CONFIG_PPC_EARLY_DEBUG_OPAL_RAW */
+
+#ifdef CONFIG_PPC_EARLY_DEBUG_OPAL_HVSI
+void __init udbg_init_debug_opal_hvsi(void)
+{
+	u32 index = CONFIG_PPC_EARLY_DEBUG_OPAL_VTERMNO;
+	hvc_opal_privs[index] = &hvc_opal_boot_priv;
+	hvc_opal_boot_termno = index;
+	udbg_init_opal_common();
+	hvsilib_init(&hvc_opal_boot_priv.hvsi, opal_get_chars, opal_put_chars,
+		     index, 1);
+	hvsilib_establish(&hvc_opal_boot_priv.hvsi);
+}
+#endif /* CONFIG_PPC_EARLY_DEBUG_OPAL_HVSI */
diff --git a/drivers/tty/hvc/hvsi_lib.c b/drivers/tty/hvc/hvsi_lib.c
index 08e3e90..215df5b 100644
--- a/drivers/tty/hvc/hvsi_lib.c
+++ b/drivers/tty/hvc/hvsi_lib.c
@@ -183,7 +183,7 @@ int hvsilib_get_chars(struct hvsi_priv *pv, char *buf, int count)
 	unsigned int tries, read = 0;
 
 	if (WARN_ON(!pv))
-		return 0;
+		return -ENXIO;
 
 	/* If we aren't open, don't do anything in order to avoid races
 	 * with connection establishment. The hvc core will call this
@@ -234,7 +234,7 @@ int hvsilib_put_chars(struct hvsi_priv *pv, const char *buf, int count)
 	int rc, adjcount = min(count, HVSI_MAX_OUTGOING_DATA);
 
 	if (WARN_ON(!pv))
-		return 0;
+		return -ENODEV;
 
 	dp.hdr.type = VS_DATA_PACKET_HEADER;
 	dp.hdr.len = adjcount + sizeof(struct hvsi_header);
-- 
1.7.4.1

[PATCH 14/21] powerpc/powernv: Hookup reboot and poweroff functions

[Benjamin Herrenschmidt]

This calls the respective HAL functions, and spin on hal_poll_event()
to ensure the HAL has a chance to communicate with the FSP to trigger
the reboot or shutdown operation

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/platforms/powernv/setup.c |   32 ++++++++++++++++++++++++++------
 1 files changed, 26 insertions(+), 6 deletions(-)

diff --git a/arch/powerpc/platforms/powernv/setup.c b/arch/powerpc/platforms/powernv/setup.c
index 07ba1ec..0fac0a6 100644
--- a/arch/powerpc/platforms/powernv/setup.c
+++ b/arch/powerpc/platforms/powernv/setup.c
@@ -83,19 +83,39 @@ static void pnv_show_cpuinfo(struct seq_file *m)
 	of_node_put(root);
 }
 
-static void pnv_restart(char *cmd)
+static void  __noreturn pnv_restart(char *cmd)
 {
-	for (;;);
+	long rc = OPAL_BUSY;
+
+	while (rc == OPAL_BUSY || rc == OPAL_BUSY_EVENT) {
+		rc = opal_cec_reboot();
+		if (rc == OPAL_BUSY_EVENT)
+			opal_poll_events(NULL);
+		else
+			mdelay(10);
+	}
+	for (;;)
+		opal_poll_events(NULL);
 }
 
-static void pnv_power_off(void)
+static void __noreturn pnv_power_off(void)
 {
-	for (;;);
+	long rc = OPAL_BUSY;
+
+	while (rc == OPAL_BUSY || rc == OPAL_BUSY_EVENT) {
+		rc = opal_cec_power_down(0);
+		if (rc == OPAL_BUSY_EVENT)
+			opal_poll_events(NULL);
+		else
+			mdelay(10);
+	}
+	for (;;)
+		opal_poll_events(NULL);
 }
 
-static void pnv_halt(void)
+static void __noreturn pnv_halt(void)
 {
-	for (;;);
+	pnv_power_off();
 }
 
 static unsigned long __init pnv_get_boot_time(void)
-- 
1.7.4.1

[PATCH 15/21] powerpc/powernv: Add RTC and NVRAM support plus RTAS fallbacks

[Benjamin Herrenschmidt]

Implements OPAL RTC and NVRAM support and wire all that up to
the powernv platform.

We use RTAS for RTC as a fallback if available. Using RTAS for nvram
is not supported yet, pending some rework/cleanup and generalization
of the pSeries & CHRP code. We also use RTAS fallbacks for power off
and reboot

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/include/asm/opal.h             |    6 ++
 arch/powerpc/platforms/powernv/Makefile     |    2 +
 arch/powerpc/platforms/powernv/opal-nvram.c |   88 ++++++++++++++++++++++++
 arch/powerpc/platforms/powernv/opal-rtc.c   |   97 +++++++++++++++++++++++++++
 arch/powerpc/platforms/powernv/setup.c      |   57 ++++++++++------
 5 files changed, 229 insertions(+), 21 deletions(-)
 create mode 100644 arch/powerpc/platforms/powernv/opal-nvram.c
 create mode 100644 arch/powerpc/platforms/powernv/opal-rtc.c

diff --git a/arch/powerpc/include/asm/opal.h b/arch/powerpc/include/asm/opal.h
index dbeabea..9bb0efd 100644
--- a/arch/powerpc/include/asm/opal.h
+++ b/arch/powerpc/include/asm/opal.h
@@ -430,6 +430,12 @@ extern int opal_put_chars(uint32_t vtermno, const char *buf, int total_len);
 
 extern void hvc_opal_init_early(void);
 
+struct rtc_time;
+extern int opal_set_rtc_time(struct rtc_time *tm);
+extern void opal_get_rtc_time(struct rtc_time *tm);
+extern unsigned long opal_get_boot_time(void);
+extern void opal_nvram_init(void);
+
 #endif /* __ASSEMBLY__ */
 
 #endif /* __OPAL_H */
diff --git a/arch/powerpc/platforms/powernv/Makefile b/arch/powerpc/platforms/powernv/Makefile
index 8f69c0d..618ad83 100644
--- a/arch/powerpc/platforms/powernv/Makefile
+++ b/arch/powerpc/platforms/powernv/Makefile
@@ -1,2 +1,4 @@
 obj-y			+= setup.o opal-takeover.o opal-wrappers.o opal.o
+obj-y			+= opal-rtc.o opal-nvram.o
+
 obj-$(CONFIG_SMP)	+= smp.o
diff --git a/arch/powerpc/platforms/powernv/opal-nvram.c b/arch/powerpc/platforms/powernv/opal-nvram.c
new file mode 100644
index 0000000..3f83e1a
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/opal-nvram.c
@@ -0,0 +1,88 @@
+/*
+ * PowerNV nvram code.
+ *
+ * Copyright 2011 IBM Corp.
+ *
+ * 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 DEBUG
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/of.h>
+
+#include <asm/opal.h>
+#include <asm/machdep.h>
+
+static unsigned int nvram_size;
+
+static ssize_t opal_nvram_size(void)
+{
+	return nvram_size;
+}
+
+static ssize_t opal_nvram_read(char *buf, size_t count, loff_t *index)
+{
+	s64 rc;
+	int off;
+
+	if (*index >= nvram_size)
+		return 0;
+	off = *index;
+	if ((off + count) > nvram_size)
+		count = nvram_size - off;
+	rc = opal_read_nvram(__pa(buf), count, off);
+	if (rc != OPAL_SUCCESS)
+		return -EIO;
+	*index += count;
+	return count;
+}
+
+static ssize_t opal_nvram_write(char *buf, size_t count, loff_t *index)
+{
+	s64 rc = OPAL_BUSY;
+	int off;
+
+	if (*index >= nvram_size)
+		return 0;
+	off = *index;
+	if ((off + count) > nvram_size)
+		count = nvram_size - off;
+
+	while (rc == OPAL_BUSY || rc == OPAL_BUSY_EVENT) {
+		rc = opal_write_nvram(__pa(buf), count, off);
+		if (rc == OPAL_BUSY_EVENT)
+			opal_poll_events(NULL);
+	}
+	*index += count;
+	return count;
+}
+
+void __init opal_nvram_init(void)
+{
+	struct device_node *np;
+	const u32 *nbytes_p;
+
+	np = of_find_compatible_node(NULL, NULL, "ibm,opal-nvram");
+	if (np == NULL)
+		return;
+
+	nbytes_p = of_get_property(np, "#bytes", NULL);
+	if (!nbytes_p) {
+		of_node_put(np);
+		return;
+	}
+	nvram_size = *nbytes_p;
+
+	printk(KERN_INFO "OPAL nvram setup, %u bytes\n", nvram_size);
+	of_node_put(np);
+
+	ppc_md.nvram_read = opal_nvram_read;
+	ppc_md.nvram_write = opal_nvram_write;
+	ppc_md.nvram_size = opal_nvram_size;
+}
+
diff --git a/arch/powerpc/platforms/powernv/opal-rtc.c b/arch/powerpc/platforms/powernv/opal-rtc.c
new file mode 100644
index 0000000..2aa7641
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/opal-rtc.c
@@ -0,0 +1,97 @@
+/*
+ * PowerNV Real Time Clock.
+ *
+ * Copyright 2011 IBM Corp.
+ *
+ * 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.
+ */
+
+
+#include <linux/kernel.h>
+#include <linux/time.h>
+#include <linux/bcd.h>
+#include <linux/rtc.h>
+#include <linux/delay.h>
+
+#include <asm/opal.h>
+#include <asm/firmware.h>
+
+static void opal_to_tm(u32 y_m_d, u64 h_m_s_ms, struct rtc_time *tm)
+{
+	tm->tm_year	= ((bcd2bin(y_m_d >> 24) * 100) +
+			   bcd2bin((y_m_d >> 16) & 0xff)) - 1900;
+	tm->tm_mon	= bcd2bin((y_m_d >> 8) & 0xff) - 1;
+	tm->tm_mday	= bcd2bin(y_m_d & 0xff);
+	tm->tm_hour	= bcd2bin((h_m_s_ms >> 56) & 0xff);
+	tm->tm_min	= bcd2bin((h_m_s_ms >> 48) & 0xff);
+	tm->tm_sec	= bcd2bin((h_m_s_ms >> 40) & 0xff);
+
+        GregorianDay(tm);
+}
+
+unsigned long __init opal_get_boot_time(void)
+{
+	struct rtc_time tm;
+	u32 y_m_d;
+	u64 h_m_s_ms;
+	long rc = OPAL_BUSY;
+
+	while (rc == OPAL_BUSY || rc == OPAL_BUSY_EVENT) {
+		rc = opal_rtc_read(&y_m_d, &h_m_s_ms);
+		if (rc == OPAL_BUSY_EVENT)
+			opal_poll_events(NULL);
+		else
+			mdelay(10);
+	}
+	if (rc != OPAL_SUCCESS)
+		return 0;
+	opal_to_tm(y_m_d, h_m_s_ms, &tm);
+	return mktime(tm.tm_year + 1900, tm.tm_mon + 1, tm.tm_mday,
+		      tm.tm_hour, tm.tm_min, tm.tm_sec);
+}
+
+void opal_get_rtc_time(struct rtc_time *tm)
+{
+	long rc = OPAL_BUSY;
+	u32 y_m_d;
+	u64 h_m_s_ms;
+
+	while (rc == OPAL_BUSY || rc == OPAL_BUSY_EVENT) {
+		rc = opal_rtc_read(&y_m_d, &h_m_s_ms);
+		if (rc == OPAL_BUSY_EVENT)
+			opal_poll_events(NULL);
+		else
+			mdelay(10);
+	}
+	if (rc != OPAL_SUCCESS)
+		return;
+	opal_to_tm(y_m_d, h_m_s_ms, tm);
+}
+
+int opal_set_rtc_time(struct rtc_time *tm)
+{
+	long rc = OPAL_BUSY;
+	u32 y_m_d = 0;
+	u64 h_m_s_ms = 0;
+
+	y_m_d |= ((u32)bin2bcd((tm->tm_year + 1900) / 100)) << 24;
+	y_m_d |= ((u32)bin2bcd((tm->tm_year + 1900) % 100)) << 16;
+	y_m_d |= ((u32)bin2bcd((tm->tm_mon + 1))) << 8;
+	y_m_d |= ((u32)bin2bcd(tm->tm_mday));
+
+	h_m_s_ms |= ((u64)bin2bcd(tm->tm_hour)) << 56;
+	h_m_s_ms |= ((u64)bin2bcd(tm->tm_min)) << 48;
+	h_m_s_ms |= ((u64)bin2bcd(tm->tm_sec)) << 40;
+
+	while (rc == OPAL_BUSY || rc == OPAL_BUSY_EVENT) {
+		rc = opal_rtc_write(y_m_d, h_m_s_ms);
+		if (rc == OPAL_BUSY_EVENT)
+			opal_poll_events(NULL);
+		else
+			mdelay(10);
+	}
+	return rc == OPAL_SUCCESS ? 0 : -EIO;
+}
diff --git a/arch/powerpc/platforms/powernv/setup.c b/arch/powerpc/platforms/powernv/setup.c
index 0fac0a6..4a2b2e2 100644
--- a/arch/powerpc/platforms/powernv/setup.c
+++ b/arch/powerpc/platforms/powernv/setup.c
@@ -29,7 +29,9 @@
 #include <asm/machdep.h>
 #include <asm/firmware.h>
 #include <asm/xics.h>
+#include <asm/rtas.h>
 #include <asm/opal.h>
+#include <asm/xics.h>
 
