Merge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc

Pull powerpc updates from Ben Herrenschmidt:
 "Here is the bulk of the powerpc changes for this merge window.  It got
  a bit delayed in part because I wasn't paying attention, and in part
  because I discovered I had a core PCI change without a PCI maintainer
  ack in it.  Bjorn eventually agreed it was ok to merge it though we'll
  probably improve it later and I didn't want to rebase to add his ack.

  There is going to be a bit more next week, essentially fixes that I
  still want to sort through and test.

  The biggest item this time is the support to build the ppc64 LE kernel
  with our new v2 ABI.  We previously supported v2 userspace but the
  kernel itself was a tougher nut to crack.  This is now sorted mostly
  thanks to Anton and Rusty.

  We also have a fairly big series from Cedric that add support for
  64-bit LE zImage boot wrapper.  This was made harder by the fact that
  traditionally our zImage wrapper was always 32-bit, but our new LE
  toolchains don't really support 32-bit anymore (it's somewhat there
  but not really "supported") so we didn't want to rely on it.  This
  meant more churn that just endian fixes.

  This brings some more LE bits as well, such as the ability to run in
  LE mode without a hypervisor (ie. under OPAL firmware) by doing the
  right OPAL call to reinitialize the CPU to take HV interrupts in the
  right mode and the usual pile of endian fixes.

  There's another series from Gavin adding EEH improvements (one day we
  *will* have a release with less than 20 EEH patches, I promise!).

  Another highlight is the support for the "Split core" functionality on
  P8 by Michael.  This allows a P8 core to be split into "sub cores" of
  4 threads which allows the subcores to run different guests under KVM
  (the HW still doesn't support a partition per thread).

  And then the usual misc bits and fixes ..."

[ Further delayed by gmail deciding that BenH is a dirty spammer.
  Google knows.  ]

* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc: (155 commits)
  powerpc/powernv: Add missing include to LPC code
  selftests/powerpc: Test the THP bug we fixed in the previous commit
  powerpc/mm: Check paca psize is up to date for huge mappings
  powerpc/powernv: Pass buffer size to OPAL validate flash call
  powerpc/pseries: hcall functions are exported to modules, need _GLOBAL_TOC()
  powerpc: Exported functions __clear_user and copy_page use r2 so need _GLOBAL_TOC()
  powerpc/powernv: Set memory_block_size_bytes to 256MB
  powerpc: Allow ppc_md platform hook to override memory_block_size_bytes
  powerpc/powernv: Fix endian issues in memory error handling code
  powerpc/eeh: Skip eeh sysfs when eeh is disabled
  powerpc: 64bit sendfile is capped at 2GB
  powerpc/powernv: Provide debugfs access to the LPC bus via OPAL
  powerpc/serial: Use saner flags when creating legacy ports
  powerpc: Add cpu family documentation
  powerpc/xmon: Fix up xmon format strings
  powerpc/powernv: Add calls to support little endian host
  powerpc: Document sysfs DSCR interface
  powerpc: Fix regression of per-CPU DSCR setting
  powerpc: Split __SYSFS_SPRSETUP macro
  arch: powerpc/fadump: Cleaning up inconsistent NULL checks
  ...

18 files changed, 1078 insertions, 257 deletions

diff --git a/arch/powerpc/platforms/powernv/Makefile b/arch/powerpc/platforms/powernv/Makefile
index 63cebb9b4d45..4ad0d345bc96 100644
--- a/arch/powerpc/platforms/powernv/Makefile
+++ b/arch/powerpc/platforms/powernv/Makefile
@@ -1,7 +1,7 @@
 obj-y			+= setup.o opal-takeover.o opal-wrappers.o opal.o opal-async.o
 obj-y			+= opal-rtc.o opal-nvram.o opal-lpc.o opal-flash.o
 obj-y			+= rng.o opal-elog.o opal-dump.o opal-sysparam.o opal-sensor.o
-obj-y			+= opal-msglog.o
+obj-y			+= opal-msglog.o subcore.o subcore-asm.o
 
 obj-$(CONFIG_SMP)	+= smp.o
 obj-$(CONFIG_PCI)	+= pci.o pci-p5ioc2.o pci-ioda.o
diff --git a/arch/powerpc/platforms/powernv/eeh-ioda.c b/arch/powerpc/platforms/powernv/eeh-ioda.c
index 5b51079f3e3b..753f08e36dfa 100644
--- a/arch/powerpc/platforms/powernv/eeh-ioda.c
+++ b/arch/powerpc/platforms/powernv/eeh-ioda.c
@@ -42,11 +42,19 @@ static int ioda_eeh_event(struct notifier_block *nb,
 {
 	uint64_t changed_evts = (uint64_t)change;
 
-	/* We simply send special EEH event */
-	if ((changed_evts & OPAL_EVENT_PCI_ERROR) &&
-	    (events & OPAL_EVENT_PCI_ERROR) &&
-	    eeh_enabled())
+	/*
+	 * We simply send special EEH event if EEH has
+	 * been enabled, or clear pending events in
+	 * case that we enable EEH soon
+	 */
+	if (!(changed_evts & OPAL_EVENT_PCI_ERROR) ||
+	    !(events & OPAL_EVENT_PCI_ERROR))
+		return 0;
+
+	if (eeh_enabled())
 		eeh_send_failure_event(NULL);
+	else
+		opal_notifier_update_evt(OPAL_EVENT_PCI_ERROR, 0x0ul);
 
 	return 0;
 }
@@ -141,7 +149,9 @@ static int ioda_eeh_post_init(struct pci_controller *hose)
 	}
 
 #ifdef CONFIG_DEBUG_FS
-	if (phb->dbgfs) {
+	if (!phb->has_dbgfs && phb->dbgfs) {
+		phb->has_dbgfs = 1;
+
 		debugfs_create_file("err_injct_outbound", 0600,
 				    phb->dbgfs, hose,
 				    &ioda_eeh_outb_dbgfs_ops);
@@ -154,7 +164,14 @@ static int ioda_eeh_post_init(struct pci_controller *hose)
 	}
 #endif
 
-	phb->eeh_state |= PNV_EEH_STATE_ENABLED;
+	/* If EEH is enabled, we're going to rely on that.
+	 * Otherwise, we restore to conventional mechanism
+	 * to clear frozen PE during PCI config access.
+	 */
+	if (eeh_enabled())
+		phb->flags |= PNV_PHB_FLAG_EEH;
+	else
+		phb->flags &= ~PNV_PHB_FLAG_EEH;
 
 	return 0;
 }
@@ -268,6 +285,21 @@ static int ioda_eeh_get_state(struct eeh_pe *pe)
 		return EEH_STATE_NOT_SUPPORT;
 	}
 
+	/*
+	 * If we're in middle of PE reset, return normal
+	 * state to keep EEH core going. For PHB reset, we
+	 * still expect to have fenced PHB cleared with
+	 * PHB reset.
+	 */
+	if (!(pe->type & EEH_PE_PHB) &&
+	    (pe->state & EEH_PE_RESET)) {
+		result = (EEH_STATE_MMIO_ACTIVE |
+			  EEH_STATE_DMA_ACTIVE |
+			  EEH_STATE_MMIO_ENABLED |
+			  EEH_STATE_DMA_ENABLED);
+		return result;
+	}
+
 	/* Retrieve PE status through OPAL */
 	pe_no = pe->addr;
 	ret = opal_pci_eeh_freeze_status(phb->opal_id, pe_no,
@@ -347,52 +379,6 @@ static int ioda_eeh_get_state(struct eeh_pe *pe)
 	return result;
 }
 
-static int ioda_eeh_pe_clear(struct eeh_pe *pe)
-{
-	struct pci_controller *hose;
-	struct pnv_phb *phb;
-	u32 pe_no;
-	u8 fstate;
-	u16 pcierr;
-	s64 ret;
-
-	pe_no = pe->addr;
-	hose = pe->phb;
-	phb = pe->phb->private_data;
-
-	/* Clear the EEH error on the PE */
-	ret = opal_pci_eeh_freeze_clear(phb->opal_id,
-			pe_no, OPAL_EEH_ACTION_CLEAR_FREEZE_ALL);
-	if (ret) {
-		pr_err("%s: Failed to clear EEH error for "
-		       "PHB#%x-PE#%x, err=%lld\n",
-		       __func__, hose->global_number, pe_no, ret);
-		return -EIO;
-	}
-
-	/*
-	 * Read the PE state back and verify that the frozen
-	 * state has been removed.
-	 */
-	ret = opal_pci_eeh_freeze_status(phb->opal_id, pe_no,
-			&fstate, &pcierr, NULL);
-	if (ret) {
-		pr_err("%s: Failed to get EEH status on "
-		       "PHB#%x-PE#%x\n, err=%lld\n",
-		       __func__, hose->global_number, pe_no, ret);
-		return -EIO;
-	}
-
-	if (fstate != OPAL_EEH_STOPPED_NOT_FROZEN) {
-		pr_err("%s: Frozen state not cleared on "
-		       "PHB#%x-PE#%x, sts=%x\n",
-		       __func__, hose->global_number, pe_no, fstate);
-		return -EIO;
-	}
-
-	return 0;
-}
-
 static s64 ioda_eeh_phb_poll(struct pnv_phb *phb)
 {
 	s64 rc = OPAL_HARDWARE;
@@ -402,13 +388,16 @@ static s64 ioda_eeh_phb_poll(struct pnv_phb *phb)
 		if (rc <= 0)
 			break;
 
-		msleep(rc);
+		if (system_state < SYSTEM_RUNNING)
+			udelay(1000 * rc);
+		else
+			msleep(rc);
 	}
 
 	return rc;
 }
 
-static int ioda_eeh_phb_reset(struct pci_controller *hose, int option)
+int ioda_eeh_phb_reset(struct pci_controller *hose, int option)
 {
 	struct pnv_phb *phb = hose->private_data;
 	s64 rc = OPAL_HARDWARE;
@@ -431,9 +420,17 @@ static int ioda_eeh_phb_reset(struct pci_controller *hose, int option)
 
 	/*
 	 * Poll state of the PHB until the request is done
-	 * successfully.
+	 * successfully. The PHB reset is usually PHB complete
+	 * reset followed by hot reset on root bus. So we also
+	 * need the PCI bus settlement delay.
 	 */
 	rc = ioda_eeh_phb_poll(phb);
+	if (option == EEH_RESET_DEACTIVATE) {
+		if (system_state < SYSTEM_RUNNING)
+			udelay(1000 * EEH_PE_RST_SETTLE_TIME);
+		else
+			msleep(EEH_PE_RST_SETTLE_TIME);
+	}
 out:
 	if (rc != OPAL_SUCCESS)
 		return -EIO;
@@ -471,6 +468,8 @@ static int ioda_eeh_root_reset(struct pci_controller *hose, int option)
 
