Merge branch 'master'

19 files changed, 534 insertions, 264 deletions

diff --git a/arch/powerpc/platforms/iseries/htab.c b/arch/powerpc/platforms/iseries/htab.c
index b3c6c3374ca..30bdcf3925d 100644
--- a/arch/powerpc/platforms/iseries/htab.c
+++ b/arch/powerpc/platforms/iseries/htab.c
@@ -39,15 +39,16 @@ static inline void iSeries_hunlock(unsigned long slot)
 	spin_unlock(&iSeries_hlocks[(slot >> 4) & 0x3f]);
 }
 
-static long iSeries_hpte_insert(unsigned long hpte_group, unsigned long va,
-				unsigned long prpn, unsigned long vflags,
-				unsigned long rflags)
+long iSeries_hpte_insert(unsigned long hpte_group, unsigned long va,
+			 unsigned long pa, unsigned long rflags,
+			 unsigned long vflags, int psize)
 {
-	unsigned long arpn;
 	long slot;
 	hpte_t lhpte;
 	int secondary = 0;
 
+	BUG_ON(psize != MMU_PAGE_4K);
+
 	/*
 	 * The hypervisor tries both primary and secondary.
 	 * If we are being called to insert in the secondary,
@@ -59,8 +60,19 @@ static long iSeries_hpte_insert(unsigned long hpte_group, unsigned long va,
 
 	iSeries_hlock(hpte_group);
 
-	slot = HvCallHpt_findValid(&lhpte, va >> PAGE_SHIFT);
-	BUG_ON(lhpte.v & HPTE_V_VALID);
+	slot = HvCallHpt_findValid(&lhpte, va >> HW_PAGE_SHIFT);
+	if (unlikely(lhpte.v & HPTE_V_VALID)) {
+		if (vflags & HPTE_V_BOLTED) {
+			HvCallHpt_setSwBits(slot, 0x10, 0);
+			HvCallHpt_setPp(slot, PP_RWXX);
+			iSeries_hunlock(hpte_group);
+			if (slot < 0)
+				return 0x8 | (slot & 7);
+			else
+				return slot & 7;
+		}
+		BUG();
+	}
 
 	if (slot == -1)	{ /* No available entry found in either group */
 		iSeries_hunlock(hpte_group);
@@ -73,10 +85,9 @@ static long iSeries_hpte_insert(unsigned long hpte_group, unsigned long va,
 		slot &= 0x7fffffffffffffff;
 	}
 
-	arpn = phys_to_abs(prpn << PAGE_SHIFT) >> PAGE_SHIFT;
 
-	lhpte.v = (va >> 23) << HPTE_V_AVPN_SHIFT | vflags | HPTE_V_VALID;
-	lhpte.r = (arpn << HPTE_R_RPN_SHIFT) | rflags;
+ 	lhpte.v = hpte_encode_v(va, MMU_PAGE_4K) | vflags | HPTE_V_VALID;
+	lhpte.r = hpte_encode_r(phys_to_abs(pa), MMU_PAGE_4K) | rflags;
 
 	/* Now fill in the actual HPTE */
 	HvCallHpt_addValidate(slot, secondary, &lhpte);
@@ -86,25 +97,6 @@ static long iSeries_hpte_insert(unsigned long hpte_group, unsigned long va,
 	return (secondary << 3) | (slot & 7);
 }
 
-long iSeries_hpte_bolt_or_insert(unsigned long hpte_group,
-		unsigned long va, unsigned long prpn, unsigned long vflags,
-		unsigned long rflags)
-{
-	long slot;
-	hpte_t lhpte;
-
-	slot = HvCallHpt_findValid(&lhpte, va >> PAGE_SHIFT);
-
-	if (lhpte.v & HPTE_V_VALID) {
-		/* Bolt the existing HPTE */
-		HvCallHpt_setSwBits(slot, 0x10, 0);
-		HvCallHpt_setPp(slot, PP_RWXX);
-		return 0;
-	}
-
-	return iSeries_hpte_insert(hpte_group, va, prpn, vflags, rflags);
-}
-
 static unsigned long iSeries_hpte_getword0(unsigned long slot)
 {
 	hpte_t hpte;
@@ -150,15 +142,17 @@ static long iSeries_hpte_remove(unsigned long hpte_group)
  *	bits 61..63 : PP2,PP1,PP0
  */
 static long iSeries_hpte_updatepp(unsigned long slot, unsigned long newpp,
-				  unsigned long va, int large, int local)
+				  unsigned long va, int psize, int local)
 {
 	hpte_t hpte;
-	unsigned long avpn = va >> 23;
+	unsigned long want_v;
 
 	iSeries_hlock(slot);
 
 	HvCallHpt_get(&hpte, slot);
-	if ((HPTE_V_AVPN_VAL(hpte.v) == avpn) && (hpte.v & HPTE_V_VALID)) {
+	want_v = hpte_encode_v(va, MMU_PAGE_4K);
+
+	if (HPTE_V_COMPARE(hpte.v, want_v) && (hpte.v & HPTE_V_VALID)) {
 		/*
 		 * Hypervisor expects bits as NPPP, which is
 		 * different from how they are mapped in our PP.
@@ -210,14 +204,17 @@ static long iSeries_hpte_find(unsigned long vpn)
  *
  * No need to lock here because we should be the only user.
  */
-static void iSeries_hpte_updateboltedpp(unsigned long newpp, unsigned long ea)
+static void iSeries_hpte_updateboltedpp(unsigned long newpp, unsigned long ea,
+					int psize)
 {
 	unsigned long vsid,va,vpn;
 	long slot;
 