 #include "powernv.h"
 
@@ -40,7 +42,9 @@ static void __init pnv_setup_arch(void)
 
 	/* XXX PCI */
 
-	/* XXX NVRAM */
+	/* Setup RTC and NVRAM callbacks */
+	if (firmware_has_feature(FW_FEATURE_OPAL))
+		opal_nvram_init();
 
 	/* Enable NAP mode */
 	powersave_nap = 1;
@@ -118,30 +122,40 @@ static void __noreturn pnv_halt(void)
 	pnv_power_off();
 }
 
-static unsigned long __init pnv_get_boot_time(void)
-{
-	return 0;
-}
-
-static void pnv_get_rtc_time(struct rtc_time *rtc_tm)
+static void pnv_progress(char *s, unsigned short hex)
 {
 }
 
-static int pnv_set_rtc_time(struct rtc_time *tm)
+#ifdef CONFIG_KEXEC
+static void pnv_kexec_cpu_down(int crash_shutdown, int secondary)
 {
-	return 0;
+	xics_kexec_teardown_cpu(secondary);
 }
+#endif /* CONFIG_KEXEC */
 
-static void pnv_progress(char *s, unsigned short hex)
+static void __init pnv_setup_machdep_opal(void)
 {
+	ppc_md.get_boot_time = opal_get_boot_time;
+	ppc_md.get_rtc_time = opal_get_rtc_time;
+	ppc_md.set_rtc_time = opal_set_rtc_time;
+	ppc_md.restart = pnv_restart;
+	ppc_md.power_off = pnv_power_off;
+	ppc_md.halt = pnv_halt;
 }
 
-#ifdef CONFIG_KEXEC
-static void pnv_kexec_cpu_down(int crash_shutdown, int secondary)
+#ifdef CONFIG_PPC_POWERNV_RTAS
+static void __init pnv_setup_machdep_rtas(void)
 {
-	xics_kexec_teardown_cpu(secondary);
+	if (rtas_token("get-time-of-day") != RTAS_UNKNOWN_SERVICE) {
+		ppc_md.get_boot_time = rtas_get_boot_time;
+		ppc_md.get_rtc_time = rtas_get_rtc_time;
+		ppc_md.set_rtc_time = rtas_set_rtc_time;
+	}
+	ppc_md.restart = rtas_restart;
+	ppc_md.power_off = rtas_power_off;
+	ppc_md.halt = rtas_halt;
 }
-#endif /* CONFIG_KEXEC */
+#endif /* CONFIG_PPC_POWERNV_RTAS */
 
 static int __init pnv_probe(void)
 {
@@ -152,6 +166,13 @@ static int __init pnv_probe(void)
 
 	hpte_init_native();
 
+	if (firmware_has_feature(FW_FEATURE_OPAL))
+		pnv_setup_machdep_opal();
+#ifdef CONFIG_PPC_POWERNV_RTAS
+	else if (rtas.base)
+		pnv_setup_machdep_rtas();
+#endif /* CONFIG_PPC_POWERNV_RTAS */
+
 	pr_debug("PowerNV detected !\n");
 
 	return 1;
@@ -160,16 +181,10 @@ static int __init pnv_probe(void)
 define_machine(powernv) {
 	.name			= "PowerNV",
 	.probe			= pnv_probe,
-	.setup_arch		= pnv_setup_arch,
 	.init_early		= pnv_init_early,
+	.setup_arch		= pnv_setup_arch,
 	.init_IRQ		= pnv_init_IRQ,
 	.show_cpuinfo		= pnv_show_cpuinfo,
-	.restart		= pnv_restart,
-	.power_off		= pnv_power_off,
-	.halt			= pnv_halt,
-	.get_boot_time		= pnv_get_boot_time,
-	.get_rtc_time		= pnv_get_rtc_time,
-	.set_rtc_time		= pnv_set_rtc_time,
 	.progress		= pnv_progress,
 	.power_save             = power7_idle,
 	.calibrate_decr		= generic_calibrate_decr,
-- 
1.7.4.1

[PATCH 16/21] powerpc/powernv: Add OPAL ICS backend

[Benjamin Herrenschmidt]

OPAL handles HW access to the various ICS or equivalent chips
for us (with the exception of p5ioc2 based HEA which uses a
different backend) similarily to what RTAS does on pSeries.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/include/asm/xics.h        |   19 +++
 arch/powerpc/sysdev/xics/Makefile      |    1 +
 arch/powerpc/sysdev/xics/ics-opal.c    |  244 ++++++++++++++++++++++++++++++++
 arch/powerpc/sysdev/xics/xics-common.c |    8 +-
 4 files changed, 266 insertions(+), 6 deletions(-)
 create mode 100644 arch/powerpc/sysdev/xics/ics-opal.c

diff --git a/arch/powerpc/include/asm/xics.h b/arch/powerpc/include/asm/xics.h
index b183a40..bd6c401 100644
--- a/arch/powerpc/include/asm/xics.h
+++ b/arch/powerpc/include/asm/xics.h
@@ -27,10 +27,18 @@
 #define MAX_NUM_PRIORITIES	3
 
 /* Native ICP */
+#ifdef CONFIG_PPC_ICP_NATIVE
 extern int icp_native_init(void);
+#else
+static inline int icp_native_init(void) { return -ENODEV; }
+#endif
 
 /* PAPR ICP */
+#ifdef CONFIG_PPC_ICP_HV
 extern int icp_hv_init(void);
+#else
+static inline int icp_hv_init(void) { return -ENODEV; }
+#endif
 
 /* ICP ops */
 struct icp_ops {
@@ -51,7 +59,18 @@ extern const struct icp_ops *icp_ops;
 extern int ics_native_init(void);
 
 /* RTAS ICS */
+#ifdef CONFIG_PPC_ICS_RTAS
 extern int ics_rtas_init(void);
+#else
+static inline int ics_rtas_init(void) { return -ENODEV; }
+#endif
+
+/* HAL ICS */
+#ifdef CONFIG_PPC_POWERNV
+extern int ics_opal_init(void);
+#else
+static inline int ics_opal_init(void) { return -ENODEV; }
+#endif
 