 	/* Poll state of the PHB until the request is done */
 	rc = ioda_eeh_phb_poll(phb);
+	if (option == EEH_RESET_DEACTIVATE)
+		msleep(EEH_PE_RST_SETTLE_TIME);
 out:
 	if (rc != OPAL_SUCCESS)
 		return -EIO;
@@ -478,32 +477,71 @@ out:
 	return 0;
 }
 
-static int ioda_eeh_bridge_reset(struct pci_controller *hose,
-		struct pci_dev *dev, int option)
+static int ioda_eeh_bridge_reset(struct pci_dev *dev, int option)
+
 {
-	u16 ctrl;
+	struct device_node *dn = pci_device_to_OF_node(dev);
+	struct eeh_dev *edev = of_node_to_eeh_dev(dn);
+	int aer = edev ? edev->aer_cap : 0;
+	u32 ctrl;
 
-	pr_debug("%s: Reset device %04x:%02x:%02x.%01x with option %d\n",
-		 __func__, hose->global_number, dev->bus->number,
-		 PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn), option);
+	pr_debug("%s: Reset PCI bus %04x:%02x with option %d\n",
+		 __func__, pci_domain_nr(dev->bus),
+		 dev->bus->number, option);
 
 	switch (option) {
 	case EEH_RESET_FUNDAMENTAL:
 	case EEH_RESET_HOT:
-		pci_read_config_word(dev, PCI_BRIDGE_CONTROL, &ctrl);
+		/* Don't report linkDown event */
+		if (aer) {
+			eeh_ops->read_config(dn, aer + PCI_ERR_UNCOR_MASK,
+					     4, &ctrl);
+			ctrl |= PCI_ERR_UNC_SURPDN;
+                        eeh_ops->write_config(dn, aer + PCI_ERR_UNCOR_MASK,
+					      4, ctrl);
+                }
+
+		eeh_ops->read_config(dn, PCI_BRIDGE_CONTROL, 2, &ctrl);
 		ctrl |= PCI_BRIDGE_CTL_BUS_RESET;
-		pci_write_config_word(dev, PCI_BRIDGE_CONTROL, ctrl);
+		eeh_ops->write_config(dn, PCI_BRIDGE_CONTROL, 2, ctrl);
+		msleep(EEH_PE_RST_HOLD_TIME);
+
 		break;
 	case EEH_RESET_DEACTIVATE:
-		pci_read_config_word(dev, PCI_BRIDGE_CONTROL, &ctrl);
+		eeh_ops->read_config(dn, PCI_BRIDGE_CONTROL, 2, &ctrl);
 		ctrl &= ~PCI_BRIDGE_CTL_BUS_RESET;
-		pci_write_config_word(dev, PCI_BRIDGE_CONTROL, ctrl);
+		eeh_ops->write_config(dn, PCI_BRIDGE_CONTROL, 2, ctrl);
+		msleep(EEH_PE_RST_SETTLE_TIME);
+
+		/* Continue reporting linkDown event */
+		if (aer) {
+			eeh_ops->read_config(dn, aer + PCI_ERR_UNCOR_MASK,
+					     4, &ctrl);
+			ctrl &= ~PCI_ERR_UNC_SURPDN;
+			eeh_ops->write_config(dn, aer + PCI_ERR_UNCOR_MASK,
+					      4, ctrl);
+		}
+
 		break;
 	}
 
 	return 0;
 }
 
+void pnv_pci_reset_secondary_bus(struct pci_dev *dev)
+{
+	struct pci_controller *hose;
+
+	if (pci_is_root_bus(dev->bus)) {
+		hose = pci_bus_to_host(dev->bus);
+		ioda_eeh_root_reset(hose, EEH_RESET_HOT);
+		ioda_eeh_root_reset(hose, EEH_RESET_DEACTIVATE);
+	} else {
+		ioda_eeh_bridge_reset(dev, EEH_RESET_HOT);
+		ioda_eeh_bridge_reset(dev, EEH_RESET_DEACTIVATE);
+	}
+}
+
 /**
  * ioda_eeh_reset - Reset the indicated PE
  * @pe: EEH PE
@@ -523,27 +561,18 @@ static int ioda_eeh_reset(struct eeh_pe *pe, int option)
 	int ret;
 
 	/*
-	 * Anyway, we have to clear the problematic state for the
-	 * corresponding PE. However, we needn't do it if the PE
-	 * is PHB associated. That means the PHB is having fatal
-	 * errors and it needs reset. Further more, the AIB interface
-	 * isn't reliable any more.
-	 */
-	if (!(pe->type & EEH_PE_PHB) &&
-	    (option == EEH_RESET_HOT ||
-	    option == EEH_RESET_FUNDAMENTAL)) {
-		ret = ioda_eeh_pe_clear(pe);
-		if (ret)
-			return -EIO;
-	}
-
-	/*
-	 * The rules applied to reset, either fundamental or hot reset:
+	 * For PHB reset, we always have complete reset. For those PEs whose
+	 * primary bus derived from root complex (root bus) or root port
+	 * (usually bus#1), we apply hot or fundamental reset on the root port.
+	 * For other PEs, we always have hot reset on the PE primary bus.
 	 *
-	 * We always reset the direct upstream bridge of the PE. If the
-	 * direct upstream bridge isn't root bridge, we always take hot
-	 * reset no matter what option (fundamental or hot) is. Otherwise,
-	 * we should do the reset according to the required option.
+	 * Here, we have different design to pHyp, which always clear the
+	 * frozen state during PE reset. However, the good idea here from
+	 * benh is to keep frozen state before we get PE reset done completely
+	 * (until BAR restore). With the frozen state, HW drops illegal IO
+	 * or MMIO access, which can incur recrusive frozen PE during PE
+	 * reset. The side effect is that EEH core has to clear the frozen
+	 * state explicitly after BAR restore.
 	 */
 	if (pe->type & EEH_PE_PHB) {
 		ret = ioda_eeh_phb_reset(hose, option);
@@ -553,7 +582,7 @@ static int ioda_eeh_reset(struct eeh_pe *pe, int option)
 		    pci_is_root_bus(bus->parent))
 			ret = ioda_eeh_root_reset(hose, option);
 		else
-			ret = ioda_eeh_bridge_reset(hose, bus->self, option);
+			ret = ioda_eeh_bridge_reset(bus->self, option);
 	}
 
 	return ret;
@@ -640,22 +669,6 @@ static void ioda_eeh_hub_diag(struct pci_controller *hose)
 	}
 }
 
-static int ioda_eeh_get_phb_pe(struct pci_controller *hose,
-			       struct eeh_pe **pe)
-{
-	struct eeh_pe *phb_pe;
-
-	phb_pe = eeh_phb_pe_get(hose);
-	if (!phb_pe) {
-		pr_warning("%s Can't find PE for PHB#%d\n",
-			   __func__, hose->global_number);
-		return -EEXIST;
-	}
-
-	*pe = phb_pe;
-	return 0;
-}
-
 static int ioda_eeh_get_pe(struct pci_controller *hose,
 			   u16 pe_no, struct eeh_pe **pe)
 {
@@ -663,7 +676,8 @@ static int ioda_eeh_get_pe(struct pci_controller *hose,
 	struct eeh_dev dev;
 
 	/* Find the PHB PE */
-	if (ioda_eeh_get_phb_pe(hose, &phb_pe))
+	phb_pe = eeh_phb_pe_get(hose);
+	if (!phb_pe)
 		return -EEXIST;
 
 	/* Find the PE according to PE# */
@@ -691,6 +705,7 @@ static int ioda_eeh_next_error(struct eeh_pe **pe)
 {
 	struct pci_controller *hose;
 	struct pnv_phb *phb;
+	struct eeh_pe *phb_pe;
 	u64 frozen_pe_no;
 	u16 err_type, severity;
 	long rc;
@@ -707,10 +722,12 @@ static int ioda_eeh_next_error(struct eeh_pe **pe)
 	list_for_each_entry(hose, &hose_list, list_node) {
 		/*
 		 * If the subordinate PCI buses of the PHB has been
-		 * removed, we needn't take care of it any more.
+		 * removed or is exactly under error recovery, we
+		 * needn't take care of it any more.
 		 */
 		phb = hose->private_data;
-		if (phb->eeh_state & PNV_EEH_STATE_REMOVED)
+		phb_pe = eeh_phb_pe_get(hose);
+		if (!phb_pe || (phb_pe->state & EEH_PE_ISOLATED))
 			continue;
 