+	BUG_ON(psize != MMU_PAGE_4K);
+
 	vsid = get_kernel_vsid(ea);
 	va = (vsid << 28) | (ea & 0x0fffffff);
-	vpn = va >> PAGE_SHIFT;
+	vpn = va >> HW_PAGE_SHIFT;
 	slot = iSeries_hpte_find(vpn);
 	if (slot == -1)
 		panic("updateboltedpp: Could not find page to bolt\n");
@@ -225,7 +222,7 @@ static void iSeries_hpte_updateboltedpp(unsigned long newpp, unsigned long ea)
 }
 
 static void iSeries_hpte_invalidate(unsigned long slot, unsigned long va,
-				    int large, int local)
+				    int psize, int local)
 {
 	unsigned long hpte_v;
 	unsigned long avpn = va >> 23;
diff --git a/arch/powerpc/platforms/iseries/hvlog.c b/arch/powerpc/platforms/iseries/hvlog.c
index 62ec7347968..f476d71194f 100644
--- a/arch/powerpc/platforms/iseries/hvlog.c
+++ b/arch/powerpc/platforms/iseries/hvlog.c
@@ -22,7 +22,7 @@ void HvCall_writeLogBuffer(const void *buffer, u64 len)
 
 	while (len) {
 		hv_buf.addr = cur;
-		left_this_page = ((cur & PAGE_MASK) + PAGE_SIZE) - cur;
+		left_this_page = ((cur & HW_PAGE_MASK) + HW_PAGE_SIZE) - cur;
 		if (left_this_page > len)
 			left_this_page = len;
 		hv_buf.len = left_this_page;
@@ -30,6 +30,6 @@ void HvCall_writeLogBuffer(const void *buffer, u64 len)
 		HvCall2(HvCallBaseWriteLogBuffer,
 				virt_to_abs(&hv_buf),
 				left_this_page);
-		cur = (cur & PAGE_MASK) + PAGE_SIZE;
+		cur = (cur & HW_PAGE_MASK) + HW_PAGE_SIZE;
 	}
 }
diff --git a/arch/powerpc/platforms/iseries/iommu.c b/arch/powerpc/platforms/iseries/iommu.c
index 1a6845b5c5a..bf081b34582 100644
--- a/arch/powerpc/platforms/iseries/iommu.c
+++ b/arch/powerpc/platforms/iseries/iommu.c
@@ -43,9 +43,12 @@ static void tce_build_iSeries(struct iommu_table *tbl, long index, long npages,
 	u64 rc;
 	union tce_entry tce;
 
+	index <<= TCE_PAGE_FACTOR;
+	npages <<= TCE_PAGE_FACTOR;
+
 	while (npages--) {
 		tce.te_word = 0;
-		tce.te_bits.tb_rpn = virt_to_abs(uaddr) >> PAGE_SHIFT;
+		tce.te_bits.tb_rpn = virt_to_abs(uaddr) >> TCE_SHIFT;
 
 		if (tbl->it_type == TCE_VB) {
 			/* Virtual Bus */
@@ -66,7 +69,7 @@ static void tce_build_iSeries(struct iommu_table *tbl, long index, long npages,
 			panic("PCI_DMA: HvCallXm_setTce failed, Rc: 0x%lx\n",
 					rc);
 		index++;
-		uaddr += PAGE_SIZE;
+		uaddr += TCE_PAGE_SIZE;
 	}
 }
 
@@ -74,6 +77,9 @@ static void tce_free_iSeries(struct iommu_table *tbl, long index, long npages)
 {
 	u64 rc;
 
+	npages <<= TCE_PAGE_FACTOR;
+	index <<= TCE_PAGE_FACTOR;
+
 	while (npages--) {
 		rc = HvCallXm_setTce((u64)tbl->it_index, (u64)index, 0);
 		if (rc)
@@ -83,27 +89,6 @@ static void tce_free_iSeries(struct iommu_table *tbl, long index, long npages)
 	}
 }
 
-#ifdef CONFIG_PCI
-/*
- * This function compares the known tables to find an iommu_table
- * that has already been built for hardware TCEs.
- */
-static struct iommu_table *iommu_table_find(struct iommu_table * tbl)
-{
-	struct pci_dn *pdn;
-
-	list_for_each_entry(pdn, &iSeries_Global_Device_List, Device_List) {
-		struct iommu_table *it = pdn->iommu_table;
-		if ((it != NULL) &&
-		    (it->it_type == TCE_PCI) &&
-		    (it->it_offset == tbl->it_offset) &&
-		    (it->it_index == tbl->it_index) &&
-		    (it->it_size == tbl->it_size))
-			return it;
-	}
-	return NULL;
-}
-
 /*
  * Call Hv with the architected data structure to get TCE table info.
  * info. Put the returned data into the Linux representation of the
@@ -113,8 +98,10 @@ static struct iommu_table *iommu_table_find(struct iommu_table * tbl)
  * 2. TCE table per Bus.
  * 3. TCE Table per IOA.
  */
-static void iommu_table_getparms(struct pci_dn *pdn,
-				 struct iommu_table* tbl)
+void iommu_table_getparms_iSeries(unsigned long busno,
+				  unsigned char slotno,
+				  unsigned char virtbus,
+				  struct iommu_table* tbl)
 {
 	struct iommu_table_cb *parms;
 
@@ -124,9 +111,9 @@ static void iommu_table_getparms(struct pci_dn *pdn,
 
 	memset(parms, 0, sizeof(*parms));
 