 /* ICS instance, hooked up to chip_data of an irq */
 struct ics {
diff --git a/arch/powerpc/sysdev/xics/Makefile b/arch/powerpc/sysdev/xics/Makefile
index b75a605..c606aa8 100644
--- a/arch/powerpc/sysdev/xics/Makefile
+++ b/arch/powerpc/sysdev/xics/Makefile
@@ -4,3 +4,4 @@ obj-y				+= xics-common.o
 obj-$(CONFIG_PPC_ICP_NATIVE)	+= icp-native.o
 obj-$(CONFIG_PPC_ICP_HV)	+= icp-hv.o
 obj-$(CONFIG_PPC_ICS_RTAS)	+= ics-rtas.o
+obj-$(CONFIG_PPC_POWERNV)	+= ics-opal.o
diff --git a/arch/powerpc/sysdev/xics/ics-opal.c b/arch/powerpc/sysdev/xics/ics-opal.c
new file mode 100644
index 0000000..1f3a676
--- /dev/null
+++ b/arch/powerpc/sysdev/xics/ics-opal.c
@@ -0,0 +1,244 @@
+/*
+ * ICS backend for OPAL managed interrupts.
+ *
+ * Copyright 2011 IBM Corp.
+ *
+ * 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.
+ */
+
+#undef DEBUG
+
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/irq.h>
+#include <linux/smp.h>
+#include <linux/interrupt.h>
+#include <linux/init.h>
+#include <linux/cpu.h>
+#include <linux/of.h>
+#include <linux/spinlock.h>
+#include <linux/msi.h>
+
+#include <asm/prom.h>
+#include <asm/smp.h>
+#include <asm/machdep.h>
+#include <asm/irq.h>
+#include <asm/errno.h>
+#include <asm/xics.h>
+#include <asm/opal.h>
+#include <asm/firmware.h>
+
+static int ics_opal_mangle_server(int server)
+{
+	/* No link for now */
+	return server << 2;
+}
+
+static int ics_opal_unmangle_server(int server)
+{
+	/* No link for now */
+	return server >> 2;
+}
+
+static void ics_opal_unmask_irq(struct irq_data *d)
+{
+	unsigned int hw_irq = (unsigned int)irqd_to_hwirq(d);
+	int64_t rc;
+	int server;
+
+	pr_devel("ics-hal: unmask virq %d [hw 0x%x]\n", d->irq, hw_irq);
+
+	if (hw_irq == XICS_IPI || hw_irq == XICS_IRQ_SPURIOUS)
+		return;
+
+	server = xics_get_irq_server(d->irq, d->affinity, 0);
+	server = ics_opal_mangle_server(server);
+
+	rc = opal_set_xive(hw_irq, server, DEFAULT_PRIORITY);
+	if (rc != OPAL_SUCCESS)
+		pr_err("%s: opal_set_xive(irq=%d [hw 0x%x] server=%x)"
+		       " error %lld\n",
+		       __func__, d->irq, hw_irq, server, rc);
+}
+
+static unsigned int ics_opal_startup(struct irq_data *d)
+{
+#ifdef CONFIG_PCI_MSI
+	/*
+	 * The generic MSI code returns with the interrupt disabled on the
+	 * card, using the MSI mask bits. Firmware doesn't appear to unmask
+	 * at that level, so we do it here by hand.
+	 */
+	if (d->msi_desc)
+		unmask_msi_irq(d);
+#endif
+
+	/* unmask it */
+	ics_opal_unmask_irq(d);
+	return 0;
+}
+
+static void ics_opal_mask_real_irq(unsigned int hw_irq)
+{
+	int server = ics_opal_mangle_server(xics_default_server);
+	int64_t rc;
+
+	if (hw_irq == XICS_IPI)
+		return;
+
+	/* Have to set XIVE to 0xff to be able to remove a slot */
+	rc = opal_set_xive(hw_irq, server, 0xff);
+	if (rc != OPAL_SUCCESS)
+		pr_err("%s: opal_set_xive(0xff) irq=%u returned %lld\n",
+		       __func__, hw_irq, rc);
+}
+
+static void ics_opal_mask_irq(struct irq_data *d)
+{
+	unsigned int hw_irq = (unsigned int)irqd_to_hwirq(d);
+
+	pr_devel("ics-hal: mask virq %d [hw 0x%x]\n", d->irq, hw_irq);
+
+	if (hw_irq == XICS_IPI || hw_irq == XICS_IRQ_SPURIOUS)
+		return;
+	ics_opal_mask_real_irq(hw_irq);
+}
+
+static int ics_opal_set_affinity(struct irq_data *d,
+				 const struct cpumask *cpumask,
+				 bool force)
+{
+	unsigned int hw_irq = (unsigned int)irqd_to_hwirq(d);
+	int16_t server;
+	int8_t priority;
+	int64_t rc;
+	int wanted_server;
+
+	if (hw_irq == XICS_IPI || hw_irq == XICS_IRQ_SPURIOUS)
+		return -1;
+
+	rc = opal_get_xive(hw_irq, &server, &priority);
+	if (rc != OPAL_SUCCESS) {
+		pr_err("%s: opal_set_xive(irq=%d [hw 0x%x] server=%x)"
+		       " error %lld\n",
+		       __func__, d->irq, hw_irq, server, rc);
+		return -1;
+	}
+
+	wanted_server = xics_get_irq_server(d->irq, cpumask, 1);
+	if (wanted_server < 0) {
+		char cpulist[128];
+		cpumask_scnprintf(cpulist, sizeof(cpulist), cpumask);
+		pr_warning("%s: No online cpus in the mask %s for irq %d\n",
+			   __func__, cpulist, d->irq);
+		return -1;
+	}
+	server = ics_opal_mangle_server(wanted_server);
+
+	pr_devel("ics-hal: set-affinity irq %d [hw 0x%x] server: 0x%x/0x%x\n",
+		 d->irq, hw_irq, wanted_server, server);
+
+	rc = opal_set_xive(hw_irq, server, priority);
+	if (rc != OPAL_SUCCESS) {
+		pr_err("%s: opal_set_xive(irq=%d [hw 0x%x] server=%x)"
+		       " error %lld\n",
+		       __func__, d->irq, hw_irq, server, rc);
+		return -1;
+	}
+	return 0;
+}
+
+static struct irq_chip ics_opal_irq_chip = {
+	.name = "HAL ICS",
+	.irq_startup = ics_opal_startup,
+	.irq_mask = ics_opal_mask_irq,
+	.irq_unmask = ics_opal_unmask_irq,
+	.irq_eoi = NULL, /* Patched at init time */
+	.irq_set_affinity = ics_opal_set_affinity
+};
+
+static int ics_opal_map(struct ics *ics, unsigned int virq);
+static void ics_opal_mask_unknown(struct ics *ics, unsigned long vec);
+static long ics_opal_get_server(struct ics *ics, unsigned long vec);
+
+static int ics_opal_host_match(struct ics *ics, struct device_node *node)
+{
+	return 1;
+}
+
+/* Only one global & state struct ics */
+static struct ics ics_hal = {
+	.map		= ics_opal_map,
+	.mask_unknown	= ics_opal_mask_unknown,
+	.get_server	= ics_opal_get_server,
+	.host_match	= ics_opal_host_match,
+};
+
+static int ics_opal_map(struct ics *ics, unsigned int virq)
+{
+	unsigned int hw_irq = (unsigned int)virq_to_hw(virq);
+	int64_t rc;
+	int16_t server;
+	int8_t priority;
+
+	if (WARN_ON(hw_irq == XICS_IPI || hw_irq == XICS_IRQ_SPURIOUS))
+		return -EINVAL;
+
+	/* Check if HAL knows about this interrupt */
+	rc = opal_get_xive(hw_irq, &server, &priority);
+	if (rc != OPAL_SUCCESS)
+		return -ENXIO;
+
+	irq_set_chip_and_handler(virq, &ics_opal_irq_chip, handle_fasteoi_irq);
+	irq_set_chip_data(virq, &ics_hal);
+
+	return 0;
+}
+
+static void ics_opal_mask_unknown(struct ics *ics, unsigned long vec)
+{
+	int64_t rc;
+	int16_t server;
+	int8_t priority;
+
+	/* Check if HAL knows about this interrupt */
+	rc = opal_get_xive(vec, &server, &priority);
+	if (rc != OPAL_SUCCESS)
+		return;
+
+	ics_opal_mask_real_irq(vec);
+}
+
+static long ics_opal_get_server(struct ics *ics, unsigned long vec)
+{
+	int64_t rc;
+	int16_t server;
+	int8_t priority;
+
+	/* Check if HAL knows about this interrupt */
+	rc = opal_get_xive(vec, &server, &priority);
+	if (rc != OPAL_SUCCESS)
+		return -1;
+	return ics_opal_unmangle_server(server);
+}
+
+int __init ics_opal_init(void)
+{
+	if (!firmware_has_feature(FW_FEATURE_OPAL))
+		return -ENODEV;
+
+	/* We need to patch our irq chip's EOI to point to the
+	 * right ICP
+	 */
+	ics_opal_irq_chip.irq_eoi = icp_ops->eoi;
+
+	/* Register ourselves */
+	xics_register_ics(&ics_hal);
+
+	pr_info("ICS OPAL backend registered\n");
+
+	return 0;
+}
diff --git a/arch/powerpc/sysdev/xics/xics-common.c b/arch/powerpc/sysdev/xics/xics-common.c
index 445c5a0..3d93a8d 100644
--- a/arch/powerpc/sysdev/xics/xics-common.c
+++ b/arch/powerpc/sysdev/xics/xics-common.c
@@ -409,14 +409,10 @@ void __init xics_init(void)
 	int rc = -1;
 
 	/* Fist locate ICP */
-#ifdef CONFIG_PPC_ICP_HV
 	if (firmware_has_feature(FW_FEATURE_LPAR))
 		rc = icp_hv_init();
-#endif
-#ifdef CONFIG_PPC_ICP_NATIVE
 	if (rc < 0)
 		rc = icp_native_init();
-#endif
 	if (rc < 0) {
 		pr_warning("XICS: Cannot find a Presentation Controller !\n");
 		return;
@@ -429,9 +425,9 @@ void __init xics_init(void)
 	xics_ipi_chip.irq_eoi = icp_ops->eoi;
 
 	/* Now locate ICS */
-#ifdef CONFIG_PPC_ICS_RTAS
 	rc = ics_rtas_init();
-#endif
+	if (rc < 0)
+		rc = ics_opal_init();
 	if (rc < 0)
 		pr_warning("XICS: Cannot find a Source Controller !\n");
 
-- 
1.7.4.1

[PATCH 17/21] powerpc/powernv: Register and handle OPAL interrupts

[Benjamin Herrenschmidt]

We do the minimum which is to "pass" interrupts to HAL, which
makes the console smoother and will allow us to implement
interrupt based completion and console.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/platforms/powernv/opal.c |   31 +++++++++++++++++++++++++++++++
 1 files changed, 31 insertions(+), 0 deletions(-)

diff --git a/arch/powerpc/platforms/powernv/opal.c b/arch/powerpc/platforms/powernv/opal.c
index 7887733..5a598ca 100644
--- a/arch/powerpc/platforms/powernv/opal.c
+++ b/arch/powerpc/platforms/powernv/opal.c
@@ -14,6 +14,7 @@
 #include <linux/types.h>
 #include <linux/of.h>
 #include <linux/of_platform.h>
+#include <linux/interrupt.h>
 #include <asm/opal.h>
 #include <asm/firmware.h>
 