 		rc = opal_pci_next_error(phb->opal_id,
@@ -743,12 +760,6 @@ static int ioda_eeh_next_error(struct eeh_pe **pe)
 		switch (err_type) {
 		case OPAL_EEH_IOC_ERROR:
 			if (severity == OPAL_EEH_SEV_IOC_DEAD) {
-				list_for_each_entry(hose, &hose_list,
-						    list_node) {
-					phb = hose->private_data;
-					phb->eeh_state |= PNV_EEH_STATE_REMOVED;
-				}
-
 				pr_err("EEH: dead IOC detected\n");
 				ret = EEH_NEXT_ERR_DEAD_IOC;
 			} else if (severity == OPAL_EEH_SEV_INF) {
@@ -761,17 +772,12 @@ static int ioda_eeh_next_error(struct eeh_pe **pe)
 			break;
 		case OPAL_EEH_PHB_ERROR:
 			if (severity == OPAL_EEH_SEV_PHB_DEAD) {
-				if (ioda_eeh_get_phb_pe(hose, pe))
-					break;
-
+				*pe = phb_pe;
 				pr_err("EEH: dead PHB#%x detected\n",
 					hose->global_number);
-				phb->eeh_state |= PNV_EEH_STATE_REMOVED;
 				ret = EEH_NEXT_ERR_DEAD_PHB;
 			} else if (severity == OPAL_EEH_SEV_PHB_FENCED) {
-				if (ioda_eeh_get_phb_pe(hose, pe))
-					break;
-
+				*pe = phb_pe;
 				pr_err("EEH: fenced PHB#%x detected\n",
 					hose->global_number);
 				ret = EEH_NEXT_ERR_FENCED_PHB;
@@ -789,17 +795,21 @@ static int ioda_eeh_next_error(struct eeh_pe **pe)
 			 * If we can't find the corresponding PE, the
 			 * PEEV / PEST would be messy. So we force an
 			 * fenced PHB so that it can be recovered.
+			 *
+			 * If the PE has been marked as isolated, that
+			 * should have been removed permanently or in
+			 * progress with recovery. We needn't report
+			 * it again.
 			 */
 			if (ioda_eeh_get_pe(hose, frozen_pe_no, pe)) {
-				if (!ioda_eeh_get_phb_pe(hose, pe)) {
-					pr_err("EEH: Escalated fenced PHB#%x "
-					       "detected for PE#%llx\n",
-						hose->global_number,
-						frozen_pe_no);
-					ret = EEH_NEXT_ERR_FENCED_PHB;
-				} else {
-					ret = EEH_NEXT_ERR_NONE;
-				}
+				*pe = phb_pe;
+				pr_err("EEH: Escalated fenced PHB#%x "
+				       "detected for PE#%llx\n",
+					hose->global_number,
+					frozen_pe_no);
+				ret = EEH_NEXT_ERR_FENCED_PHB;
+			} else if ((*pe)->state & EEH_PE_ISOLATED) {
+				ret = EEH_NEXT_ERR_NONE;
 			} else {
 				pr_err("EEH: Frozen PE#%x on PHB#%x detected\n",
 					(*pe)->addr, (*pe)->phb->global_number);
diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
index a59788e83b8b..56a206f32f77 100644
--- a/arch/powerpc/platforms/powernv/eeh-powernv.c
+++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
@@ -126,6 +126,7 @@ static int powernv_eeh_dev_probe(struct pci_dev *dev, void *flag)
 	edev->mode	&= 0xFFFFFF00;
 	if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE)
 		edev->mode |= EEH_DEV_BRIDGE;
+	edev->pcix_cap = pci_find_capability(dev, PCI_CAP_ID_PCIX);
 	if (pci_is_pcie(dev)) {
 		edev->pcie_cap = pci_pcie_cap(dev);
 
@@ -133,6 +134,9 @@ static int powernv_eeh_dev_probe(struct pci_dev *dev, void *flag)
 			edev->mode |= EEH_DEV_ROOT_PORT;
 		else if (pci_pcie_type(dev) == PCI_EXP_TYPE_DOWNSTREAM)
 			edev->mode |= EEH_DEV_DS_PORT;
+
+		edev->aer_cap = pci_find_ext_capability(dev,
+							PCI_EXT_CAP_ID_ERR);
 	}
 
 	edev->config_addr	= ((dev->bus->number << 8) | dev->devfn);
diff --git a/arch/powerpc/platforms/powernv/opal-flash.c b/arch/powerpc/platforms/powernv/opal-flash.c
index dc487ff04704..5c21d9c07f45 100644
--- a/arch/powerpc/platforms/powernv/opal-flash.c
+++ b/arch/powerpc/platforms/powernv/opal-flash.c
@@ -20,6 +20,7 @@
 #include <linux/mm.h>
 #include <linux/vmalloc.h>
 #include <linux/pagemap.h>
+#include <linux/delay.h>
 
 #include <asm/opal.h>
 
@@ -130,7 +131,8 @@ static inline void opal_flash_validate(void)
 {
 	long ret;
 	void *buf = validate_flash_data.buf;
-	__be32 size, result;
+	__be32 size = cpu_to_be32(validate_flash_data.buf_size);
+	__be32 result;
 
 	ret = opal_validate_flash(__pa(buf), &size, &result);
 
@@ -290,11 +292,6 @@ static int opal_flash_update(int op)
 	/* First entry address */
 	addr = __pa(list);
 
-	pr_alert("FLASH: Image is %u bytes\n", image_data.size);
-	pr_alert("FLASH: Image update requested\n");
-	pr_alert("FLASH: Image will be updated during system reboot\n");
-	pr_alert("FLASH: This will take several minutes. Do not power off!\n");
-
 flash:
 	rc = opal_update_flash(addr);
 
@@ -302,6 +299,47 @@ invalid_img:
 	return rc;
 }
 
+/* Return CPUs to OPAL before starting FW update */
+static void flash_return_cpu(void *info)
+{
+	int cpu = smp_processor_id();
+
+	if (!cpu_online(cpu))
+		return;
+
+	/* Disable IRQ */
+	hard_irq_disable();
+
+	/* Return the CPU to OPAL */
+	opal_return_cpu();
+}
+
+/* This gets called just before system reboots */
+void opal_flash_term_callback(void)
+{
+	struct cpumask mask;
+
+	if (update_flash_data.status != FLASH_IMG_READY)
+		return;
+
+	pr_alert("FLASH: Flashing new firmware\n");
+	pr_alert("FLASH: Image is %u bytes\n", image_data.size);
+	pr_alert("FLASH: Performing flash and reboot/shutdown\n");
+	pr_alert("FLASH: This will take several minutes. Do not power off!\n");
+
+	/* Small delay to help getting the above message out */
+	msleep(500);
+
+	/* Return secondary CPUs to firmware */
+	cpumask_copy(&mask, cpu_online_mask);
+	cpumask_clear_cpu(smp_processor_id(), &mask);
+	if (!cpumask_empty(&mask))
+		smp_call_function_many(&mask,
+				       flash_return_cpu, NULL, false);
+	/* Hard disable interrupts */
+	hard_irq_disable();
+}
+
 /*
  * Show candidate image status
  */
diff --git a/arch/powerpc/platforms/powernv/opal-lpc.c b/arch/powerpc/platforms/powernv/opal-lpc.c
index 79d83cad3d67..f04b4d8aca5a 100644
--- a/arch/powerpc/platforms/powernv/opal-lpc.c
+++ b/arch/powerpc/platforms/powernv/opal-lpc.c
@@ -12,12 +12,17 @@
 #include <linux/kernel.h>
 #include <linux/of.h>
 #include <linux/bug.h>
+#include <linux/debugfs.h>
+#include <linux/io.h>
+#include <linux/slab.h>
 
 #include <asm/machdep.h>
 #include <asm/firmware.h>
 #include <asm/xics.h>
 #include <asm/opal.h>
 #include <asm/prom.h>
+#include <asm/uaccess.h>
+#include <asm/debug.h>
 
 static int opal_lpc_chip_id = -1;
 
@@ -176,6 +181,152 @@ static const struct ppc_pci_io opal_lpc_io = {
 	.outsl	= opal_lpc_outsl,
 };
 
+#ifdef CONFIG_DEBUG_FS
+struct lpc_debugfs_entry {
+	enum OpalLPCAddressType lpc_type;
+};
+
+static ssize_t lpc_debug_read(struct file *filp, char __user *ubuf,
+			      size_t count, loff_t *ppos)
+{
+	struct lpc_debugfs_entry *lpc = filp->private_data;
+	u32 data, pos, len, todo;
+	int rc;
+
+	if (!access_ok(VERIFY_WRITE, ubuf, count))
+		return -EFAULT;
+
+	todo = count;
+	while (todo) {
+		pos = *ppos;
+
+		/*
+		 * Select access size based on count and alignment and
+		 * access type. IO and MEM only support byte acceses,
+		 * FW supports all 3.
+		 */
+		len = 1;
+		if (lpc->lpc_type == OPAL_LPC_FW) {
+			if (todo > 3 && (pos & 3) == 0)
+				len = 4;
+			else if (todo > 1 && (pos & 1) == 0)
+				len = 2;
+		}
+		rc = opal_lpc_read(opal_lpc_chip_id, lpc->lpc_type, pos,
+				   &data, len);
+		if (rc)
+			return -ENXIO;
+		switch(len) {
+		case 4:
+			rc = __put_user((u32)data, (u32 __user *)ubuf);
+			break;
+		case 2:
+			rc = __put_user((u16)data, (u16 __user *)ubuf);
+			break;
+		default:
+			rc = __put_user((u8)data, (u8 __user *)ubuf);
+			break;
+		}
+		if (rc)
+			return -EFAULT;
+		*ppos += len;
+		ubuf += len;
+		todo -= len;
+	}
+
+	return count;
+}
+
+static ssize_t lpc_debug_write(struct file *filp, const char __user *ubuf,
+			       size_t count, loff_t *ppos)
+{
+	struct lpc_debugfs_entry *lpc = filp->private_data;
+	u32 data, pos, len, todo;
+	int rc;
+
+	if (!access_ok(VERIFY_READ, ubuf, count))
+		return -EFAULT;
+
+	todo = count;
+	while (todo) {
+		pos = *ppos;
+
+		/*
+		 * Select access size based on count and alignment and
+		 * access type. IO and MEM only support byte acceses,
+		 * FW supports all 3.
+		 */
+		len = 1;
+		if (lpc->lpc_type == OPAL_LPC_FW) {
+			if (todo > 3 && (pos & 3) == 0)
+				len = 4;
+			else if (todo > 1 && (pos & 1) == 0)
+				len = 2;
+		}
+		switch(len) {
+		case 4:
+			rc = __get_user(data, (u32 __user *)ubuf);
+			break;
+		case 2:
+			rc = __get_user(data, (u16 __user *)ubuf);
+			break;
+		default:
+			rc = __get_user(data, (u8 __user *)ubuf);
+			break;
+		}
+		if (rc)
+			return -EFAULT;
+
+		rc = opal_lpc_write(opal_lpc_chip_id, lpc->lpc_type, pos,
+				    data, len);
+		if (rc)
+			return -ENXIO;
+		*ppos += len;
+		ubuf += len;
+		todo -= len;
+	}
+
+	return count;
+}
+
+static const struct file_operations lpc_fops = {
+	.read =		lpc_debug_read,
+	.write =	lpc_debug_write,
+	.open =		simple_open,
+	.llseek =	default_llseek,
+};
+
+static int opal_lpc_debugfs_create_type(struct dentry *folder,
+					const char *fname,
+					enum OpalLPCAddressType type)
+{
+	struct lpc_debugfs_entry *entry;
+	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
+	if (!entry)
+		return -ENOMEM;
+	entry->lpc_type = type;
+	debugfs_create_file(fname, 0600, folder, entry, &lpc_fops);
+	return 0;
+}
+
+static int opal_lpc_init_debugfs(void)
+{
+	struct dentry *root;
+	int rc = 0;
+
+	if (opal_lpc_chip_id < 0)
+		return -ENODEV;
+
+	root = debugfs_create_dir("lpc", powerpc_debugfs_root);
+
+	rc |= opal_lpc_debugfs_create_type(root, "io", OPAL_LPC_IO);
+	rc |= opal_lpc_debugfs_create_type(root, "mem", OPAL_LPC_MEM);
+	rc |= opal_lpc_debugfs_create_type(root, "fw", OPAL_LPC_FW);
+	return rc;
+}
+device_initcall(opal_lpc_init_debugfs);
+#endif  /* CONFIG_DEBUG_FS */
+
 void opal_lpc_init(void)
 {
 	struct device_node *np;
diff --git a/arch/powerpc/platforms/powernv/opal-memory-errors.c b/arch/powerpc/platforms/powernv/opal-memory-errors.c
index ec4132239cdf..b17a34b695ef 100644
--- a/arch/powerpc/platforms/powernv/opal-memory-errors.c
+++ b/arch/powerpc/platforms/powernv/opal-memory-errors.c
@@ -47,12 +47,12 @@ static void handle_memory_error_event(struct OpalMemoryErrorData *merr_evt)
 		  __func__, merr_evt->type);
 	switch (merr_evt->type) {
 	case OPAL_MEM_ERR_TYPE_RESILIENCE:
-		paddr_start = merr_evt->u.resilience.physical_address_start;
-		paddr_end = merr_evt->u.resilience.physical_address_end;
+		paddr_start = be64_to_cpu(merr_evt->u.resilience.physical_address_start);
+		paddr_end = be64_to_cpu(merr_evt->u.resilience.physical_address_end);
 		break;
 	case OPAL_MEM_ERR_TYPE_DYN_DALLOC:
-		paddr_start = merr_evt->u.dyn_dealloc.physical_address_start;
-		paddr_end = merr_evt->u.dyn_dealloc.physical_address_end;
+		paddr_start = be64_to_cpu(merr_evt->u.dyn_dealloc.physical_address_start);
+		paddr_end = be64_to_cpu(merr_evt->u.dyn_dealloc.physical_address_end);
 		break;
 	default:
 		return;
diff --git a/arch/powerpc/platforms/powernv/opal-takeover.S b/arch/powerpc/platforms/powernv/opal-takeover.S
index 3cd262897c27..11a3169ee583 100644
--- a/arch/powerpc/platforms/powernv/opal-takeover.S
+++ b/arch/powerpc/platforms/powernv/opal-takeover.S
@@ -21,11 +21,13 @@
 _GLOBAL(opal_query_takeover)
 	mfcr	r0
 	stw	r0,8(r1)
+	stdu	r1,-STACKFRAMESIZE(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
+	addi	r1,r1,STACKFRAMESIZE
 	ld	r10,STK_PARAM(R3)(r1)
 	std	r4,0(r10)
 	ld	r10,STK_PARAM(R4)(r1)
diff --git a/arch/powerpc/platforms/powernv/opal-wrappers.S b/arch/powerpc/platforms/powernv/opal-wrappers.S
index f531ffe35b3e..4abbff22a61f 100644
--- a/arch/powerpc/platforms/powernv/opal-wrappers.S
+++ b/arch/powerpc/platforms/powernv/opal-wrappers.S
@@ -32,7 +32,7 @@
 	std	r12,PACASAVEDMSR(r13);	\
 	andc	r12,r12,r0;		\
 	mtmsrd	r12,1;			\
-	LOAD_REG_ADDR(r0,.opal_return);	\
+	LOAD_REG_ADDR(r0,opal_return);	\
 	mtlr	r0;			\
 	li	r0,MSR_DR|MSR_IR|MSR_LE;\
 	andc	r12,r12,r0;		\
@@ -44,7 +44,7 @@
 	mtspr	SPRN_HSRR0,r12;		\
 	hrfid
 