-	parms->itc_busno = pdn->busno;
-	parms->itc_slotno = pdn->LogicalSlot;
-	parms->itc_virtbus = 0;
+	parms->itc_busno = busno;
+	parms->itc_slotno = slotno;
+	parms->itc_virtbus = virtbus;
 
 	HvCallXm_getTceTableParms(iseries_hv_addr(parms));
 
@@ -134,17 +121,40 @@ static void iommu_table_getparms(struct pci_dn *pdn,
 		panic("PCI_DMA: parms->size is zero, parms is 0x%p", parms);
 
 	/* itc_size is in pages worth of table, it_size is in # of entries */
-	tbl->it_size = (parms->itc_size * PAGE_SIZE) / sizeof(union tce_entry);
+	tbl->it_size = ((parms->itc_size * TCE_PAGE_SIZE) /
+			sizeof(union tce_entry)) >> TCE_PAGE_FACTOR;
 	tbl->it_busno = parms->itc_busno;
-	tbl->it_offset = parms->itc_offset;
+	tbl->it_offset = parms->itc_offset >> TCE_PAGE_FACTOR;
 	tbl->it_index = parms->itc_index;
 	tbl->it_blocksize = 1;
-	tbl->it_type = TCE_PCI;
+	tbl->it_type = virtbus ? TCE_VB : TCE_PCI;
 
 	kfree(parms);
 }
 
 
+#ifdef CONFIG_PCI
+/*
+ * This function compares the known tables to find an iommu_table
+ * that has already been built for hardware TCEs.
+ */
+static struct iommu_table *iommu_table_find(struct iommu_table * tbl)
+{
+	struct pci_dn *pdn;
+
+	list_for_each_entry(pdn, &iSeries_Global_Device_List, Device_List) {
+		struct iommu_table *it = pdn->iommu_table;
+		if ((it != NULL) &&
+		    (it->it_type == TCE_PCI) &&
+		    (it->it_offset == tbl->it_offset) &&
+		    (it->it_index == tbl->it_index) &&
+		    (it->it_size == tbl->it_size))
+			return it;
+	}
+	return NULL;
+}
+
+
 void iommu_devnode_init_iSeries(struct device_node *dn)
 {
 	struct iommu_table *tbl;
@@ -152,7 +162,7 @@ void iommu_devnode_init_iSeries(struct device_node *dn)
 
 	tbl = kmalloc(sizeof(struct iommu_table), GFP_KERNEL);
 
-	iommu_table_getparms(pdn, tbl);
+	iommu_table_getparms_iSeries(pdn->busno, pdn->LogicalSlot, 0, tbl);
 
 	/* Look for existing tce table */
 	pdn->iommu_table = iommu_table_find(tbl);
diff --git a/arch/powerpc/platforms/iseries/irq.c b/arch/powerpc/platforms/iseries/irq.c
index c1135912cc0..a06603d84a4 100644
--- a/arch/powerpc/platforms/iseries/irq.c
+++ b/arch/powerpc/platforms/iseries/irq.c
@@ -35,7 +35,6 @@
 #include <linux/irq.h>
 #include <linux/spinlock.h>
 
-#include <asm/ppcdebug.h>
 #include <asm/iseries/hv_types.h>
 #include <asm/iseries/hv_lp_event.h>
 #include <asm/iseries/hv_call_xm.h>
@@ -227,8 +226,6 @@ static void iSeries_enable_IRQ(unsigned int irq)
 	/* Unmask secondary INTA */
 	mask = 0x80000000;
 	HvCallPci_unmaskInterrupts(bus, subBus, deviceId, mask);
-	PPCDBG(PPCDBG_BUSWALK, "iSeries_enable_IRQ 0x%02X.%02X.%02X 0x%04X\n",
-			bus, subBus, deviceId, irq);
 }
 
 /* This is called by iSeries_activate_IRQs */
@@ -310,8 +307,6 @@ static void iSeries_disable_IRQ(unsigned int irq)
 	/* Mask secondary INTA   */
 	mask = 0x80000000;
 	HvCallPci_maskInterrupts(bus, subBus, deviceId, mask);
-	PPCDBG(PPCDBG_BUSWALK, "iSeries_disable_IRQ 0x%02X.%02X.%02X 0x%04X\n",
-			bus, subBus, deviceId, irq);
 }
 
 /*
diff --git a/arch/powerpc/platforms/iseries/pci.c b/arch/powerpc/platforms/iseries/pci.c
index 7d7d5884343..4b75131773a 100644
--- a/arch/powerpc/platforms/iseries/pci.c
+++ b/arch/powerpc/platforms/iseries/pci.c
@@ -32,7 +32,6 @@
 #include <asm/prom.h>
 #include <asm/machdep.h>
 #include <asm/pci-bridge.h>
-#include <asm/ppcdebug.h>
 #include <asm/iommu.h>
 #include <asm/abs_addr.h>
 
@@ -207,10 +206,6 @@ static struct device_node *build_device_node(HvBusNumber Bus,
 	struct device_node *node;
 	struct pci_dn *pdn;
 
-	PPCDBG(PPCDBG_BUSWALK,
-			"-build_device_node 0x%02X.%02X.%02X Function: %02X\n",
-			Bus, SubBus, AgentId, Function);
-
 	node = kmalloc(sizeof(struct device_node), GFP_KERNEL);
 	if (node == NULL)
 		return NULL;
@@ -243,8 +238,6 @@ unsigned long __init find_and_init_phbs(void)
 	struct pci_controller *phb;
 	HvBusNumber bus;
 