@@ -135,9 +136,22 @@ int opal_put_chars(uint32_t vtermno, const char *data, int total_len)
 	return written;
 }
 
+static irqreturn_t opal_interrupt(int irq, void *data)
+{
+	uint64_t events;
+
+	opal_handle_interrupt(virq_to_hw(irq), &events);
+
+	/* XXX TODO: Do something with the events */
+
+	return IRQ_HANDLED;
+}
+
 static int __init opal_init(void)
 {
 	struct device_node *np, *consoles;
+	const u32 *irqs;
+	int rc, i, irqlen;
 
 	opal_node = of_find_node_by_path("/ibm,opal");
 	if (!opal_node) {
@@ -156,6 +170,23 @@ static int __init opal_init(void)
 		of_platform_device_create(np, NULL, NULL);
 	}
 	of_node_put(consoles);
+
+	/* Find all OPAL interrupts and request them */
+	irqs = of_get_property(opal_node, "opal-interrupts", &irqlen);
+	pr_debug("opal: Found %d interrupts reserved for OPAL\n",
+		 irqs ? (irqlen / 4) : 0);
+	for (i = 0; irqs && i < (irqlen / 4); i++, irqs++) {
+		unsigned int hwirq = be32_to_cpup(irqs);
+		unsigned int irq = irq_create_mapping(NULL, hwirq);
+		if (irq == NO_IRQ) {
+			pr_warning("opal: Failed to map irq 0x%x\n", hwirq);
+			continue;
+		}
+		rc = request_irq(irq, opal_interrupt, 0, "opal", NULL);
+		if (rc)
+			pr_warning("opal: Error %d requesting irq %d"
+				   " (0x%x)\n", rc, irq, hwirq);
+	}
 	return 0;
 }
 subsys_initcall(opal_init);
-- 
1.7.4.1

[PATCH 18/21] powerpc/powernv: Machine check and other system interrupts

[Benjamin Herrenschmidt]

OPAL can handle various interrupt for us such as Machine Checks (it
performs all sorts of recovery tasks and passes back control to us with
informations about the error), Hardware Management Interrupts and Softpatch
interrupts.

This wires up the mechanisms and prints out specific informations returned
by HAL when a machine check occurs.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/include/asm/opal.h        |    2 +
 arch/powerpc/include/asm/paca.h        |    8 ++
 arch/powerpc/kernel/asm-offsets.c      |   10 +++
 arch/powerpc/kernel/exceptions-64s.S   |   27 ++++++-
 arch/powerpc/platforms/powernv/opal.c  |  130 ++++++++++++++++++++++++++++++++
 arch/powerpc/platforms/powernv/setup.c |    1 +
 6 files changed, 174 insertions(+), 4 deletions(-)

diff --git a/arch/powerpc/include/asm/opal.h b/arch/powerpc/include/asm/opal.h
index 9bb0efd..fca611b 100644
--- a/arch/powerpc/include/asm/opal.h
+++ b/arch/powerpc/include/asm/opal.h
@@ -436,6 +436,8 @@ extern void opal_get_rtc_time(struct rtc_time *tm);
 extern unsigned long opal_get_boot_time(void);
 extern void opal_nvram_init(void);
 
+extern int opal_machine_check(struct pt_regs *regs);
+
 #endif /* __ASSEMBLY__ */
 
 #endif /* __OPAL_H */
diff --git a/arch/powerpc/include/asm/paca.h b/arch/powerpc/include/asm/paca.h
index 516bfb3..17722c7 100644
--- a/arch/powerpc/include/asm/paca.h
+++ b/arch/powerpc/include/asm/paca.h
@@ -43,6 +43,7 @@ extern unsigned int debug_smp_processor_id(void); /* from linux/smp.h */
 #define get_slb_shadow()	(get_paca()->slb_shadow_ptr)
 
 struct task_struct;
+struct opal_machine_check_event;
 
 /*
  * Defines the layout of the paca.
@@ -135,6 +136,13 @@ struct paca_struct {
 	u8 io_sync;			/* writel() needs spin_unlock sync */
 	u8 irq_work_pending;		/* IRQ_WORK interrupt while soft-disable */
 
+#ifdef CONFIG_PPC_POWERNV
+	/* Pointer to OPAL machine check event structure set by the
+	 * early exception handler for use by high level C handler
+	 */
+	struct opal_machine_check_event *opal_mc_evt;
+#endif
+
 	/* Stuff for accurate time accounting */
 	u64 user_time;			/* accumulated usermode TB ticks */
 	u64 system_time;		/* accumulated system TB ticks */
diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c
index 5f078bc..536ffa8 100644
--- a/arch/powerpc/kernel/asm-offsets.c
+++ b/arch/powerpc/kernel/asm-offsets.c
@@ -48,6 +48,9 @@
 #ifdef CONFIG_PPC_ISERIES
 #include <asm/iseries/alpaca.h>
 #endif
+#ifdef CONFIG_PPC_POWERNV
+#include <asm/opal.h>
+#endif
 #if defined(CONFIG_KVM) || defined(CONFIG_KVM_GUEST)
 #include <linux/kvm_host.h>
 #endif
@@ -609,5 +612,12 @@ int main(void)
 					arch.timing_last_enter.tv32.tbl));
 #endif
 
+#ifdef CONFIG_PPC_POWERNV
+	DEFINE(OPAL_MC_GPR3, offsetof(struct opal_machine_check_event, gpr3));
+	DEFINE(OPAL_MC_SRR0, offsetof(struct opal_machine_check_event, srr0));
+	DEFINE(OPAL_MC_SRR1, offsetof(struct opal_machine_check_event, srr1));
+	DEFINE(PACA_OPAL_MC_EVT, offsetof(struct paca_struct, opal_mc_evt));
+#endif
+
 	return 0;
 }
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 41b02c7..d51458f 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -1143,7 +1143,7 @@ _GLOBAL(do_stab_bolted)
 	rfid
 	b	.	/* prevent speculative execution */
 
-#ifdef CONFIG_PPC_PSERIES
+#if defined(CONFIG_PPC_PSERIES) || defined(CONFIG_PPC_POWERNV)
 /*
  * Data area reserved for FWNMI option.
  * This address (0x7000) is fixed by the RPA.
@@ -1151,7 +1151,7 @@ _GLOBAL(do_stab_bolted)
 	.= 0x7000
 	.globl fwnmi_data_area
 fwnmi_data_area:
-#endif /* CONFIG_PPC_PSERIES */
+#endif /* defined(CONFIG_PPC_PSERIES) || defined(CONFIG_PPC_POWERNV) */
 
 	/* iSeries does not use the FWNMI stuff, so it is safe to put
 	 * this here, even if we later allow kernels that will boot on
@@ -1176,9 +1176,12 @@ xLparMap:
 
 #endif /* CONFIG_PPC_ISERIES */
 
-#ifdef CONFIG_PPC_PSERIES
+#if defined(CONFIG_PPC_PSERIES) || defined(CONFIG_PPC_POWERNV)
+	/* pseries and powernv need to keep the whole page from
+	 * 0x7000 to 0x8000 free for use by the firmware
+	 */
         . = 0x8000
-#endif /* CONFIG_PPC_PSERIES */
+#endif /* defined(CONFIG_PPC_PSERIES) || defined(CONFIG_PPC_POWERNV) */
 
 /*
  * Space for CPU0's segment table.
@@ -1193,3 +1196,19 @@ xLparMap:
 	.globl initial_stab
 initial_stab:
 	.space	4096
+#ifdef CONFIG_PPC_POWERNV
+_GLOBAL(opal_mc_secondary_handler)
+	HMT_MEDIUM
+	SET_SCRATCH0(r13)
+	GET_PACA(r13)
+	clrldi	r3,r3,2
+	tovirt(r3,r3)
+	std	r3,PACA_OPAL_MC_EVT(r13)
+	ld	r13,OPAL_MC_SRR0(r3)
+	mtspr	SPRN_SRR0,r13
+	ld	r13,OPAL_MC_SRR1(r3)
+	mtspr	SPRN_SRR1,r13
+	ld	r3,OPAL_MC_GPR3(r3)
+	GET_SCRATCH0(r13)
+	b	machine_check_pSeries
+#endif /* CONFIG_PPC_POWERNV */
diff --git a/arch/powerpc/platforms/powernv/opal.c b/arch/powerpc/platforms/powernv/opal.c
index 5a598ca..aaa0dba 100644
--- a/arch/powerpc/platforms/powernv/opal.c
+++ b/arch/powerpc/platforms/powernv/opal.c
@@ -27,12 +27,14 @@ struct opal {
 
 static struct device_node *opal_node;
 static DEFINE_SPINLOCK(opal_write_lock);
+extern u64 opal_mc_secondary_handler[];
 
 int __init early_init_dt_scan_opal(unsigned long node,
 				   const char *uname, int depth, void *data)
 {
 	const void *basep, *entryp;
 	unsigned long basesz, entrysz;
+	u64 glue;
 
 	if (depth != 1 || strcmp(uname, "ibm,opal") != 0)
 		return 0;
@@ -59,6 +61,19 @@ int __init early_init_dt_scan_opal(unsigned long node,
 		printk("OPAL V1 detected !\n");
 	}
 
+	/* Hookup some exception handlers. We use the fwnmi area at 0x7000
+	 * to provide the glue space to OPAL
+	 */
+	glue = 0x7000;
+	opal_register_exception_handler(OPAL_MACHINE_CHECK_HANDLER,
+					__pa(opal_mc_secondary_handler[0]),
+					glue);
+	glue += 128;
+	opal_register_exception_handler(OPAL_HYPERVISOR_MAINTENANCE_HANDLER,
+					0, glue);
+	glue += 128;
+	opal_register_exception_handler(OPAL_SOFTPATCH_HANDLER, 0, glue);
+
 	return 1;
 }
 
@@ -136,6 +151,121 @@ int opal_put_chars(uint32_t vtermno, const char *data, int total_len)
 	return written;
 }
 