-_STATIC(opal_return)
+opal_return:
 	/*
 	 * Fixup endian on OPAL return... we should be able to simplify
 	 * this by instead converting the below trampoline to a set of
@@ -124,6 +124,7 @@ OPAL_CALL(opal_xscom_write,			OPAL_XSCOM_WRITE);
 OPAL_CALL(opal_lpc_read,			OPAL_LPC_READ);
 OPAL_CALL(opal_lpc_write,			OPAL_LPC_WRITE);
 OPAL_CALL(opal_return_cpu,			OPAL_RETURN_CPU);
+OPAL_CALL(opal_reinit_cpus,			OPAL_REINIT_CPUS);
 OPAL_CALL(opal_read_elog,			OPAL_ELOG_READ);
 OPAL_CALL(opal_send_ack_elog,			OPAL_ELOG_ACK);
 OPAL_CALL(opal_get_elog_size,			OPAL_ELOG_SIZE);
diff --git a/arch/powerpc/platforms/powernv/opal.c b/arch/powerpc/platforms/powernv/opal.c
index f343183add07..199975613fe9 100644
--- a/arch/powerpc/platforms/powernv/opal.c
+++ b/arch/powerpc/platforms/powernv/opal.c
@@ -57,6 +57,21 @@ static DEFINE_SPINLOCK(opal_notifier_lock);
 static uint64_t last_notified_mask = 0x0ul;
 static atomic_t opal_notifier_hold = ATOMIC_INIT(0);
 
+static void opal_reinit_cores(void)
+{
+	/* Do the actual re-init, This will clobber all FPRs, VRs, etc...
+	 *
+	 * It will preserve non volatile GPRs and HSPRG0/1. It will
+	 * also restore HIDs and other SPRs to their original value
+	 * but it might clobber a bunch.
+	 */
+#ifdef __BIG_ENDIAN__
+	opal_reinit_cpus(OPAL_REINIT_CPUS_HILE_BE);
+#else
+	opal_reinit_cpus(OPAL_REINIT_CPUS_HILE_LE);
+#endif
+}
+
 int __init early_init_dt_scan_opal(unsigned long node,
 				   const char *uname, int depth, void *data)
 {
@@ -96,6 +111,13 @@ int __init early_init_dt_scan_opal(unsigned long node,
 		printk("OPAL V1 detected !\n");
 	}
 
+	/* Reinit all cores with the right endian */
+	opal_reinit_cores();
+
+	/* Restore some bits */
+	if (cur_cpu_spec->cpu_restore)
+		cur_cpu_spec->cpu_restore();
+
 	return 1;
 }
 
diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c
index 98824aa99173..de19edeaa7a7 100644
--- a/arch/powerpc/platforms/powernv/pci-ioda.c
+++ b/arch/powerpc/platforms/powernv/pci-ioda.c
@@ -13,6 +13,7 @@
 
 #include <linux/kernel.h>
 #include <linux/pci.h>
+#include <linux/crash_dump.h>
 #include <linux/debugfs.h>
 #include <linux/delay.h>
 #include <linux/string.h>
@@ -663,15 +664,15 @@ static void pnv_pci_ioda_setup_dma_pe(struct pnv_phb *phb,
 		 * errors, and on the first pass the data will be a relative
 		 * bus number, print that out instead.
 		 */
-		tbl->it_busno = 0;
 		pe->tce_inval_reg_phys = be64_to_cpup(swinvp);
 		tbl->it_index = (unsigned long)ioremap(pe->tce_inval_reg_phys,
 				8);
-		tbl->it_type = TCE_PCI_SWINV_CREATE | TCE_PCI_SWINV_FREE |
-			       TCE_PCI_SWINV_PAIR;
+		tbl->it_type |= (TCE_PCI_SWINV_CREATE |
+				 TCE_PCI_SWINV_FREE   |
+				 TCE_PCI_SWINV_PAIR);
 	}
 	iommu_init_table(tbl, phb->hose->node);
-	iommu_register_group(tbl, pci_domain_nr(pe->pbus), pe->pe_number);
+	iommu_register_group(tbl, phb->hose->global_number, pe->pe_number);
 
 	if (pe->pdev)
 		set_iommu_table_base_and_group(&pe->pdev->dev, tbl);
@@ -793,14 +794,13 @@ static void pnv_pci_ioda2_setup_dma_pe(struct pnv_phb *phb,
 		 * errors, and on the first pass the data will be a relative
 		 * bus number, print that out instead.
 		 */
-		tbl->it_busno = 0;
 		pe->tce_inval_reg_phys = be64_to_cpup(swinvp);
 		tbl->it_index = (unsigned long)ioremap(pe->tce_inval_reg_phys,
 				8);
-		tbl->it_type = TCE_PCI_SWINV_CREATE | TCE_PCI_SWINV_FREE;
+		tbl->it_type |= (TCE_PCI_SWINV_CREATE | TCE_PCI_SWINV_FREE);
 	}
 	iommu_init_table(tbl, phb->hose->node);
-	iommu_register_group(tbl, pci_domain_nr(pe->pbus), pe->pe_number);
+	iommu_register_group(tbl, phb->hose->global_number, pe->pe_number);
 
 	if (pe->pdev)
 		set_iommu_table_base_and_group(&pe->pdev->dev, tbl);
@@ -1386,12 +1386,24 @@ void __init pnv_pci_init_ioda_phb(struct device_node *np,
 	ppc_md.pcibios_fixup = pnv_pci_ioda_fixup;
 	ppc_md.pcibios_enable_device_hook = pnv_pci_enable_device_hook;
 	ppc_md.pcibios_window_alignment = pnv_pci_window_alignment;
+	ppc_md.pcibios_reset_secondary_bus = pnv_pci_reset_secondary_bus;
 	pci_add_flags(PCI_REASSIGN_ALL_RSRC);
 