-	PPCDBG(PPCDBG_BUSWALK, "find_and_init_phbs Entry\n");
-
 	/* Check all possible buses. */
 	for (bus = 0; bus < 256; bus++) {
 		int ret = HvCallXm_testBus(bus);
@@ -261,9 +254,6 @@ unsigned long __init find_and_init_phbs(void)
 			phb->last_busno = bus;
 			phb->ops = &iSeries_pci_ops;
 
-			PPCDBG(PPCDBG_BUSWALK, "PCI:Create iSeries pci_controller(%p), Bus: %04X\n",
-					phb, bus);
-
 			/* Find and connect the devices. */
 			scan_PHB_slots(phb);
 		}
@@ -285,11 +275,9 @@ unsigned long __init find_and_init_phbs(void)
  */
 void iSeries_pcibios_init(void)
 {
-	PPCDBG(PPCDBG_BUSWALK, "iSeries_pcibios_init Entry.\n");
 	iomm_table_initialize();
 	find_and_init_phbs();
 	io_page_mask = -1;
-	PPCDBG(PPCDBG_BUSWALK, "iSeries_pcibios_init Exit.\n");
 }
 
 /*
@@ -301,8 +289,6 @@ void __init iSeries_pci_final_fixup(void)
 	struct device_node *node;
 	int DeviceCount = 0;
 
-	PPCDBG(PPCDBG_BUSWALK, "iSeries_pcibios_fixup Entry.\n");
-
 	/* Fix up at the device node and pci_dev relationship */
 	mf_display_src(0xC9000100);
 
@@ -316,9 +302,6 @@ void __init iSeries_pci_final_fixup(void)
 			++DeviceCount;
 			pdev->sysdata = (void *)node;
 			PCI_DN(node)->pcidev = pdev;
-			PPCDBG(PPCDBG_BUSWALK,
-					"pdev 0x%p <==> DevNode 0x%p\n",
-					pdev, node);
 			allocate_device_bars(pdev);
 			iSeries_Device_Information(pdev, DeviceCount);
 			iommu_devnode_init_iSeries(node);
@@ -333,13 +316,10 @@ void __init iSeries_pci_final_fixup(void)
 
 void pcibios_fixup_bus(struct pci_bus *PciBus)
 {
-	PPCDBG(PPCDBG_BUSWALK, "iSeries_pcibios_fixup_bus(0x%04X) Entry.\n",
-			PciBus->number);
 }
 
 void pcibios_fixup_resources(struct pci_dev *pdev)
 {
-	PPCDBG(PPCDBG_BUSWALK, "fixup_resources pdev %p\n", pdev);
 }
 
 /*
@@ -401,9 +381,6 @@ static void scan_EADS_bridge(HvBusNumber bus, HvSubBusNumber SubBus,
 			printk("found device at bus %d idsel %d func %d (AgentId %x)\n",
 			       bus, IdSel, Function, AgentId);
 			/*  Connect EADs: 0x18.00.12 = 0x00 */
-			PPCDBG(PPCDBG_BUSWALK,
-					"PCI:Connect EADs: 0x%02X.%02X.%02X\n",
-					bus, SubBus, AgentId);
 			HvRc = HvCallPci_getBusUnitInfo(bus, SubBus, AgentId,
 					iseries_hv_addr(BridgeInfo),
 					sizeof(struct HvCallPci_BridgeInfo));
@@ -414,14 +391,6 @@ static void scan_EADS_bridge(HvBusNumber bus, HvSubBusNumber SubBus,
 					BridgeInfo->maxAgents,
 					BridgeInfo->maxSubBusNumber,
 					BridgeInfo->logicalSlotNumber);
-				PPCDBG(PPCDBG_BUSWALK,
-					"PCI: BridgeInfo, Type:0x%02X, SubBus:0x%02X, MaxAgents:0x%02X, MaxSubBus: 0x%02X, LSlot: 0x%02X\n",
-					BridgeInfo->busUnitInfo.deviceType,
-					BridgeInfo->subBusNumber,
-					BridgeInfo->maxAgents,
-					BridgeInfo->maxSubBusNumber,
-					BridgeInfo->logicalSlotNumber);
-
 				if (BridgeInfo->busUnitInfo.deviceType ==
 						HvCallPci_BridgeDevice)  {
 					/* Scan_Bridge_Slot...: 0x18.00.12 */
@@ -454,9 +423,6 @@ static int scan_bridge_slot(HvBusNumber Bus,
 
 	/* iSeries_allocate_IRQ.: 0x18.00.12(0xA3) */
 	Irq = iSeries_allocate_IRQ(Bus, 0, EADsIdSel);
-	PPCDBG(PPCDBG_BUSWALK,
-		"PCI:- allocate and assign IRQ 0x%02X.%02X.%02X = 0x%02X\n",
-		Bus, 0, EADsIdSel, Irq);
 
 	/*
 	 * Connect all functions of any device found.
@@ -482,9 +448,6 @@ static int scan_bridge_slot(HvBusNumber Bus,
 			printk("read vendor ID: %x\n", VendorId);
 
 			/* FoundDevice: 0x18.28.10 = 0x12AE */
-			PPCDBG(PPCDBG_BUSWALK,
-			       "PCI:- FoundDevice: 0x%02X.%02X.%02X = 0x%04X, irq %d\n",
-			       Bus, SubBus, AgentId, VendorId, Irq);
 			HvRc = HvCallPci_configStore8(Bus, SubBus, AgentId,
 						      PCI_INTERRUPT_LINE, Irq);
 			if (HvRc != 0)
diff --git a/arch/powerpc/platforms/iseries/setup.c b/arch/powerpc/platforms/iseries/setup.c
index fda712b4216..7f8f0cda6a7 100644
--- a/arch/powerpc/platforms/iseries/setup.c
+++ b/arch/powerpc/platforms/iseries/setup.c
@@ -71,8 +71,6 @@ extern void hvlog(char *fmt, ...);
 #endif
 