+int opal_machine_check(struct pt_regs *regs)
+{
+	struct opal_machine_check_event *opal_evt = get_paca()->opal_mc_evt;
+	struct opal_machine_check_event evt;
+	const char *level, *sevstr, *subtype;
+	static const char *opal_mc_ue_types[] = {
+		"Indeterminate",
+		"Instruction fetch",
+		"Page table walk ifetch",
+		"Load/Store",
+		"Page table walk Load/Store",
+	};
+	static const char *opal_mc_slb_types[] = {
+		"Indeterminate",
+		"Parity",
+		"Multihit",
+	};
+	static const char *opal_mc_erat_types[] = {
+		"Indeterminate",
+		"Parity",
+		"Multihit",
+	};
+	static const char *opal_mc_tlb_types[] = {
+		"Indeterminate",
+		"Parity",
+		"Multihit",
+	};
+
+	/* Copy the event structure and release the original */
+	evt = *opal_evt;
+	opal_evt->in_use = 0;
+
+	/* Print things out */
+	if (evt.version != OpalMCE_V1) {
+		pr_err("Machine Check Exception, Unknown event version %d !\n",
+		       evt.version);
+		return 0;
+	}
+	switch(evt.severity) {
+	case OpalMCE_SEV_NO_ERROR:
+		level = KERN_INFO;
+		sevstr = "Harmless";
+		break;
+	case OpalMCE_SEV_WARNING:
+		level = KERN_WARNING;
+		sevstr = "";
+		break;
+	case OpalMCE_SEV_ERROR_SYNC:
+		level = KERN_ERR;
+		sevstr = "Severe";
+		break;
+	case OpalMCE_SEV_FATAL:
+	default:
+		level = KERN_ERR;
+		sevstr = "Fatal";
+		break;
+	}
+
+	printk("%s%s Machine check interrupt [%s]\n", level, sevstr,
+	       evt.disposition == OpalMCE_DISPOSITION_RECOVERED ?
+	       "Recovered" : "[Not recovered");
+	printk("%s  Initiator: %s\n", level,
+	       evt.initiator == OpalMCE_INITIATOR_CPU ? "CPU" : "Unknown");
+	switch(evt.error_type) {
+	case OpalMCE_ERROR_TYPE_UE:
+		subtype = evt.u.ue_error.ue_error_type <
+			ARRAY_SIZE(opal_mc_ue_types) ?
+			opal_mc_ue_types[evt.u.ue_error.ue_error_type]
+			: "Unknown";
+		printk("%s  Error type: UE [%s]\n", level, subtype);
+		if (evt.u.ue_error.effective_address_provided)
+			printk("%s    Effective address: %016llx\n",
+			       level, evt.u.ue_error.effective_address);
+		if (evt.u.ue_error.physical_address_provided)
+			printk("%s      Physial address: %016llx\n",
+			       level, evt.u.ue_error.physical_address);
+		break;
+	case OpalMCE_ERROR_TYPE_SLB:
+		subtype = evt.u.slb_error.slb_error_type <
+			ARRAY_SIZE(opal_mc_slb_types) ?
+			opal_mc_slb_types[evt.u.slb_error.slb_error_type]
+			: "Unknown";
+		printk("%s  Error type: SLB [%s]\n", level, subtype);
+		if (evt.u.slb_error.effective_address_provided)
+			printk("%s    Effective address: %016llx\n",
+			       level, evt.u.slb_error.effective_address);
+		break;
+	case OpalMCE_ERROR_TYPE_ERAT:
+		subtype = evt.u.erat_error.erat_error_type <
+			ARRAY_SIZE(opal_mc_erat_types) ?
+			opal_mc_erat_types[evt.u.erat_error.erat_error_type]
+			: "Unknown";
+		printk("%s  Error type: ERAT [%s]\n", level, subtype);
+		if (evt.u.erat_error.effective_address_provided)
+			printk("%s    Effective address: %016llx\n",
+			       level, evt.u.erat_error.effective_address);
+		break;
+	case OpalMCE_ERROR_TYPE_TLB:
+		subtype = evt.u.tlb_error.tlb_error_type <
+			ARRAY_SIZE(opal_mc_tlb_types) ?
+			opal_mc_tlb_types[evt.u.tlb_error.tlb_error_type]
+			: "Unknown";
+		printk("%s  Error type: TLB [%s]\n", level, subtype);
+		if (evt.u.tlb_error.effective_address_provided)
+			printk("%s    Effective address: %016llx\n",
+			       level, evt.u.tlb_error.effective_address);
+		break;
+	default:
+	case OpalMCE_ERROR_TYPE_UNKNOWN:
+		printk("%s  Error type: Unknown\n", level);
+		break;
+	}
+	return evt.severity == OpalMCE_SEV_FATAL ? 0 : 1;
+}
+
 static irqreturn_t opal_interrupt(int irq, void *data)
 {
 	uint64_t events;
diff --git a/arch/powerpc/platforms/powernv/setup.c b/arch/powerpc/platforms/powernv/setup.c
index 4a2b2e2..f0242f3 100644
--- a/arch/powerpc/platforms/powernv/setup.c
+++ b/arch/powerpc/platforms/powernv/setup.c
@@ -141,6 +141,7 @@ static void __init pnv_setup_machdep_opal(void)
 	ppc_md.restart = pnv_restart;
 	ppc_md.power_off = pnv_power_off;
 	ppc_md.halt = pnv_halt;
+	ppc_md.machine_check_exception = opal_machine_check;
 }
 
 #ifdef CONFIG_PPC_POWERNV_RTAS
-- 
1.7.4.1

[PATCH 19/21] powerpc/powernv: Add support for p5ioc2 PCI-X and PCIe

[Benjamin Herrenschmidt]

This adds support for PCI-X and PCIe on the p5ioc2 IO hub using
OPAL. This includes allocating & setting up TCE tables and config
space access routines.

This also supports fallbacks via RTAS when OPAL is absent, using
legacy TCE format pre-allocated via the device-tree (BML style)

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/platforms/powernv/Makefile     |    2 +
 arch/powerpc/platforms/powernv/pci-p5ioc2.c |  185 +++++++++++++++++
 arch/powerpc/platforms/powernv/pci.c        |  286 +++++++++++++++++++++++++++
 arch/powerpc/platforms/powernv/pci.h        |   38 ++++
 arch/powerpc/platforms/powernv/powernv.h    |    6 +
 arch/powerpc/platforms/powernv/setup.c      |    3 +-
 6 files changed, 519 insertions(+), 1 deletions(-)
 create mode 100644 arch/powerpc/platforms/powernv/pci-p5ioc2.c
 create mode 100644 arch/powerpc/platforms/powernv/pci.c
 create mode 100644 arch/powerpc/platforms/powernv/pci.h

diff --git a/arch/powerpc/platforms/powernv/Makefile b/arch/powerpc/platforms/powernv/Makefile
index 618ad83..cf9dec8 100644
--- a/arch/powerpc/platforms/powernv/Makefile
+++ b/arch/powerpc/platforms/powernv/Makefile
@@ -2,3 +2,5 @@ obj-y			+= setup.o opal-takeover.o opal-wrappers.o opal.o
 obj-y			+= opal-rtc.o opal-nvram.o
 