 	/* Reset IODA tables to a clean state */
 	rc = opal_pci_reset(phb_id, OPAL_PCI_IODA_TABLE_RESET, OPAL_ASSERT_RESET);
 	if (rc)
 		pr_warning("  OPAL Error %ld performing IODA table reset !\n", rc);
+
+	/* If we're running in kdump kerenl, the previous kerenl never
+	 * shutdown PCI devices correctly. We already got IODA table
+	 * cleaned out. So we have to issue PHB reset to stop all PCI
+	 * transactions from previous kerenl.
+	 */
+	if (is_kdump_kernel()) {
+		pr_info("  Issue PHB reset ...\n");
+		ioda_eeh_phb_reset(hose, EEH_RESET_FUNDAMENTAL);
+		ioda_eeh_phb_reset(hose, OPAL_DEASSERT_RESET);
+	}
 }
 
 void __init pnv_pci_init_ioda2_phb(struct device_node *np)
diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c
index 8518817dcdfd..eefbfcc3fd8c 100644
--- a/arch/powerpc/platforms/powernv/pci.c
+++ b/arch/powerpc/platforms/powernv/pci.c
@@ -131,65 +131,60 @@ static void pnv_pci_dump_p7ioc_diag_data(struct pci_controller *hose,
 	int i;
 
 	data = (struct OpalIoP7IOCPhbErrorData *)common;
-	pr_info("P7IOC PHB#%d Diag-data (Version: %d)\n\n",
+	pr_info("P7IOC PHB#%d Diag-data (Version: %d)\n",
 		hose->global_number, common->version);
 
 	if (data->brdgCtl)
-		pr_info("  brdgCtl:     %08x\n",
+		pr_info("brdgCtl:     %08x\n",
 			data->brdgCtl);
 	if (data->portStatusReg || data->rootCmplxStatus ||
 	    data->busAgentStatus)
-		pr_info("  UtlSts:      %08x %08x %08x\n",
+		pr_info("UtlSts:      %08x %08x %08x\n",
 			data->portStatusReg, data->rootCmplxStatus,
 			data->busAgentStatus);
 	if (data->deviceStatus || data->slotStatus   ||
 	    data->linkStatus   || data->devCmdStatus ||
 	    data->devSecStatus)
-		pr_info("  RootSts:     %08x %08x %08x %08x %08x\n",
+		pr_info("RootSts:     %08x %08x %08x %08x %08x\n",
 			data->deviceStatus, data->slotStatus,
 			data->linkStatus, data->devCmdStatus,
 			data->devSecStatus);
 	if (data->rootErrorStatus   || data->uncorrErrorStatus ||
 	    data->corrErrorStatus)
-		pr_info("  RootErrSts:  %08x %08x %08x\n",
+		pr_info("RootErrSts:  %08x %08x %08x\n",
 			data->rootErrorStatus, data->uncorrErrorStatus,
 			data->corrErrorStatus);
 	if (data->tlpHdr1 || data->tlpHdr2 ||
 	    data->tlpHdr3 || data->tlpHdr4)
-		pr_info("  RootErrLog:  %08x %08x %08x %08x\n",
+		pr_info("RootErrLog:  %08x %08x %08x %08x\n",
 			data->tlpHdr1, data->tlpHdr2,
 			data->tlpHdr3, data->tlpHdr4);
 	if (data->sourceId || data->errorClass ||
 	    data->correlator)
-		pr_info("  RootErrLog1: %08x %016llx %016llx\n",
+		pr_info("RootErrLog1: %08x %016llx %016llx\n",
 			data->sourceId, data->errorClass,
 			data->correlator);
 	if (data->p7iocPlssr || data->p7iocCsr)
-		pr_info("  PhbSts:      %016llx %016llx\n",
+		pr_info("PhbSts:      %016llx %016llx\n",
 			data->p7iocPlssr, data->p7iocCsr);
-	if (data->lemFir || data->lemErrorMask ||
-	    data->lemWOF)
-		pr_info("  Lem:         %016llx %016llx %016llx\n",
+	if (data->lemFir)
+		pr_info("Lem:         %016llx %016llx %016llx\n",
 			data->lemFir, data->lemErrorMask,
 			data->lemWOF);
-	if (data->phbErrorStatus || data->phbFirstErrorStatus ||
-	    data->phbErrorLog0   || data->phbErrorLog1)
-		pr_info("  PhbErr:      %016llx %016llx %016llx %016llx\n",
+	if (data->phbErrorStatus)
+		pr_info("PhbErr:      %016llx %016llx %016llx %016llx\n",
 			data->phbErrorStatus, data->phbFirstErrorStatus,
 			data->phbErrorLog0, data->phbErrorLog1);
-	if (data->mmioErrorStatus || data->mmioFirstErrorStatus ||
-	    data->mmioErrorLog0   || data->mmioErrorLog1)
-		pr_info("  OutErr:      %016llx %016llx %016llx %016llx\n",
+	if (data->mmioErrorStatus)
+		pr_info("OutErr:      %016llx %016llx %016llx %016llx\n",
 			data->mmioErrorStatus, data->mmioFirstErrorStatus,
 			data->mmioErrorLog0, data->mmioErrorLog1);
-	if (data->dma0ErrorStatus || data->dma0FirstErrorStatus ||
-	    data->dma0ErrorLog0   || data->dma0ErrorLog1)
-		pr_info("  InAErr:      %016llx %016llx %016llx %016llx\n",
+	if (data->dma0ErrorStatus)
+		pr_info("InAErr:      %016llx %016llx %016llx %016llx\n",
 			data->dma0ErrorStatus, data->dma0FirstErrorStatus,
 			data->dma0ErrorLog0, data->dma0ErrorLog1);
-	if (data->dma1ErrorStatus || data->dma1FirstErrorStatus ||
-	    data->dma1ErrorLog0   || data->dma1ErrorLog1)
-		pr_info("  InBErr:      %016llx %016llx %016llx %016llx\n",
+	if (data->dma1ErrorStatus)
+		pr_info("InBErr:      %016llx %016llx %016llx %016llx\n",
 			data->dma1ErrorStatus, data->dma1FirstErrorStatus,
 			data->dma1ErrorLog0, data->dma1ErrorLog1);
 
@@ -198,7 +193,7 @@ static void pnv_pci_dump_p7ioc_diag_data(struct pci_controller *hose,
 		    (data->pestB[i] >> 63) == 0)
 			continue;
 
-		pr_info("  PE[%3d] A/B: %016llx %016llx\n",
+		pr_info("PE[%3d] A/B: %016llx %016llx\n",
 			i, data->pestA[i], data->pestB[i]);
 	}
 }
@@ -210,69 +205,63 @@ static void pnv_pci_dump_phb3_diag_data(struct pci_controller *hose,
 	int i;
 
 	data = (struct OpalIoPhb3ErrorData*)common;
-	pr_info("PHB3 PHB#%d Diag-data (Version: %d)\n\n",
+	pr_info("PHB3 PHB#%d Diag-data (Version: %d)\n",
 		hose->global_number, common->version);
 	if (data->brdgCtl)
-		pr_info("  brdgCtl:     %08x\n",
+		pr_info("brdgCtl:     %08x\n",
 			data->brdgCtl);
 	if (data->portStatusReg || data->rootCmplxStatus ||
 	    data->busAgentStatus)
-		pr_info("  UtlSts:      %08x %08x %08x\n",
+		pr_info("UtlSts:      %08x %08x %08x\n",
 			data->portStatusReg, data->rootCmplxStatus,
 			data->busAgentStatus);
 	if (data->deviceStatus || data->slotStatus   ||
 	    data->linkStatus   || data->devCmdStatus ||
 	    data->devSecStatus)
-		pr_info("  RootSts:     %08x %08x %08x %08x %08x\n",
+		pr_info("RootSts:     %08x %08x %08x %08x %08x\n",
 			data->deviceStatus, data->slotStatus,
 			data->linkStatus, data->devCmdStatus,
 			data->devSecStatus);
 	if (data->rootErrorStatus || data->uncorrErrorStatus ||
 	    data->corrErrorStatus)
-		pr_info("  RootErrSts:  %08x %08x %08x\n",
+		pr_info("RootErrSts:  %08x %08x %08x\n",
 			data->rootErrorStatus, data->uncorrErrorStatus,
 			data->corrErrorStatus);
 	if (data->tlpHdr1 || data->tlpHdr2 ||
 	    data->tlpHdr3 || data->tlpHdr4)
-		pr_info("  RootErrLog:  %08x %08x %08x %08x\n",
+		pr_info("RootErrLog:  %08x %08x %08x %08x\n",
 			data->tlpHdr1, data->tlpHdr2,
 			data->tlpHdr3, data->tlpHdr4);
 	if (data->sourceId || data->errorClass ||
 	    data->correlator)
-		pr_info("  RootErrLog1: %08x %016llx %016llx\n",
+		pr_info("RootErrLog1: %08x %016llx %016llx\n",
 			data->sourceId, data->errorClass,
 			data->correlator);
-	if (data->nFir || data->nFirMask ||
-	    data->nFirWOF)
-		pr_info("  nFir:        %016llx %016llx %016llx\n",
+	if (data->nFir)
+		pr_info("nFir:        %016llx %016llx %016llx\n",
 			data->nFir, data->nFirMask,
 			data->nFirWOF);
 	if (data->phbPlssr || data->phbCsr)
-		pr_info("  PhbSts:      %016llx %016llx\n",
+		pr_info("PhbSts:      %016llx %016llx\n",
 			data->phbPlssr, data->phbCsr);
-	if (data->lemFir || data->lemErrorMask ||
-	    data->lemWOF)
-		pr_info("  Lem:         %016llx %016llx %016llx\n",
+	if (data->lemFir)
+		pr_info("Lem:         %016llx %016llx %016llx\n",
 			data->lemFir, data->lemErrorMask,
 			data->lemWOF);
-	if (data->phbErrorStatus || data->phbFirstErrorStatus ||
-	    data->phbErrorLog0   || data->phbErrorLog1)
-		pr_info("  PhbErr:      %016llx %016llx %016llx %016llx\n",
+	if (data->phbErrorStatus)
+		pr_info("PhbErr:      %016llx %016llx %016llx %016llx\n",
 			data->phbErrorStatus, data->phbFirstErrorStatus,
 			data->phbErrorLog0, data->phbErrorLog1);
-	if (data->mmioErrorStatus || data->mmioFirstErrorStatus ||
-	    data->mmioErrorLog0   || data->mmioErrorLog1)
-		pr_info("  OutErr:      %016llx %016llx %016llx %016llx\n",
+	if (data->mmioErrorStatus)
+		pr_info("OutErr:      %016llx %016llx %016llx %016llx\n",
 			data->mmioErrorStatus, data->mmioFirstErrorStatus,
 			data->mmioErrorLog0, data->mmioErrorLog1);
-	if (data->dma0ErrorStatus || data->dma0FirstErrorStatus ||
-	    data->dma0ErrorLog0   || data->dma0ErrorLog1)
-		pr_info("  InAErr:      %016llx %016llx %016llx %016llx\n",
+	if (data->dma0ErrorStatus)
+		pr_info("InAErr:      %016llx %016llx %016llx %016llx\n",
 			data->dma0ErrorStatus, data->dma0FirstErrorStatus,
 			data->dma0ErrorLog0, data->dma0ErrorLog1);
-	if (data->dma1ErrorStatus || data->dma1FirstErrorStatus ||
-	    data->dma1ErrorLog0   || data->dma1ErrorLog1)
-		pr_info("  InBErr:      %016llx %016llx %016llx %016llx\n",
+	if (data->dma1ErrorStatus)
+		pr_info("InBErr:      %016llx %016llx %016llx %016llx\n",
 			data->dma1ErrorStatus, data->dma1FirstErrorStatus,
 			data->dma1ErrorLog0, data->dma1ErrorLog1);
 