 /* Function Prototypes */
-extern void ppcdbg_initialize(void);
-
 static void build_iSeries_Memory_Map(void);
 static void iseries_shared_idle(void);
 static void iseries_dedicated_idle(void);
@@ -309,8 +307,6 @@ static void __init iSeries_init_early(void)
 
 	ppc64_firmware_features = FW_FEATURE_ISERIES;
 
-	ppcdbg_initialize();
-
 	ppc64_interrupt_controller = IC_ISERIES;
 
 #if defined(CONFIG_BLK_DEV_INITRD)
@@ -320,11 +316,11 @@ static void __init iSeries_init_early(void)
 	 */
 	if (naca.xRamDisk) {
 		initrd_start = (unsigned long)__va(naca.xRamDisk);
-		initrd_end = initrd_start + naca.xRamDiskSize * PAGE_SIZE;
+		initrd_end = initrd_start + naca.xRamDiskSize * HW_PAGE_SIZE;
 		initrd_below_start_ok = 1;	// ramdisk in kernel space
 		ROOT_DEV = Root_RAM0;
-		if (((rd_size * 1024) / PAGE_SIZE) < naca.xRamDiskSize)
-			rd_size = (naca.xRamDiskSize * PAGE_SIZE) / 1024;
+		if (((rd_size * 1024) / HW_PAGE_SIZE) < naca.xRamDiskSize)
+			rd_size = (naca.xRamDiskSize * HW_PAGE_SIZE) / 1024;
 	} else
 #endif /* CONFIG_BLK_DEV_INITRD */
 	{
@@ -470,13 +466,14 @@ static void __init build_iSeries_Memory_Map(void)
 	 */
 	hptFirstChunk = (u32)addr_to_chunk(HvCallHpt_getHptAddress());
 	hptSizePages = (u32)HvCallHpt_getHptPages();
-	hptSizeChunks = hptSizePages >> (MSCHUNKS_CHUNK_SHIFT - PAGE_SHIFT);
+	hptSizeChunks = hptSizePages >>
+		(MSCHUNKS_CHUNK_SHIFT - HW_PAGE_SHIFT);
 	hptLastChunk = hptFirstChunk + hptSizeChunks - 1;
 
 	printk("HPT absolute addr = %016lx, size = %dK\n",
 			chunk_to_addr(hptFirstChunk), hptSizeChunks * 256);
 
-	ppc64_pft_size = __ilog2(hptSizePages * PAGE_SIZE);
+	ppc64_pft_size = __ilog2(hptSizePages * HW_PAGE_SIZE);
 
 	/*
 	 * The actual hashed page table is in the hypervisor,
@@ -629,7 +626,7 @@ static void __init iSeries_fixup_klimit(void)
 	 */
 	if (naca.xRamDisk)
 		klimit = KERNELBASE + (u64)naca.xRamDisk +
-			(naca.xRamDiskSize * PAGE_SIZE);
+			(naca.xRamDiskSize * HW_PAGE_SIZE);
 	else {
 		/*
 		 * No ram disk was included - check and see if there
@@ -697,20 +694,19 @@ static void iseries_shared_idle(void)
 		if (hvlpevent_is_pending())
 			process_iSeries_events();
 
+		preempt_enable_no_resched();
 		schedule();
+		preempt_disable();
 	}
 }
 
 static void iseries_dedicated_idle(void)
 {
 	long oldval;
+	set_thread_flag(TIF_POLLING_NRFLAG);
 
 	while (1) {
-		oldval = test_and_clear_thread_flag(TIF_NEED_RESCHED);
-
-		if (!oldval) {
-			set_thread_flag(TIF_POLLING_NRFLAG);
-
+		if (!need_resched()) {
 			while (!need_resched()) {
 				ppc64_runlatch_off();
 				HMT_low();
@@ -723,13 +719,12 @@ static void iseries_dedicated_idle(void)
 			}
 
 			HMT_medium();
-			clear_thread_flag(TIF_POLLING_NRFLAG);
-		} else {
-			set_need_resched();
 		}
 
 		ppc64_runlatch_on();
+		preempt_enable_no_resched();
 		schedule();
+		preempt_disable();
 	}
 }
 
diff --git a/arch/powerpc/platforms/iseries/smp.c b/arch/powerpc/platforms/iseries/smp.c
index 3336bad6772..fcb094ec6ae 100644
--- a/arch/powerpc/platforms/iseries/smp.c
+++ b/arch/powerpc/platforms/iseries/smp.c
@@ -40,7 +40,6 @@
 #include <asm/paca.h>
 #include <asm/iseries/hv_call.h>
 #include <asm/time.h>
-#include <asm/ppcdebug.h>
 #include <asm/machdep.h>
 #include <asm/cputable.h>
 #include <asm/system.h>
diff --git a/arch/powerpc/platforms/iseries/vio.c b/arch/powerpc/platforms/iseries/vio.c
index c27a66876c2..384360ee06e 100644
--- a/arch/powerpc/platforms/iseries/vio.c
+++ b/arch/powerpc/platforms/iseries/vio.c
@@ -30,41 +30,14 @@ static struct iommu_table vio_iommu_table;
 
 static void __init iommu_vio_init(void)
 {
-	struct iommu_table *t;
-	struct iommu_table_cb cb;
-	unsigned long cbp;
-	unsigned long itc_entries;
+	iommu_table_getparms_iSeries(255, 0, 0xff, &veth_iommu_table);
+	veth_iommu_table.it_size /= 2;
+	vio_iommu_table = veth_iommu_table;
+	vio_iommu_table.it_offset += veth_iommu_table.it_size;
 