 obj-$(CONFIG_SMP)	+= smp.o
+obj-$(CONFIG_PCI)	+= pci.o pci-p5ioc2.o
+
diff --git a/arch/powerpc/platforms/powernv/pci-p5ioc2.c b/arch/powerpc/platforms/powernv/pci-p5ioc2.c
new file mode 100644
index 0000000..afabc2b
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/pci-p5ioc2.c
@@ -0,0 +1,185 @@
+/*
+ * Support PCI/PCIe on PowerNV platforms
+ *
+ * Currently supports only P5IOC2
+ *
+ * Copyright 2011 Benjamin Herrenschmidt, IBM Corp.
+ *
+ * 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.
+ */
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/string.h>
+#include <linux/init.h>
+#include <linux/bootmem.h>
+#include <linux/irq.h>
+#include <linux/io.h>
+
+#include <asm/sections.h>
+#include <asm/io.h>
+#include <asm/prom.h>
+#include <asm/pci-bridge.h>
+#include <asm/machdep.h>
+#include <asm/ppc-pci.h>
+#include <asm/opal.h>
+#include <asm/iommu.h>
+#include <asm/tce.h>
+#include <asm/abs_addr.h>
+
+#include "powernv.h"
+#include "pci.h"
+
+/* For now, use a fixed amount of TCE memory for each p5ioc2
+ * hub, 16M will do
+ */
+#define P5IOC2_TCE_MEMORY	0x01000000
+
+static void __devinit pnv_pci_p5ioc2_dma_dev_setup(struct pnv_phb *phb,
+						   struct pci_dev *pdev)
+{
+	if (phb->p5ioc2.iommu_table.it_map == NULL)
+		iommu_init_table(&phb->p5ioc2.iommu_table, phb->hose->node);
+
+	set_iommu_table_base(&pdev->dev, &phb->p5ioc2.iommu_table);
+}
+
+static void __init pnv_pci_init_p5ioc2_phb(struct device_node *np,
+					   void *tce_mem, u64 tce_size)
+{
+	struct pnv_phb *phb;
+	const u64 *prop64;
+	u64 phb_id;
+	int64_t rc;
+	static int primary = 1;
+
+	pr_info(" Initializing p5ioc2 PHB %s\n", np->full_name);
+
+	prop64 = of_get_property(np, "ibm,opal-phbid", NULL);
+	if (!prop64) {
+		pr_err("  Missing \"ibm,opal-phbid\" property !\n");
+		return;
+	}
+	phb_id = be64_to_cpup(prop64);
+	pr_devel("  PHB-ID  : 0x%016llx\n", phb_id);
+	pr_devel("  TCE AT  : 0x%016lx\n", __pa(tce_mem));
+	pr_devel("  TCE SZ  : 0x%016llx\n", tce_size);
+
+	rc = opal_pci_set_phb_tce_memory(phb_id, __pa(tce_mem), tce_size);
+	if (rc != OPAL_SUCCESS) {
+		pr_err("  Failed to set TCE memory, OPAL error %lld\n", rc);
+		return;
+	}
+
+	phb = alloc_bootmem(sizeof(struct pnv_phb));
+	if (phb) {
+		memset(phb, 0, sizeof(struct pnv_phb));
+		phb->hose = pcibios_alloc_controller(np);
+	}
+	if (!phb || !phb->hose) {
+		pr_err("  Failed to allocate PCI controller\n");
+		return;
+	}
+
+	spin_lock_init(&phb->lock);
+	phb->hose->first_busno = 0;
+	phb->hose->last_busno = 0xff;
+	phb->hose->private_data = phb;
+	phb->opal_id = phb_id;
+	phb->type = PNV_PHB_P5IOC2;
+
+	phb->regs = of_iomap(np, 0);
+
+	if (phb->regs == NULL)
+		pr_err("  Failed to map registers !\n");
+	else {
+		pr_devel("  P_BUID     = 0x%08x\n", in_be32(phb->regs + 0x100));
+		pr_devel("  P_IOSZ     = 0x%08x\n", in_be32(phb->regs + 0x1b0));
+		pr_devel("  P_IO_ST    = 0x%08x\n", in_be32(phb->regs + 0x1e0));
+		pr_devel("  P_MEM1_H   = 0x%08x\n", in_be32(phb->regs + 0x1a0));
+		pr_devel("  P_MEM1_L   = 0x%08x\n", in_be32(phb->regs + 0x190));
+		pr_devel("  P_MSZ1_L   = 0x%08x\n", in_be32(phb->regs + 0x1c0));
+		pr_devel("  P_MEM_ST   = 0x%08x\n", in_be32(phb->regs + 0x1d0));
+		pr_devel("  P_MEM2_H   = 0x%08x\n", in_be32(phb->regs + 0x2c0));
+		pr_devel("  P_MEM2_L   = 0x%08x\n", in_be32(phb->regs + 0x2b0));
+		pr_devel("  P_MSZ2_H   = 0x%08x\n", in_be32(phb->regs + 0x2d0));
+		pr_devel("  P_MSZ2_L   = 0x%08x\n", in_be32(phb->regs + 0x2e0));
+	}
+
+	/* Interpret the "ranges" property */
+	/* This also maps the I/O region and sets isa_io/mem_base */
+	pci_process_bridge_OF_ranges(phb->hose, np, primary);
+	primary = 0;
+
+	phb->hose->ops = &pnv_pci_ops;
+
+	/* Setup TCEs */
+	phb->dma_dev_setup = pnv_pci_p5ioc2_dma_dev_setup;
+	pnv_pci_setup_iommu_table(&phb->p5ioc2.iommu_table,
+				  tce_mem, tce_size, 0);
+}
+
+void __init pnv_pci_init_p5ioc2_hub(struct device_node *np)
+{
+	struct device_node *phbn;
+	const u64 *prop64;
+	u64 hub_id;
+	void *tce_mem;
+	uint64_t tce_per_phb;
+	int64_t rc;
+	int phb_count = 0;
+
+	pr_info("Probing p5ioc2 IO-Hub %s\n", np->full_name);
+
+	prop64 = of_get_property(np, "ibm,opal-hubid", NULL);
+	if (!prop64) {
+		pr_err(" Missing \"ibm,opal-hubid\" property !\n");
+		return;
+	}
+	hub_id = be64_to_cpup(prop64);
+	pr_info(" HUB-ID : 0x%016llx\n", hub_id);
+
+	/* Currently allocate 16M of TCE memory for every Hub
+	 *
+	 * XXX TODO: Make it chip local if possible
+	 */
+	tce_mem = __alloc_bootmem(P5IOC2_TCE_MEMORY, P5IOC2_TCE_MEMORY,
+				  __pa(MAX_DMA_ADDRESS));
+	if (!tce_mem) {
+		pr_err(" Failed to allocate TCE Memory !\n");
+		return;
+	}
+	pr_debug(" TCE    : 0x%016lx..0x%016lx\n",
+		__pa(tce_mem), __pa(tce_mem) + P5IOC2_TCE_MEMORY - 1);
+	rc = opal_pci_set_hub_tce_memory(hub_id, __pa(tce_mem),
+					P5IOC2_TCE_MEMORY);
+	if (rc != OPAL_SUCCESS) {
+		pr_err(" Failed to allocate TCE memory, OPAL error %lld\n", rc);
+		return;
+	}
+
+	/* Count child PHBs */
+	for_each_child_of_node(np, phbn) {
+		if (of_device_is_compatible(phbn, "ibm,p5ioc2-pcix") ||
+		    of_device_is_compatible(phbn, "ibm,p5ioc2-pciex"))
+			phb_count++;
+	}
+
+	/* Calculate how much TCE space we can give per PHB */
+	tce_per_phb = __rounddown_pow_of_two(P5IOC2_TCE_MEMORY / phb_count);
+	pr_info(" Allocating %lld MB of TCE memory per PHB\n",
+		tce_per_phb >> 20);
+
+	/* Initialize PHBs */
+	for_each_child_of_node(np, phbn) {
+		if (of_device_is_compatible(phbn, "ibm,p5ioc2-pcix") ||
+		    of_device_is_compatible(phbn, "ibm,p5ioc2-pciex")) {
+			pnv_pci_init_p5ioc2_phb(phbn, tce_mem, tce_per_phb);
+			tce_mem += tce_per_phb;
+		}
+	}
+}
diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c
new file mode 100644
index 0000000..b512489
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/pci.c
@@ -0,0 +1,286 @@
+/*
+ * Support PCI/PCIe on PowerNV platforms
+ *
+ * Currently supports only P5IOC2
+ *
+ * Copyright 2011 Benjamin Herrenschmidt, IBM Corp.
+ *
+ * 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.
+ */
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/string.h>
+#include <linux/init.h>
+#include <linux/bootmem.h>
+#include <linux/irq.h>
+#include <linux/io.h>
+
+#include <asm/sections.h>
+#include <asm/io.h>
+#include <asm/prom.h>
+#include <asm/pci-bridge.h>
+#include <asm/machdep.h>
+#include <asm/ppc-pci.h>
+#include <asm/opal.h>
+#include <asm/iommu.h>
+#include <asm/tce.h>
+#include <asm/abs_addr.h>
+
+#include "powernv.h"
+#include "pci.h"
+
+
+#define cfg_dbg(fmt...)	do { } while(0)
+//#define cfg_dbg(fmt...)	printk(fmt)
+
+
+static void pnv_pci_config_check_eeh(struct pnv_phb *phb, struct pci_bus *bus,
+				     u32 bdfn)
+{
+	s64	rc;
+	u8	fstate;
+	u16	pcierr;
+	u32	pe_no;
+
+	/* Get PE# if we support IODA */
+	pe_no = phb->bdfn_to_pe ? phb->bdfn_to_pe(phb, bus, bdfn & 0xff) : 0;
+
+	/* Read freeze status */
+	rc = opal_pci_eeh_freeze_status(phb->opal_id, pe_no, &fstate, &pcierr, 
+					NULL);
+	if (rc) {
+		pr_warning("PCI %d: Failed to read EEH status for PE#%d,"
+			   " err %lld\n", phb->hose->global_number, pe_no, rc);
+		return;
+	}
+	cfg_dbg(" -> EEH check, bdfn=%04x PE%d fstate=%x\n",
+		bdfn, pe_no, fstate);
+	if (fstate != 0) {		
+		rc = opal_pci_eeh_freeze_clear(phb->opal_id, pe_no,
+					      OPAL_EEH_ACTION_CLEAR_FREEZE_ALL);
+		if (rc) {
+			pr_warning("PCI %d: Failed to clear EEH freeze state"
+				   " for PE#%d, err %lld\n",
+				   phb->hose->global_number, pe_no, rc);
+		}
+	}
+}
+
+static int pnv_pci_read_config(struct pci_bus *bus,
+			       unsigned int devfn,
+			       int where, int size, u32 *val)
+{
+	struct pci_controller *hose = pci_bus_to_host(bus);
+	struct pnv_phb *phb = hose->private_data;
+	u32 bdfn = (((uint64_t)bus->number) << 8) | devfn;
+	s64 rc;
+
+	if (hose == NULL)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	switch (size) {
+	case 1: {
+		u8 v8;
+		rc = opal_pci_config_read_byte(phb->opal_id, bdfn, where, &v8);
+		*val = (rc == OPAL_SUCCESS) ? v8 : 0xff;
+		break;
+	}
+	case 2: {
+		u16 v16;
+		rc = opal_pci_config_read_half_word(phb->opal_id, bdfn, where,
+						   &v16);
+		*val = (rc == OPAL_SUCCESS) ? v16 : 0xffff;
+		break;
+	}
+	case 4: {
+		u32 v32;
+		rc = opal_pci_config_read_word(phb->opal_id, bdfn, where, &v32);
+		*val = (rc == OPAL_SUCCESS) ? v32 : 0xffffffff;
+		break;
+	}
+	default:
+		return PCIBIOS_FUNC_NOT_SUPPORTED;
+	}
+	cfg_dbg("pnv_pci_read_config bus: %x devfn: %x +%x/%x -> %08x\n",
+		bus->number, devfn, where, size, *val);
+
+	/* Check if the PHB got frozen due to an error (no response) */
+	pnv_pci_config_check_eeh(phb, bus, bdfn);
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int pnv_pci_write_config(struct pci_bus *bus,
+				unsigned int devfn,
+				int where, int size, u32 val)
+{
+	struct pci_controller *hose = pci_bus_to_host(bus);
+	struct pnv_phb *phb = hose->private_data;
+	u32 bdfn = (((uint64_t)bus->number) << 8) | devfn;
+
+	if (hose == NULL)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	cfg_dbg("pnv_pci_write_config bus: %x devfn: %x +%x/%x -> %08x\n",
+		bus->number, devfn, where, size, val);
+	switch (size) {
+	case 1:
+		opal_pci_config_write_byte(phb->opal_id, bdfn, where, val);
+		break;
+	case 2:
+		opal_pci_config_write_half_word(phb->opal_id, bdfn, where, val);
+		break;
+	case 4:
+		opal_pci_config_write_word(phb->opal_id, bdfn, where, val);
+		break;
+	default:
+		return PCIBIOS_FUNC_NOT_SUPPORTED;
+	}
+	/* Check if the PHB got frozen due to an error (no response) */
+	pnv_pci_config_check_eeh(phb, bus, bdfn);
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+struct pci_ops pnv_pci_ops = {
+	.read = pnv_pci_read_config,
+	.write = pnv_pci_write_config,
+};
+
+static int pnv_tce_build(struct iommu_table *tbl, long index, long npages,
+			 unsigned long uaddr, enum dma_data_direction direction,
+			 struct dma_attrs *attrs)
+{
+	u64 proto_tce;
+	u64 *tcep;
+	u64 rpn;
+
+	proto_tce = TCE_PCI_READ; // Read allowed
+
+	if (direction != DMA_TO_DEVICE)
+		proto_tce |= TCE_PCI_WRITE;
+
+	tcep = ((u64 *)tbl->it_base) + index;
+
+	while (npages--) {
+		/* can't move this out since we might cross LMB boundary */
+		rpn = (virt_to_abs(uaddr)) >> TCE_SHIFT;
+		*tcep = proto_tce | (rpn & TCE_RPN_MASK) << TCE_RPN_SHIFT;
+
+		uaddr += TCE_PAGE_SIZE;
+		tcep++;
+	}
+	return 0;
+}
+
+static void pnv_tce_free(struct iommu_table *tbl, long index, long npages)
+{
+	u64 *tcep = ((u64 *)tbl->it_base) + index;
+
+	while (npages--)
+		*(tcep++) = 0;
+}
+
+void pnv_pci_setup_iommu_table(struct iommu_table *tbl,
+			       void *tce_mem, u64 tce_size,
+			       u64 dma_offset)
+{
+	tbl->it_blocksize = 16;
+	tbl->it_base = (unsigned long)tce_mem;
+	tbl->it_offset = dma_offset >> IOMMU_PAGE_SHIFT;
+	tbl->it_index = 0;
+	tbl->it_size = tce_size >> 3;
+	tbl->it_busno = 0;
+	tbl->it_type = TCE_PCI;
+}
+
+static struct iommu_table * __devinit
+pnv_pci_setup_bml_iommu(struct pci_controller *hose)
+{
+	struct iommu_table *tbl;
+	const __be64 *basep;
+	const __be32 *sizep;
+
+	basep = of_get_property(hose->dn, "linux,tce-base", NULL);
+	sizep = of_get_property(hose->dn, "linux,tce-size", NULL);
+	if (basep == NULL || sizep == NULL) {
+		pr_err("PCI: %s has missing tce entries !\n", hose->dn->full_name);
+		return NULL;
+	}
+	tbl = kzalloc_node(sizeof(struct iommu_table), GFP_KERNEL, hose->node);
+	if (WARN_ON(!tbl))
+		return NULL;
+	pnv_pci_setup_iommu_table(tbl, __va(be64_to_cpup(basep)),
+				  be32_to_cpup(sizep), 0);
+	iommu_init_table(tbl, hose->node);
+	return tbl;
+}
+
+static void __devinit pnv_pci_dma_fallback_setup(struct pci_controller *hose,
+						 struct pci_dev *pdev)
+{
+	struct device_node *np = pci_bus_to_OF_node(hose->bus);
+	struct pci_dn *pdn;
+
+	if (np == NULL)
+		return;
+	pdn = PCI_DN(np);
+	if (!pdn->iommu_table)
+		pdn->iommu_table = pnv_pci_setup_bml_iommu(hose);
+	if (!pdn->iommu_table)
+		return;
+	set_iommu_table_base(&pdev->dev, pdn->iommu_table);
+}
+
+static void __devinit pnv_pci_dma_dev_setup(struct pci_dev *pdev)
+{
+	struct pci_controller *hose = pci_bus_to_host(pdev->bus);
+	struct pnv_phb *phb = hose->private_data;
+
+	/* If we have no phb structure, try to setup a fallback based on
+	 * the device-tree (RTAS PCI for example)
+	 */
+	if (phb && phb->dma_dev_setup)
+		phb->dma_dev_setup(phb, pdev);
+	else
+		pnv_pci_dma_fallback_setup(hose, pdev);
+}
+
+void __init pnv_pci_init(void)
+{
+	struct device_node *np;
+
+	pci_set_flags(PCI_CAN_SKIP_ISA_ALIGN);
+
+	/* We do not want to just probe */
+	pci_probe_only = 0;
+
+	/* OPAL absent, try POPAL first then RTAS detection of PHBs */
+	if (!firmware_has_feature(FW_FEATURE_OPAL)) {
+#ifdef CONFIG_PPC_POWERNV_RTAS
+		init_pci_config_tokens();
+		find_and_init_phbs();
+#endif /* CONFIG_PPC_POWERNV_RTAS */
+	} else {
+		/* OPAL is here, do our normal stuff */
+
+		/* Look for p5ioc2 IO-Hubs */
+		for_each_compatible_node(np, NULL, "ibm,p5ioc2")
+			pnv_pci_init_p5ioc2_hub(np);
+	}
+
+	/* Setup the linkage between OF nodes and PHBs */
+	pci_devs_phb_init();
+
+	/* Configure IOMMU DMA hooks */
+	ppc_md.pci_dma_dev_setup = pnv_pci_dma_dev_setup;
+	ppc_md.tce_build = pnv_tce_build;
+	ppc_md.tce_free = pnv_tce_free;
+	set_pci_dma_ops(&dma_iommu_ops);
+
+}
diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h
new file mode 100644
index 0000000..b8be721
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/pci.h
@@ -0,0 +1,38 @@
+#ifndef __POWERNV_PCI_H
+#define __POWERNV_PCI_H
+
+struct pci_dn;
+
+enum pnv_phb_type {
+	PNV_PHB_P5IOC2,
+	PNV_PHB_IODA1,
+	PNV_PHB_IODA2,
+};
+
+struct pnv_phb {
+	struct pci_controller	*hose;
+	enum pnv_phb_type	type;
+	u64			opal_id;
+	void __iomem		*regs;
+	spinlock_t		lock;
+
+	void (*dma_dev_setup)(struct pnv_phb *phb, struct pci_dev *pdev);
+	void (*fixup_phb)(struct pci_controller *hose);
+	u32 (*bdfn_to_pe)(struct pnv_phb *phb, struct pci_bus *bus, u32 devfn);
+	
+	union {
+		struct {
+			struct iommu_table iommu_table;
+		} p5ioc2;
+	};
+};
+
+extern struct pci_ops pnv_pci_ops;
+
+extern void pnv_pci_setup_iommu_table(struct iommu_table *tbl,
+				      void *tce_mem, u64 tce_size,
+				      u64 dma_offset);
+extern void pnv_pci_init_p5ioc2_hub(struct device_node *np);
+
+
+#endif /* __POWERNV_PCI_H */
diff --git a/arch/powerpc/platforms/powernv/powernv.h b/arch/powerpc/platforms/powernv/powernv.h
index 35b7160..8a9df7f 100644
--- a/arch/powerpc/platforms/powernv/powernv.h
+++ b/arch/powerpc/platforms/powernv/powernv.h
@@ -7,4 +7,10 @@ extern void pnv_smp_init(void);
 static inline void pnv_smp_init(void) { }
 #endif
 