@@ -281,7 +270,7 @@ static void pnv_pci_dump_phb3_diag_data(struct pci_controller *hose,
 		    (data->pestB[i] >> 63) == 0)
 			continue;
 
-		pr_info("  PE[%3d] A/B: %016llx %016llx\n",
+		pr_info("PE[%3d] A/B: %016llx %016llx\n",
 			i, data->pestA[i], data->pestB[i]);
 	}
 }
@@ -384,9 +373,6 @@ int pnv_pci_cfg_read(struct device_node *dn,
 	struct pci_dn *pdn = PCI_DN(dn);
 	struct pnv_phb *phb = pdn->phb->private_data;
 	u32 bdfn = (pdn->busno << 8) | pdn->devfn;
-#ifdef CONFIG_EEH
-	struct eeh_pe *phb_pe = NULL;
-#endif
 	s64 rc;
 
 	switch (size) {
@@ -412,31 +398,9 @@ int pnv_pci_cfg_read(struct device_node *dn,
 	default:
 		return PCIBIOS_FUNC_NOT_SUPPORTED;
 	}
+
 	cfg_dbg("%s: bus: %x devfn: %x +%x/%x -> %08x\n",
 		__func__, pdn->busno, pdn->devfn, where, size, *val);
-
-	/*
-	 * Check if the specified PE has been put into frozen
-	 * state. On the other hand, we needn't do that while
-	 * the PHB has been put into frozen state because of
-	 * PHB-fatal errors.
-	 */
-#ifdef CONFIG_EEH
-	phb_pe = eeh_phb_pe_get(pdn->phb);
-	if (phb_pe && (phb_pe->state & EEH_PE_ISOLATED))
-		return PCIBIOS_SUCCESSFUL;
-
-	if (phb->eeh_state & PNV_EEH_STATE_ENABLED) {
-		if (*val == EEH_IO_ERROR_VALUE(size) &&
-		    eeh_dev_check_failure(of_node_to_eeh_dev(dn)))
-			return PCIBIOS_DEVICE_NOT_FOUND;
-	} else {
-		pnv_pci_config_check_eeh(phb, dn);
-	}
-#else
-	pnv_pci_config_check_eeh(phb, dn);
-#endif
-
 	return PCIBIOS_SUCCESSFUL;
 }
 
@@ -463,33 +427,74 @@ int pnv_pci_cfg_write(struct device_node *dn,
 		return PCIBIOS_FUNC_NOT_SUPPORTED;
 	}
 
-	/* Check if the PHB got frozen due to an error (no response) */
-#ifdef CONFIG_EEH
-	if (!(phb->eeh_state & PNV_EEH_STATE_ENABLED))
-		pnv_pci_config_check_eeh(phb, dn);
-#else
-	pnv_pci_config_check_eeh(phb, dn);
-#endif
-
 	return PCIBIOS_SUCCESSFUL;
 }
 
+#if CONFIG_EEH
+static bool pnv_pci_cfg_check(struct pci_controller *hose,
+			      struct device_node *dn)
+{
+	struct eeh_dev *edev = NULL;
+	struct pnv_phb *phb = hose->private_data;
+
+	/* EEH not enabled ? */
+	if (!(phb->flags & PNV_PHB_FLAG_EEH))
+		return true;
+
+	/* PE reset or device removed ? */
+	edev = of_node_to_eeh_dev(dn);
+	if (edev) {
+		if (edev->pe &&
+		    (edev->pe->state & EEH_PE_RESET))
+			return false;
+
+		if (edev->mode & EEH_DEV_REMOVED)
+			return false;
+	}
+
+	return true;
+}
+#else
+static inline pnv_pci_cfg_check(struct pci_controller *hose,
+				struct device_node *dn)
+{
+	return true;
+}
+#endif /* CONFIG_EEH */
+
 static int pnv_pci_read_config(struct pci_bus *bus,
 			       unsigned int devfn,
 			       int where, int size, u32 *val)
 {
 	struct device_node *dn, *busdn = pci_bus_to_OF_node(bus);
 	struct pci_dn *pdn;
+	struct pnv_phb *phb;
+	bool found = false;
+	int ret;
 
+	*val = 0xFFFFFFFF;
 	for (dn = busdn->child; dn; dn = dn->sibling) {
 		pdn = PCI_DN(dn);
-		if (pdn && pdn->devfn == devfn)
-			return pnv_pci_cfg_read(dn, where, size, val);
+		if (pdn && pdn->devfn == devfn) {
+			phb = pdn->phb->private_data;
+			found = true;
+			break;
+		}
 	}
 
-	*val = 0xFFFFFFFF;
-	return PCIBIOS_DEVICE_NOT_FOUND;
+	if (!found || !pnv_pci_cfg_check(pdn->phb, dn))
+		return PCIBIOS_DEVICE_NOT_FOUND;
 
+	ret = pnv_pci_cfg_read(dn, where, size, val);
+	if (phb->flags & PNV_PHB_FLAG_EEH) {
+		if (*val == EEH_IO_ERROR_VALUE(size) &&
+		    eeh_dev_check_failure(of_node_to_eeh_dev(dn)))
+                        return PCIBIOS_DEVICE_NOT_FOUND;
+	} else {
+		pnv_pci_config_check_eeh(phb, dn);
+	}
+
+	return ret;
 }
 
 static int pnv_pci_write_config(struct pci_bus *bus,
@@ -498,14 +503,27 @@ static int pnv_pci_write_config(struct pci_bus *bus,
 {
 	struct device_node *dn, *busdn = pci_bus_to_OF_node(bus);
 	struct pci_dn *pdn;
+	struct pnv_phb *phb;
+	bool found = false;
+	int ret;
 
 	for (dn = busdn->child; dn; dn = dn->sibling) {
 		pdn = PCI_DN(dn);
-		if (pdn && pdn->devfn == devfn)
-			return pnv_pci_cfg_write(dn, where, size, val);
+		if (pdn && pdn->devfn == devfn) {
+			phb = pdn->phb->private_data;
+			found = true;
+			break;
+		}
 	}
 
-	return PCIBIOS_DEVICE_NOT_FOUND;
+	if (!found || !pnv_pci_cfg_check(pdn->phb, dn))
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	ret = pnv_pci_cfg_write(dn, where, size, val);
+	if (!(phb->flags & PNV_PHB_FLAG_EEH))
+		pnv_pci_config_check_eeh(phb, dn);
+
+	return ret;
 }
 
 struct pci_ops pnv_pci_ops = {
diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h
index cde169442775..676232c34328 100644
--- a/arch/powerpc/platforms/powernv/pci.h
+++ b/arch/powerpc/platforms/powernv/pci.h
@@ -81,28 +81,27 @@ struct pnv_eeh_ops {
 	int (*configure_bridge)(struct eeh_pe *pe);
 	int (*next_error)(struct eeh_pe **pe);
 };
-
-#define PNV_EEH_STATE_ENABLED	(1 << 0)	/* EEH enabled	*/
-#define PNV_EEH_STATE_REMOVED	(1 << 1)	/* PHB removed	*/
-
 #endif /* CONFIG_EEH */
 
+#define PNV_PHB_FLAG_EEH	(1 << 0)
+
 struct pnv_phb {
 	struct pci_controller	*hose;
 	enum pnv_phb_type	type;
 	enum pnv_phb_model	model;
 	u64			hub_id;
 	u64			opal_id;
+	int			flags;
 	void __iomem		*regs;
 	int			initialized;
 	spinlock_t		lock;
 
 #ifdef CONFIG_EEH
 	struct pnv_eeh_ops	*eeh_ops;
-	int			eeh_state;
 #endif
 
 #ifdef CONFIG_DEBUG_FS
+	int			has_dbgfs;
 	struct dentry		*dbgfs;
 #endif
 
@@ -205,5 +204,7 @@ extern void pnv_pci_init_ioda_hub(struct device_node *np);
 extern void pnv_pci_init_ioda2_phb(struct device_node *np);
 extern void pnv_pci_ioda_tce_invalidate(struct iommu_table *tbl,
 					__be64 *startp, __be64 *endp, bool rm);
+extern void pnv_pci_reset_secondary_bus(struct pci_dev *dev);
+extern int ioda_eeh_phb_reset(struct pci_controller *hose, int option);
 
 #endif /* __POWERNV_PCI_H */
diff --git a/arch/powerpc/platforms/powernv/powernv.h b/arch/powerpc/platforms/powernv/powernv.h
index 0051e108ef0f..75501bfede7f 100644
--- a/arch/powerpc/platforms/powernv/powernv.h
+++ b/arch/powerpc/platforms/powernv/powernv.h
@@ -25,4 +25,6 @@ static inline int pnv_pci_dma_set_mask(struct pci_dev *pdev, u64 dma_mask)
 
 extern void pnv_lpc_init(void);
 
+bool cpu_core_split_required(void);
+
 #endif /* _POWERNV_H */
diff --git a/arch/powerpc/platforms/powernv/setup.c b/arch/powerpc/platforms/powernv/setup.c
index 8723d32632f5..8c16a5f96728 100644
--- a/arch/powerpc/platforms/powernv/setup.c
+++ b/arch/powerpc/platforms/powernv/setup.c
@@ -27,6 +27,7 @@
 #include <linux/interrupt.h>
 #include <linux/bug.h>
 #include <linux/pci.h>
+#include <linux/cpufreq.h>
 
 #include <asm/machdep.h>
 #include <asm/firmware.h>
@@ -98,11 +99,32 @@ static void pnv_show_cpuinfo(struct seq_file *m)
 	of_node_put(root);
 }
 