-	cb.itc_busno = 255;    /* Bus 255 is the virtual bus */
-	cb.itc_virtbus = 0xff; /* Ask for virtual bus */
-
-	cbp = virt_to_abs(&cb);
-	HvCallXm_getTceTableParms(cbp);
-
-	itc_entries = cb.itc_size * PAGE_SIZE / sizeof(union tce_entry);
-	veth_iommu_table.it_size        = itc_entries / 2;
-	veth_iommu_table.it_busno       = cb.itc_busno;
-	veth_iommu_table.it_offset      = cb.itc_offset;
-	veth_iommu_table.it_index       = cb.itc_index;
-	veth_iommu_table.it_type        = TCE_VB;
-	veth_iommu_table.it_blocksize	= 1;
-
-	t = iommu_init_table(&veth_iommu_table);
-
-	if (!t)
+	if (!iommu_init_table(&veth_iommu_table))
 		printk("Virtual Bus VETH TCE table failed.\n");
-
-	vio_iommu_table.it_size         = itc_entries - veth_iommu_table.it_size;
-	vio_iommu_table.it_busno        = cb.itc_busno;
-	vio_iommu_table.it_offset       = cb.itc_offset +
-					  veth_iommu_table.it_size;
-	vio_iommu_table.it_index        = cb.itc_index;
-	vio_iommu_table.it_type         = TCE_VB;
-	vio_iommu_table.it_blocksize	= 1;
-
-	t = iommu_init_table(&vio_iommu_table);
-
-	if (!t)
+	if (!iommu_init_table(&vio_iommu_table))
 		printk("Virtual Bus VIO TCE table failed.\n");
 }
 
diff --git a/arch/powerpc/platforms/iseries/viopath.c b/arch/powerpc/platforms/iseries/viopath.c
index fe97bfbf746..84267269559 100644
--- a/arch/powerpc/platforms/iseries/viopath.c
+++ b/arch/powerpc/platforms/iseries/viopath.c
@@ -68,7 +68,8 @@ static DEFINE_SPINLOCK(statuslock);
  * For each kind of event we allocate a buffer that is
  * guaranteed not to cross a page boundary
  */
-static unsigned char event_buffer[VIO_MAX_SUBTYPES * 256] __page_aligned;
+static unsigned char event_buffer[VIO_MAX_SUBTYPES * 256]
+	__attribute__((__aligned__(4096)));
 static atomic_t event_buffer_available[VIO_MAX_SUBTYPES];
 static int event_buffer_initialised;
 
@@ -116,12 +117,12 @@ static int proc_viopath_show(struct seq_file *m, void *v)
 	HvLpEvent_Rc hvrc;
 	DECLARE_MUTEX_LOCKED(Semaphore);
 
-	buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	buf = kmalloc(HW_PAGE_SIZE, GFP_KERNEL);
 	if (!buf)
 		return 0;
-	memset(buf, 0, PAGE_SIZE);
+	memset(buf, 0, HW_PAGE_SIZE);
 
-	handle = dma_map_single(iSeries_vio_dev, buf, PAGE_SIZE,
+	handle = dma_map_single(iSeries_vio_dev, buf, HW_PAGE_SIZE,
 				DMA_FROM_DEVICE);
 
 	hvrc = HvCallEvent_signalLpEventFast(viopath_hostLp,
@@ -131,7 +132,7 @@ static int proc_viopath_show(struct seq_file *m, void *v)
 			viopath_sourceinst(viopath_hostLp),
 			viopath_targetinst(viopath_hostLp),
 			(u64)(unsigned long)&Semaphore, VIOVERSION << 16,
-			((u64)handle) << 32, PAGE_SIZE, 0, 0);
+			((u64)handle) << 32, HW_PAGE_SIZE, 0, 0);
 
 	if (hvrc != HvLpEvent_Rc_Good)
 		printk(VIOPATH_KERN_WARN "hv error on op %d\n", (int)hvrc);
@@ -140,7 +141,7 @@ static int proc_viopath_show(struct seq_file *m, void *v)
 
 	vlanMap = HvLpConfig_getVirtualLanIndexMap();
 
-	buf[PAGE_SIZE-1] = '\0';
+	buf[HW_PAGE_SIZE-1] = '\0';
 	seq_printf(m, "%s", buf);
 	seq_printf(m, "AVAILABLE_VETH=%x\n", vlanMap);
 	seq_printf(m, "SRLNBR=%c%c%c%c%c%c%c\n",
@@ -152,7 +153,8 @@ static int proc_viopath_show(struct seq_file *m, void *v)
 		   e2a(xItExtVpdPanel.systemSerial[4]),
 		   e2a(xItExtVpdPanel.systemSerial[5]));
 