+#ifdef CONFIG_PCI
+extern void pnv_pci_init(void);
+#else
+static inline void pnv_pci_init(void) { }
+#endif
+
 #endif /* _POWERNV_H */
diff --git a/arch/powerpc/platforms/powernv/setup.c b/arch/powerpc/platforms/powernv/setup.c
index f0242f3..467bd4a 100644
--- a/arch/powerpc/platforms/powernv/setup.c
+++ b/arch/powerpc/platforms/powernv/setup.c
@@ -40,7 +40,8 @@ static void __init pnv_setup_arch(void)
 	/* Initialize SMP */
 	pnv_smp_init();
 
-	/* XXX PCI */
+	/* Setup PCI */
+	pnv_pci_init();
 
 	/* Setup RTC and NVRAM callbacks */
 	if (firmware_has_feature(FW_FEATURE_OPAL))
-- 
1.7.4.1

[PATCH 20/21] powerpc/powernv: Implement MSI support for p5ioc2 PCIe

[Benjamin Herrenschmidt]

This implements support for MSIs on p5ioc2 PHBs. We only support
MSIs on the PCIe PHBs, not the PCI-X ones as the later hasn't been
properly verified in HW.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/platforms/powernv/pci-p5ioc2.c |   49 ++++++++++++
 arch/powerpc/platforms/powernv/pci.c        |  108 +++++++++++++++++++++++++++
 arch/powerpc/platforms/powernv/pci.h        |   10 +++
 3 files changed, 167 insertions(+), 0 deletions(-)

diff --git a/arch/powerpc/platforms/powernv/pci-p5ioc2.c b/arch/powerpc/platforms/powernv/pci-p5ioc2.c
index afabc2b..fd1509a 100644
--- a/arch/powerpc/platforms/powernv/pci-p5ioc2.c
+++ b/arch/powerpc/platforms/powernv/pci-p5ioc2.c
@@ -19,6 +19,7 @@
 #include <linux/bootmem.h>
 #include <linux/irq.h>
 #include <linux/io.h>
+#include <linux/msi.h>
 
 #include <asm/sections.h>
 #include <asm/io.h>
@@ -39,6 +40,51 @@
  */
 #define P5IOC2_TCE_MEMORY	0x01000000
 
+#ifdef CONFIG_PCI_MSI
+static int pnv_pci_p5ioc2_msi_setup(struct pnv_phb *phb, struct pci_dev *dev,
+				    unsigned int hwirq, unsigned int is_64,
+				    struct msi_msg *msg)
+{
+	if (WARN_ON(!is_64))
+		return -ENXIO;
+	msg->data = hwirq - phb->msi_base;
+	msg->address_hi = 0x10000000;
+	msg->address_lo = 0;
+
+	return 0;
+}
+
+static void pnv_pci_init_p5ioc2_msis(struct pnv_phb *phb)
+{
+	unsigned int bmap_size;
+	const __be32 *prop = of_get_property(phb->hose->dn,
+					     "ibm,opal-msi-ranges", NULL);
+	if (!prop)
+		return;
+
+	/* Don't do MSI's on p5ioc2 PCI-X are they are not properly
+	 * verified in HW
+	 */
+	if (of_device_is_compatible(phb->hose->dn, "ibm,p5ioc2-pcix"))
+		return;
+	phb->msi_base = be32_to_cpup(prop);
+	phb->msi_count = be32_to_cpup(prop + 1);
+	bmap_size = BITS_TO_LONGS(phb->msi_count) * sizeof(unsigned long);
+	phb->msi_map = zalloc_maybe_bootmem(bmap_size, GFP_KERNEL);
+	if (!phb->msi_map) {
+		pr_err("PCI %d: Failed to allocate MSI bitmap !\n",
+		       phb->hose->global_number);
+		return;
+	}
+	phb->msi_setup = pnv_pci_p5ioc2_msi_setup;
+	phb->msi32_support = 0;
+	pr_info("  Allocated bitmap for %d MSIs (base IRQ 0x%x)\n",
+		phb->msi_count, phb->msi_base);
+}
+#else
+static void pnv_pci_setup_p5ioc2_msis(struct pnv_phb *phb) { }
+#endif /* CONFIG_PCI_MSI */
+
 static void __devinit pnv_pci_p5ioc2_dma_dev_setup(struct pnv_phb *phb,
 						   struct pci_dev *pdev)
 {
@@ -117,6 +163,9 @@ static void __init pnv_pci_init_p5ioc2_phb(struct device_node *np,
 
 	phb->hose->ops = &pnv_pci_ops;
 