+static void pnv_prepare_going_down(void)
+{
+	/*
+	 * Disable all notifiers from OPAL, we can't
+	 * service interrupts anymore anyway
+	 */
+	opal_notifier_disable();
+
+	/* Soft disable interrupts */
+	local_irq_disable();
+
+	/*
+	 * Return secondary CPUs to firwmare if a flash update
+	 * is pending otherwise we will get all sort of error
+	 * messages about CPU being stuck etc.. This will also
+	 * have the side effect of hard disabling interrupts so
+	 * past this point, the kernel is effectively dead.
+	 */
+	opal_flash_term_callback();
+}
+
 static void  __noreturn pnv_restart(char *cmd)
 {
 	long rc = OPAL_BUSY;
 
-	opal_notifier_disable();
+	pnv_prepare_going_down();
 
 	while (rc == OPAL_BUSY || rc == OPAL_BUSY_EVENT) {
 		rc = opal_cec_reboot();
@@ -119,7 +141,7 @@ static void __noreturn pnv_power_off(void)
 {
 	long rc = OPAL_BUSY;
 
-	opal_notifier_disable();
+	pnv_prepare_going_down();
 
 	while (rc == OPAL_BUSY || rc == OPAL_BUSY_EVENT) {
 		rc = opal_cec_power_down(0);
@@ -222,6 +244,13 @@ static void pnv_kexec_cpu_down(int crash_shutdown, int secondary)
 }
 #endif /* CONFIG_KEXEC */
 
+#ifdef CONFIG_MEMORY_HOTPLUG_SPARSE
+static unsigned long pnv_memory_block_size(void)
+{
+	return 256UL * 1024 * 1024;
+}
+#endif
+
 static void __init pnv_setup_machdep_opal(void)
 {
 	ppc_md.get_boot_time = opal_get_boot_time;
@@ -269,6 +298,25 @@ static int __init pnv_probe(void)
 	return 1;
 }
 
+/*
+ * Returns the cpu frequency for 'cpu' in Hz. This is used by
+ * /proc/cpuinfo
+ */
+unsigned long pnv_get_proc_freq(unsigned int cpu)
+{
+	unsigned long ret_freq;
+
+	ret_freq = cpufreq_quick_get(cpu) * 1000ul;
+
+	/*
+	 * If the backend cpufreq driver does not exist,
+         * then fallback to old way of reporting the clockrate.
+	 */
+	if (!ret_freq)
+		ret_freq = ppc_proc_freq;
+	return ret_freq;
+}
+
 define_machine(powernv) {
 	.name			= "PowerNV",
 	.probe			= pnv_probe,
@@ -276,6 +324,7 @@ define_machine(powernv) {
 	.setup_arch		= pnv_setup_arch,
 	.init_IRQ		= pnv_init_IRQ,
 	.show_cpuinfo		= pnv_show_cpuinfo,
+	.get_proc_freq          = pnv_get_proc_freq,
 	.progress		= pnv_progress,
 	.machine_shutdown	= pnv_shutdown,
 	.power_save             = power7_idle,
@@ -284,4 +333,7 @@ define_machine(powernv) {
 #ifdef CONFIG_KEXEC
 	.kexec_cpu_down		= pnv_kexec_cpu_down,
 #endif
+#ifdef CONFIG_MEMORY_HOTPLUG_SPARSE
+	.memory_block_size	= pnv_memory_block_size,
+#endif
 };
diff --git a/arch/powerpc/platforms/powernv/smp.c b/arch/powerpc/platforms/powernv/smp.c
index bf5fcd452168..0062a43a2e0d 100644
--- a/arch/powerpc/platforms/powernv/smp.c
+++ b/arch/powerpc/platforms/powernv/smp.c
@@ -31,6 +31,7 @@
 #include <asm/xics.h>
 #include <asm/opal.h>
 #include <asm/runlatch.h>
+#include <asm/code-patching.h>
 
 #include "powernv.h"
 
@@ -50,8 +51,8 @@ static void pnv_smp_setup_cpu(int cpu)
 int 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));
+	unsigned long start_here =
+			__pa(ppc_function_entry(generic_secondary_smp_init));
 	long rc;
 