-	dma_unmap_single(iSeries_vio_dev, handle, PAGE_SIZE, DMA_FROM_DEVICE);
+	dma_unmap_single(iSeries_vio_dev, handle, HW_PAGE_SIZE,
+			 DMA_FROM_DEVICE);
 	kfree(buf);
 
 	return 0;
diff --git a/arch/powerpc/platforms/powermac/Makefile b/arch/powerpc/platforms/powermac/Makefile
index 4369676f1d5..c9df44fcf57 100644
--- a/arch/powerpc/platforms/powermac/Makefile
+++ b/arch/powerpc/platforms/powermac/Makefile
@@ -1,7 +1,8 @@
 obj-y				+= pic.o setup.o time.o feature.o pci.o \
 				   sleep.o low_i2c.o cache.o
 obj-$(CONFIG_PMAC_BACKLIGHT)	+= backlight.o
-obj-$(CONFIG_CPU_FREQ_PMAC)	+= cpufreq.o
+obj-$(CONFIG_CPU_FREQ_PMAC)	+= cpufreq_32.o
+obj-$(CONFIG_CPU_FREQ_PMAC64)	+= cpufreq_64.o
 obj-$(CONFIG_NVRAM)		+= nvram.o
 # ppc64 pmac doesn't define CONFIG_NVRAM but needs nvram stuff
 obj-$(CONFIG_PPC64)		+= nvram.o
diff --git a/arch/powerpc/platforms/powermac/cpufreq.c b/arch/powerpc/platforms/powermac/cpufreq_32.c
index c47f8b69725..56fd4e05fed 100644
--- a/arch/powerpc/platforms/powermac/cpufreq.c
+++ b/arch/powerpc/platforms/powermac/cpufreq_32.c
@@ -397,18 +397,16 @@ static int pmac_cpufreq_target(	struct cpufreq_policy *policy,
 					unsigned int relation)
 {
 	unsigned int    newstate = 0;
+	int		rc;
 
 	if (cpufreq_frequency_table_target(policy, pmac_cpu_freqs,
 			target_freq, relation, &newstate))
 		return -EINVAL;
 
-	return do_set_cpu_speed(newstate, 1);
-}
+	rc = do_set_cpu_speed(newstate, 1);
 
-unsigned int pmac_get_one_cpufreq(int i)
-{
-	/* Supports only one CPU for now */
-	return (i == 0) ? cur_freq : 0;
+	ppc_proc_freq = cur_freq * 1000ul;
+	return rc;
 }
 
 static int pmac_cpufreq_cpu_init(struct cpufreq_policy *policy)
@@ -474,6 +472,8 @@ static int pmac_cpufreq_resume(struct cpufreq_policy *policy)
 	do_set_cpu_speed(sleep_freq == low_freq ?
 			 CPUFREQ_LOW : CPUFREQ_HIGH, 0);
 
+	ppc_proc_freq = cur_freq * 1000ul;
+
 	no_schedule = 0;
 	return 0;
 }
@@ -547,7 +547,7 @@ static int pmac_cpufreq_init_MacRISC3(struct device_node *cpunode)
 		 */
 		if (low_freq < 98000000)
 			low_freq = 101000000;
-			
+
 		/* Convert those to CPU core clocks */
 		low_freq = (low_freq * (*ratio)) / 2000;
 		hi_freq = (hi_freq * (*ratio)) / 2000;
@@ -714,6 +714,7 @@ out:
 
 	pmac_cpu_freqs[CPUFREQ_LOW].frequency = low_freq;
 	pmac_cpu_freqs[CPUFREQ_HIGH].frequency = hi_freq;
+	ppc_proc_freq = cur_freq * 1000ul;
 