+	/* Setup MSI support */
+	pnv_pci_init_p5ioc2_msis(phb);
+
 	/* Setup TCEs */
 	phb->dma_dev_setup = pnv_pci_p5ioc2_dma_dev_setup;
 	pnv_pci_setup_iommu_table(&phb->p5ioc2.iommu_table,
diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c
index b512489..f85fb13 100644
--- a/arch/powerpc/platforms/powernv/pci.c
+++ b/arch/powerpc/platforms/powernv/pci.c
@@ -19,6 +19,7 @@
 #include <linux/bootmem.h>
 #include <linux/irq.h>
 #include <linux/io.h>
+#include <linux/msi.h>
 
 #include <asm/sections.h>
 #include <asm/io.h>
@@ -38,7 +39,108 @@
 #define cfg_dbg(fmt...)	do { } while(0)
 //#define cfg_dbg(fmt...)	printk(fmt)
 
+#ifdef CONFIG_PCI_MSI
+static int pnv_msi_check_device(struct pci_dev* pdev, int nvec, int type)
+{
+	struct pci_controller *hose = pci_bus_to_host(pdev->bus);
+	struct pnv_phb *phb = hose->private_data;
+
+	return (phb && phb->msi_map) ? 0 : -ENODEV;
+}
+
+static unsigned int pnv_get_one_msi(struct pnv_phb *phb)
+{
+	unsigned int id;
+
+	spin_lock(&phb->lock);
+	id = find_next_zero_bit(phb->msi_map, phb->msi_count, phb->msi_next);
+	if (id >= phb->msi_count && phb->msi_next)
+		id = find_next_zero_bit(phb->msi_map, phb->msi_count, 0);
+	if (id >= phb->msi_count) {
+		spin_unlock(&phb->lock);
+		return 0;
+	}
+	__set_bit(id, phb->msi_map);
+	spin_unlock(&phb->lock);
+	return id + phb->msi_base;
+}
+
+static void pnv_put_msi(struct pnv_phb *phb, unsigned int hwirq)
+{
+	unsigned int id;
+
+	if (WARN_ON(hwirq < phb->msi_base ||
+		    hwirq >= (phb->msi_base + phb->msi_count)))
+		return;
+	id = hwirq - phb->msi_base;
+	spin_lock(&phb->lock);
+	__clear_bit(id, phb->msi_map);
+	spin_unlock(&phb->lock);
+}
+
+static int pnv_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type)
+{
+	struct pci_controller *hose = pci_bus_to_host(pdev->bus);
+	struct pnv_phb *phb = hose->private_data;
+	struct msi_desc *entry;
+	struct msi_msg msg;
+	unsigned int hwirq, virq;
+	int rc;
+
+	if (WARN_ON(!phb))
+		return -ENODEV;
+
+	list_for_each_entry(entry, &pdev->msi_list, list) {
+		if (!entry->msi_attrib.is_64 && !phb->msi32_support) {
+			pr_warn("%s: Supports only 64-bit MSIs\n",
+				pci_name(pdev));
+			return -ENXIO;
+		}
+		hwirq = pnv_get_one_msi(phb);
+		if (!hwirq) {
+			pr_warn("%s: Failed to find a free MSI\n",
+				pci_name(pdev));
+			return -ENOSPC;
+		}
+		virq = irq_create_mapping(NULL, hwirq);
+		if (virq == NO_IRQ) {
+			pr_warn("%s: Failed to map MSI to linux irq\n",
+				pci_name(pdev));
+			pnv_put_msi(phb, hwirq);
+			return -ENOMEM;
+		}
+		rc = phb->msi_setup(phb, pdev, hwirq, entry->msi_attrib.is_64,
+				    &msg);
+		if (rc) {
+			pr_warn("%s: Failed to setup MSI\n", pci_name(pdev));
+			irq_dispose_mapping(virq);
+			pnv_put_msi(phb, hwirq);
+			return rc;
+		}
+		irq_set_msi_desc(virq, entry);
+		write_msi_msg(virq, &msg);
+	}
+	return 0;
+}
 
+static void pnv_teardown_msi_irqs(struct pci_dev *pdev)
+{
+	struct pci_controller *hose = pci_bus_to_host(pdev->bus);
+	struct pnv_phb *phb = hose->private_data;
+	struct msi_desc *entry;
+
+	if (WARN_ON(!phb))
+		return;
+
+	list_for_each_entry(entry, &pdev->msi_list, list) {
+		if (entry->irq == NO_IRQ)
+			continue;
+		irq_set_msi_desc(entry->irq, NULL);
+		pnv_put_msi(phb, virq_to_hw(entry->irq));
+		irq_dispose_mapping(entry->irq);
+	}
+}
+#endif /* CONFIG_PCI_MSI */
 static void pnv_pci_config_check_eeh(struct pnv_phb *phb, struct pci_bus *bus,
 				     u32 bdfn)
 {
@@ -283,4 +385,10 @@ void __init pnv_pci_init(void)
 	ppc_md.tce_free = pnv_tce_free;
 	set_pci_dma_ops(&dma_iommu_ops);
 
+	/* Configure MSIs */
+#ifdef CONFIG_PCI_MSI
+	ppc_md.msi_check_device = pnv_msi_check_device;
+	ppc_md.setup_msi_irqs = pnv_setup_msi_irqs;
+	ppc_md.teardown_msi_irqs = pnv_teardown_msi_irqs;
+#endif
 }
diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h
index b8be721..a468c9b 100644
--- a/arch/powerpc/platforms/powernv/pci.h
+++ b/arch/powerpc/platforms/powernv/pci.h
@@ -16,6 +16,16 @@ struct pnv_phb {
 	void __iomem		*regs;
 	spinlock_t		lock;
 
+#ifdef CONFIG_PCI_MSI
+	unsigned long		*msi_map;
+	unsigned int		msi_base;
+	unsigned int		msi_count;
+	unsigned int		msi_next;
+	unsigned int		msi32_support;
+#endif
+	int (*msi_setup)(struct pnv_phb *phb, struct pci_dev *dev,
+			 unsigned int hwirq, unsigned int is_64,
+			 struct msi_msg *msg);
 	void (*dma_dev_setup)(struct pnv_phb *phb, struct pci_dev *pdev);
 	void (*fixup_phb)(struct pci_controller *hose);
 	u32 (*bdfn_to_pe)(struct pnv_phb *phb, struct pci_bus *bus, u32 devfn);
-- 
1.7.4.1

[PATCH 21/21] powerpc/powernv: Handle PCI-X/PCIe reset delay

[Benjamin Herrenschmidt]

The firmware doesn't wait after lifting the PCI reset. However it does
timestamp it in the device tree. We use that to ensure we wait long
enough (3s is our current arbitrary setting) from that timestamp to
actually probing the bus.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/platforms/powernv/pci.c |   32 ++++++++++++++++++++++++++++++++
 1 files changed, 32 insertions(+), 0 deletions(-)

diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c
index f85fb13..1835db5 100644
--- a/arch/powerpc/platforms/powernv/pci.c
+++ b/arch/powerpc/platforms/powernv/pci.c
@@ -35,6 +35,8 @@
 #include "powernv.h"
 #include "pci.h"
 
+/* Delay in usec */
+#define PCI_RESET_DELAY_US	3000000
 
 #define cfg_dbg(fmt...)	do { } while(0)
 //#define cfg_dbg(fmt...)	printk(fmt)
@@ -353,6 +355,35 @@ static void __devinit pnv_pci_dma_dev_setup(struct pci_dev *pdev)
 		pnv_pci_dma_fallback_setup(hose, pdev);
 }
 
+static int pnv_pci_probe_mode(struct pci_bus *bus)
+{
+	struct pci_controller *hose = pci_bus_to_host(bus);
+	const __be64 *tstamp;
+	u64 now, target;
+
+
+	/* We hijack this as a way to ensure we have waited long
+	 * enough since the reset was lifted on the PCI bus
+	 */
+	if (bus != hose->bus)
+		return PCI_PROBE_NORMAL;
+	tstamp = of_get_property(hose->dn, "reset-clear-timestamp", NULL);
+	if (!tstamp || !*tstamp)
+		return PCI_PROBE_NORMAL;
+
+	now = mftb() / tb_ticks_per_usec;
+	target = (be64_to_cpup(tstamp) / tb_ticks_per_usec)
+		+ PCI_RESET_DELAY_US;
+
+	pr_devel("pci %04d: Reset target: 0x%llx now: 0x%llx\n",
+		 hose->global_number, target, now);
+
+	if (now < target)
+		msleep((target - now + 999) / 1000);
+
+	return PCI_PROBE_NORMAL;
+}
+
 void __init pnv_pci_init(void)
 {
 	struct device_node *np;
@@ -383,6 +414,7 @@ void __init pnv_pci_init(void)
 	ppc_md.pci_dma_dev_setup = pnv_pci_dma_dev_setup;
 	ppc_md.tce_build = pnv_tce_build;
 	ppc_md.tce_free = pnv_tce_free;
+	ppc_md.pci_probe_mode = pnv_pci_probe_mode;
 	set_pci_dma_ops(&dma_iommu_ops);
 
 	/* Configure MSIs */
-- 
1.7.4.1