 	BUG_ON(nr < 0 || nr >= NR_CPUS);
@@ -158,17 +159,19 @@ static void pnv_smp_cpu_kill_self(void)
 	mtspr(SPRN_LPCR, mfspr(SPRN_LPCR) & ~(u64)LPCR_PECE1);
 	while (!generic_check_cpu_restart(cpu)) {
 		ppc64_runlatch_off();
-		power7_nap();
+		power7_nap(1);
 		ppc64_runlatch_on();
-		if (!generic_check_cpu_restart(cpu)) {
+
+		/* Reenable IRQs briefly to clear the IPI that woke us */
+		local_irq_enable();
+		local_irq_disable();
+		mb();
+
+		if (cpu_core_split_required())
+			continue;
+
+		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);
diff --git a/arch/powerpc/platforms/powernv/subcore-asm.S b/arch/powerpc/platforms/powernv/subcore-asm.S
new file mode 100644
index 000000000000..39bb24aa8f34
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/subcore-asm.S
@@ -0,0 +1,95 @@
+/*
+ * Copyright 2013, Michael (Ellerman|Neuling), IBM Corporation.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <asm/asm-offsets.h>
+#include <asm/ppc_asm.h>
+#include <asm/reg.h>
+
+#include "subcore.h"
+
+
+_GLOBAL(split_core_secondary_loop)
+	/*
+	 * r3 = u8 *state, used throughout the routine
+	 * r4 = temp
+	 * r5 = temp
+	 * ..
+	 * r12 = MSR
+	 */
+	mfmsr	r12
+
+	/* Disable interrupts so SRR0/1 don't get trashed */
+	li	r4,0
+	ori	r4,r4,MSR_EE|MSR_SE|MSR_BE|MSR_RI
+	andc	r4,r12,r4
+	sync
+	mtmsrd	r4
+
+	/* Switch to real mode and leave interrupts off */
+	li	r5, MSR_IR|MSR_DR
+	andc	r5, r4, r5
+
+	LOAD_REG_ADDR(r4, real_mode)
+
+	mtspr	SPRN_SRR0,r4
+	mtspr	SPRN_SRR1,r5
+	rfid
+	b	.	/* prevent speculative execution */
+
+real_mode:
+	/* Grab values from unsplit SPRs */
+	mfspr	r6,  SPRN_LDBAR
+	mfspr	r7,  SPRN_PMMAR
+	mfspr	r8,  SPRN_PMCR
+	mfspr	r9,  SPRN_RPR
+	mfspr	r10, SPRN_SDR1
+
+	/* Order reading the SPRs vs telling the primary we are ready to split */
+	sync
+
+	/* Tell thread 0 we are in real mode */
+	li	r4, SYNC_STEP_REAL_MODE
+	stb	r4, 0(r3)
+
+	li	r5, (HID0_POWER8_4LPARMODE | HID0_POWER8_2LPARMODE)@highest
+	sldi	r5, r5, 48
+
+	/* Loop until we see the split happen in HID0 */
+1:	mfspr	r4, SPRN_HID0
+	and.	r4, r4, r5
+	beq	1b
+
+	/*
+	 * We only need to initialise the below regs once for each subcore,
+	 * but it's simpler and harmless to do it on each thread.
+	 */
+
+	/* Make sure various SPRS have sane values */
+	li	r4, 0
+	mtspr	SPRN_LPID, r4
+	mtspr	SPRN_PCR, r4
+	mtspr	SPRN_HDEC, r4
+
+	/* Restore SPR values now we are split */
+	mtspr	SPRN_LDBAR, r6
+	mtspr	SPRN_PMMAR, r7
+	mtspr	SPRN_PMCR, r8
+	mtspr	SPRN_RPR, r9
+	mtspr	SPRN_SDR1, r10
+
+	LOAD_REG_ADDR(r5, virtual_mode)
+
+	/* Get out of real mode */
+	mtspr	SPRN_SRR0,r5
+	mtspr	SPRN_SRR1,r12
+	rfid
+	b	.	/* prevent speculative execution */
+
+virtual_mode:
+	blr
diff --git a/arch/powerpc/platforms/powernv/subcore.c b/arch/powerpc/platforms/powernv/subcore.c
new file mode 100644
index 000000000000..894ecb3eb596
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/subcore.c
@@ -0,0 +1,392 @@
+/*
+ * Copyright 2013, Michael (Ellerman|Neuling), IBM Corporation.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#define pr_fmt(fmt)	"powernv: " fmt
+
+#include <linux/kernel.h>
+#include <linux/cpu.h>
+#include <linux/cpumask.h>
+#include <linux/device.h>
+#include <linux/gfp.h>
+#include <linux/smp.h>
+#include <linux/stop_machine.h>
+
+#include <asm/cputhreads.h>
+#include <asm/kvm_ppc.h>
+#include <asm/machdep.h>
+#include <asm/opal.h>
+#include <asm/smp.h>
+
+#include "subcore.h"
+
+
+/*
+ * Split/unsplit procedure:
+ *
+ * A core can be in one of three states, unsplit, 2-way split, and 4-way split.
+ *
+ * The mapping to subcores_per_core is simple:
+ *
+ *  State       | subcores_per_core
+ *  ------------|------------------
+ *  Unsplit     |        1
+ *  2-way split |        2
+ *  4-way split |        4
+ *
+ * The core is split along thread boundaries, the mapping between subcores and
+ * threads is as follows:
+ *
+ *  Unsplit:
+ *          ----------------------------
+ *  Subcore |            0             |
+ *          ----------------------------
+ *  Thread  |  0  1  2  3  4  5  6  7  |
+ *          ----------------------------
+ *
+ *  2-way split:
+ *          -------------------------------------
+ *  Subcore |        0        |        1        |
+ *          -------------------------------------
+ *  Thread  |  0   1   2   3  |  4   5   6   7  |
+ *          -------------------------------------
+ *
+ *  4-way split:
+ *          -----------------------------------------
+ *  Subcore |    0    |    1    |    2    |    3    |
+ *          -----------------------------------------
+ *  Thread  |  0   1  |  2   3  |  4   5  |  6   7  |
+ *          -----------------------------------------
+ *
+ *
+ * Transitions
+ * -----------
+ *
+ * It is not possible to transition between either of the split states, the
+ * core must first be unsplit. The legal transitions are:
+ *
+ *  -----------          ---------------
+ *  |         |  <---->  | 2-way split |
+ *  |         |          ---------------
+ *  | Unsplit |
+ *  |         |          ---------------
+ *  |         |  <---->  | 4-way split |
+ *  -----------          ---------------
+ *
+ * Unsplitting
+ * -----------
+ *
+ * Unsplitting is the simpler procedure. It requires thread 0 to request the
+ * unsplit while all other threads NAP.
+ *
+ * Thread 0 clears HID0_POWER8_DYNLPARDIS (Dynamic LPAR Disable). This tells
+ * the hardware that if all threads except 0 are napping, the hardware should
+ * unsplit the core.
+ *
+ * Non-zero threads are sent to a NAP loop, they don't exit the loop until they
+ * see the core unsplit.
+ *
+ * Core 0 spins waiting for the hardware to see all the other threads napping
+ * and perform the unsplit.
+ *
+ * Once thread 0 sees the unsplit, it IPIs the secondary threads to wake them
+ * out of NAP. They will then see the core unsplit and exit the NAP loop.
+ *
+ * Splitting
+ * ---------
+ *
+ * The basic splitting procedure is fairly straight forward. However it is
+ * complicated by the fact that after the split occurs, the newly created
+ * subcores are not in a fully initialised state.
+ *
+ * Most notably the subcores do not have the correct value for SDR1, which
+ * means they must not be running in virtual mode when the split occurs. The
+ * subcores have separate timebases SPRs but these are pre-synchronised by
+ * opal.
+ *
+ * To begin with secondary threads are sent to an assembly routine. There they
+ * switch to real mode, so they are immune to the uninitialised SDR1 value.
+ * Once in real mode they indicate that they are in real mode, and spin waiting
+ * to see the core split.
+ *
+ * Thread 0 waits to see that all secondaries are in real mode, and then begins
+ * the splitting procedure. It firstly sets HID0_POWER8_DYNLPARDIS, which
+ * prevents the hardware from unsplitting. Then it sets the appropriate HID bit
+ * to request the split, and spins waiting to see that the split has happened.
+ *
+ * Concurrently the secondaries will notice the split. When they do they set up
+ * their SPRs, notably SDR1, and then they can return to virtual mode and exit
+ * the procedure.
+ */
+
+/* Initialised at boot by subcore_init() */
+static int subcores_per_core;
+
+/*
+ * Used to communicate to offline cpus that we want them to pop out of the
+ * offline loop and do a split or unsplit.
+ *
+ * 0 - no split happening
+ * 1 - unsplit in progress
+ * 2 - split to 2 in progress
+ * 4 - split to 4 in progress
+ */
+static int new_split_mode;
+
+static cpumask_var_t cpu_offline_mask;
+
+struct split_state {
+	u8 step;
+	u8 master;
+};
+
+static DEFINE_PER_CPU(struct split_state, split_state);
+
+static void wait_for_sync_step(int step)
+{
+	int i, cpu = smp_processor_id();
+
+	for (i = cpu + 1; i < cpu + threads_per_core; i++)
+		while(per_cpu(split_state, i).step < step)
+			barrier();
+
+	/* Order the wait loop vs any subsequent loads/stores. */
+	mb();
+}
+
+static void unsplit_core(void)
+{
+	u64 hid0, mask;
+	int i, cpu;
+
+	mask = HID0_POWER8_2LPARMODE | HID0_POWER8_4LPARMODE;
+
+	cpu = smp_processor_id();
+	if (cpu_thread_in_core(cpu) != 0) {
+		while (mfspr(SPRN_HID0) & mask)
+			power7_nap(0);
+
+		per_cpu(split_state, cpu).step = SYNC_STEP_UNSPLIT;
+		return;
+	}
+
+	hid0 = mfspr(SPRN_HID0);
+	hid0 &= ~HID0_POWER8_DYNLPARDIS;
+	mtspr(SPRN_HID0, hid0);
+
+	while (mfspr(SPRN_HID0) & mask)
+		cpu_relax();
+
+	/* Wake secondaries out of NAP */
+	for (i = cpu + 1; i < cpu + threads_per_core; i++)
+		smp_send_reschedule(i);
+
+	wait_for_sync_step(SYNC_STEP_UNSPLIT);
+}
+
+static void split_core(int new_mode)
+{
+	struct {  u64 value; u64 mask; } split_parms[2] = {
+		{ HID0_POWER8_1TO2LPAR, HID0_POWER8_2LPARMODE },
+		{ HID0_POWER8_1TO4LPAR, HID0_POWER8_4LPARMODE }
+	};
+	int i, cpu;
+	u64 hid0;
+
+	/* Convert new_mode (2 or 4) into an index into our parms array */
+	i = (new_mode >> 1) - 1;
+	BUG_ON(i < 0 || i > 1);
+
+	cpu = smp_processor_id();
+	if (cpu_thread_in_core(cpu) != 0) {
+		split_core_secondary_loop(&per_cpu(split_state, cpu).step);
+		return;
+	}
+
+	wait_for_sync_step(SYNC_STEP_REAL_MODE);
+
+	/* Write new mode */
+	hid0  = mfspr(SPRN_HID0);
+	hid0 |= HID0_POWER8_DYNLPARDIS | split_parms[i].value;
+	mtspr(SPRN_HID0, hid0);
+
+	/* Wait for it to happen */
+	while (!(mfspr(SPRN_HID0) & split_parms[i].mask))
+		cpu_relax();
+}
+
+static void cpu_do_split(int new_mode)
+{
+	/*
+	 * At boot subcores_per_core will be 0, so we will always unsplit at
+	 * boot. In the usual case where the core is already unsplit it's a
+	 * nop, and this just ensures the kernel's notion of the mode is
+	 * consistent with the hardware.
+	 */
+	if (subcores_per_core != 1)
+		unsplit_core();
+
+	if (new_mode != 1)
+		split_core(new_mode);
+
+	mb();
+	per_cpu(split_state, smp_processor_id()).step = SYNC_STEP_FINISHED;
+}
+
+bool cpu_core_split_required(void)
+{
+	smp_rmb();
+
+	if (!new_split_mode)
+		return false;
+
+	cpu_do_split(new_split_mode);
+
+	return true;
+}
+
+static int cpu_update_split_mode(void *data)
+{
+	int cpu, new_mode = *(int *)data;
+
+	if (this_cpu_ptr(&split_state)->master) {
+		new_split_mode = new_mode;
+		smp_wmb();
+
+		cpumask_andnot(cpu_offline_mask, cpu_present_mask,
+			       cpu_online_mask);
+
+		/* This should work even though the cpu is offline */
+		for_each_cpu(cpu, cpu_offline_mask)
+			smp_send_reschedule(cpu);
+	}
+
+	cpu_do_split(new_mode);
+
+	if (this_cpu_ptr(&split_state)->master) {
+		/* Wait for all cpus to finish before we touch subcores_per_core */
+		for_each_present_cpu(cpu) {
+			if (cpu >= setup_max_cpus)
+				break;
+
+			while(per_cpu(split_state, cpu).step < SYNC_STEP_FINISHED)
+				barrier();
+		}
+
+		new_split_mode = 0;
+
+		/* Make the new mode public */
+		subcores_per_core = new_mode;
+		threads_per_subcore = threads_per_core / subcores_per_core;
+
+		/* Make sure the new mode is written before we exit */
+		mb();
+	}
+
+	return 0;
+}
+
+static int set_subcores_per_core(int new_mode)
+{
+	struct split_state *state;
+	int cpu;
+
+	if (kvm_hv_mode_active()) {
+		pr_err("Unable to change split core mode while KVM active.\n");
+		return -EBUSY;
+	}
+
+	/*
+	 * We are only called at boot, or from the sysfs write. If that ever
+	 * changes we'll need a lock here.
+	 */
+	BUG_ON(new_mode < 1 || new_mode > 4 || new_mode == 3);
+
+	for_each_present_cpu(cpu) {
+		state = &per_cpu(split_state, cpu);
+		state->step = SYNC_STEP_INITIAL;
+		state->master = 0;
+	}
+
+	get_online_cpus();
+
+	/* This cpu will update the globals before exiting stop machine */
+	this_cpu_ptr(&split_state)->master = 1;
+
+	/* Ensure state is consistent before we call the other cpus */
+	mb();
+
+	stop_machine(cpu_update_split_mode, &new_mode, cpu_online_mask);
+
+	put_online_cpus();
+
+	return 0;
+}
+
+static ssize_t __used store_subcores_per_core(struct device *dev,
+		struct device_attribute *attr, const char *buf,
+		size_t count)
+{
+	unsigned long val;
+	int rc;
+
+	/* We are serialised by the attribute lock */
+
+	rc = sscanf(buf, "%lx", &val);
+	if (rc != 1)
+		return -EINVAL;
+
+	switch (val) {
+	case 1:
+	case 2:
+	case 4:
+		if (subcores_per_core == val)
+			/* Nothing to do */
+			goto out;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	rc = set_subcores_per_core(val);
+	if (rc)
+		return rc;
+
+out:
+	return count;
+}
+
+static ssize_t show_subcores_per_core(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	return sprintf(buf, "%x\n", subcores_per_core);
+}
+
+static DEVICE_ATTR(subcores_per_core, 0644,
+		show_subcores_per_core, store_subcores_per_core);
+
+static int subcore_init(void)
+{
+	if (!cpu_has_feature(CPU_FTR_ARCH_207S))
+		return 0;
+
+	/*
+	 * We need all threads in a core to be present to split/unsplit so
+         * continue only if max_cpus are aligned to threads_per_core.
+	 */
+	if (setup_max_cpus % threads_per_core)
+		return 0;
+
+	BUG_ON(!alloc_cpumask_var(&cpu_offline_mask, GFP_KERNEL));
+
+	set_subcores_per_core(1);
+
+	return device_create_file(cpu_subsys.dev_root,
+				  &dev_attr_subcores_per_core);
+}
+machine_device_initcall(powernv, subcore_init);
diff --git a/arch/powerpc/platforms/powernv/subcore.h b/arch/powerpc/platforms/powernv/subcore.h
new file mode 100644
index 000000000000..148abc91debf
--- /dev/null
+++ b/arch/powerpc/platforms/powernv/subcore.h
@@ -0,0 +1,18 @@
+/*
+ * Copyright 2013, Michael Ellerman, IBM Corporation.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+/* These are ordered and tested with <= */
+#define SYNC_STEP_INITIAL	0
+#define SYNC_STEP_UNSPLIT	1	/* Set by secondary when it sees unsplit */
+#define SYNC_STEP_REAL_MODE	2	/* Set by secondary when in real mode  */
+#define SYNC_STEP_FINISHED	3	/* Set by secondary when split/unsplit is done */
+
+#ifndef __ASSEMBLY__
+void split_core_secondary_loop(u8 *state);
+#endif