 	printk(KERN_INFO "Registering PowerMac CPU frequency driver\n");
 	printk(KERN_INFO "Low: %d Mhz, High: %d Mhz, Boot: %d Mhz\n",
diff --git a/arch/powerpc/platforms/powermac/cpufreq_64.c b/arch/powerpc/platforms/powermac/cpufreq_64.c
new file mode 100644
index 00000000000..39150342c6f
--- /dev/null
+++ b/arch/powerpc/platforms/powermac/cpufreq_64.c
@@ -0,0 +1,323 @@
+/*
+ *  Copyright (C) 2002 - 2005 Benjamin Herrenschmidt <benh@kernel.crashing.org>
+ *  and                       Markus Demleitner <msdemlei@cl.uni-heidelberg.de>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This driver adds basic cpufreq support for SMU & 970FX based G5 Macs,
+ * that is iMac G5 and latest single CPU desktop.
+ */
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/delay.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/cpufreq.h>
+#include <linux/init.h>
+#include <linux/completion.h>
+#include <asm/prom.h>
+#include <asm/machdep.h>
+#include <asm/irq.h>
+#include <asm/sections.h>
+#include <asm/cputable.h>
+#include <asm/time.h>
+#include <asm/smu.h>
+
+#undef DEBUG
+
+#ifdef DEBUG
+#define DBG(fmt...) printk(fmt)
+#else
+#define DBG(fmt...)
+#endif
+
+/* see 970FX user manual */
+
+#define SCOM_PCR 0x0aa001			/* PCR scom addr */
+
+#define PCR_HILO_SELECT		0x80000000U	/* 1 = PCR, 0 = PCRH */
+#define PCR_SPEED_FULL		0x00000000U	/* 1:1 speed value */
+#define PCR_SPEED_HALF		0x00020000U	/* 1:2 speed value */
+#define PCR_SPEED_QUARTER	0x00040000U	/* 1:4 speed value */
+#define PCR_SPEED_MASK		0x000e0000U	/* speed mask */
+#define PCR_SPEED_SHIFT		17
+#define PCR_FREQ_REQ_VALID	0x00010000U	/* freq request valid */
+#define PCR_VOLT_REQ_VALID	0x00008000U	/* volt request valid */
+#define PCR_TARGET_TIME_MASK	0x00006000U	/* target time */
+#define PCR_STATLAT_MASK	0x00001f00U	/* STATLAT value */
+#define PCR_SNOOPLAT_MASK	0x000000f0U	/* SNOOPLAT value */
+#define PCR_SNOOPACC_MASK	0x0000000fU	/* SNOOPACC value */
+
+#define SCOM_PSR 0x408001			/* PSR scom addr */
+/* warning: PSR is a 64 bits register */
+#define PSR_CMD_RECEIVED	0x2000000000000000U   /* command received */
+#define PSR_CMD_COMPLETED	0x1000000000000000U   /* command completed */
+#define PSR_CUR_SPEED_MASK	0x0300000000000000U   /* current speed */
+#define PSR_CUR_SPEED_SHIFT	(56)
+
+/*
+ * The G5 only supports two frequencies (Quarter speed is not supported)
+ */
+#define CPUFREQ_HIGH                  0
+#define CPUFREQ_LOW                   1
+
+static struct cpufreq_frequency_table g5_cpu_freqs[] = {
+	{CPUFREQ_HIGH, 		0},
+	{CPUFREQ_LOW,		0},
+	{0,			CPUFREQ_TABLE_END},
+};
+
+static struct freq_attr* g5_cpu_freqs_attr[] = {
+	&cpufreq_freq_attr_scaling_available_freqs,
+	NULL,
+};
+
+/* Power mode data is an array of the 32 bits PCR values to use for
+ * the various frequencies, retreived from the device-tree
+ */
+static u32 *g5_pmode_data;
+static int g5_pmode_max;
+static int g5_pmode_cur;
+
+static DECLARE_MUTEX(g5_switch_mutex);
+
+
+static struct smu_sdbp_fvt *g5_fvt_table;	/* table of op. points */
+static int g5_fvt_count;			/* number of op. points */
+static int g5_fvt_cur;				/* current op. point */
+
+/* ----------------- real hardware interface */
+
+static void g5_switch_volt(int speed_mode)
+{
+	struct smu_simple_cmd	cmd;
+
+	DECLARE_COMPLETION(comp);
+	smu_queue_simple(&cmd, SMU_CMD_POWER_COMMAND, 8, smu_done_complete,
+			 &comp, 'V', 'S', 'L', 'E', 'W',
+			 0xff, g5_fvt_cur+1, speed_mode);
+	wait_for_completion(&comp);
+}
+
+static int g5_switch_freq(int speed_mode)
+{
+	struct cpufreq_freqs freqs;
+	int to;
+
+	if (g5_pmode_cur == speed_mode)
+		return 0;
+
+	down(&g5_switch_mutex);
+
+	freqs.old = g5_cpu_freqs[g5_pmode_cur].frequency;
+	freqs.new = g5_cpu_freqs[speed_mode].frequency;
+	freqs.cpu = 0;
+
+	cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
+
+	/* If frequency is going up, first ramp up the voltage */
+	if (speed_mode < g5_pmode_cur)
+		g5_switch_volt(speed_mode);
+
+	/* Clear PCR high */
+	scom970_write(SCOM_PCR, 0);
+	/* Clear PCR low */
+       	scom970_write(SCOM_PCR, PCR_HILO_SELECT | 0);
+	/* Set PCR low */
+	scom970_write(SCOM_PCR, PCR_HILO_SELECT |
+		      g5_pmode_data[speed_mode]);
+
+	/* Wait for completion */
+	for (to = 0; to < 10; to++) {
+		unsigned long psr = scom970_read(SCOM_PSR);
+
+		if ((psr & PSR_CMD_RECEIVED) == 0 &&
+		    (((psr >> PSR_CUR_SPEED_SHIFT) ^
+		      (g5_pmode_data[speed_mode] >> PCR_SPEED_SHIFT)) & 0x3)
+		    == 0)
+			break;
+		if (psr & PSR_CMD_COMPLETED)
+			break;
+		udelay(100);
+	}
+
+	/* If frequency is going down, last ramp the voltage */
+	if (speed_mode > g5_pmode_cur)
+		g5_switch_volt(speed_mode);
+
+	g5_pmode_cur = speed_mode;
+	ppc_proc_freq = g5_cpu_freqs[speed_mode].frequency * 1000ul;
+
+	cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
+
+	up(&g5_switch_mutex);
+
+	return 0;
+}
+
+static int g5_query_freq(void)
+{
+	unsigned long psr = scom970_read(SCOM_PSR);
+	int i;
+
+	for (i = 0; i <= g5_pmode_max; i++)
+		if ((((psr >> PSR_CUR_SPEED_SHIFT) ^
+		      (g5_pmode_data[i] >> PCR_SPEED_SHIFT)) & 0x3) == 0)
+			break;
+	return i;
+}
+
+/* ----------------- cpufreq bookkeeping */
+
+static int g5_cpufreq_verify(struct cpufreq_policy *policy)
+{
+	return cpufreq_frequency_table_verify(policy, g5_cpu_freqs);
+}
+
+static int g5_cpufreq_target(struct cpufreq_policy *policy,
+	unsigned int target_freq, unsigned int relation)
+{
+	unsigned int    newstate = 0;
+
+	if (cpufreq_frequency_table_target(policy, g5_cpu_freqs,
+			target_freq, relation, &newstate))
+		return -EINVAL;
+
+	return g5_switch_freq(newstate);
+}
+
+static unsigned int g5_cpufreq_get_speed(unsigned int cpu)
+{
+	return g5_cpu_freqs[g5_pmode_cur].frequency;
+}
+
+static int g5_cpufreq_cpu_init(struct cpufreq_policy *policy)
+{
+	if (policy->cpu != 0)
+		return -ENODEV;
+
+	policy->governor = CPUFREQ_DEFAULT_GOVERNOR;
+	policy->cpuinfo.transition_latency = CPUFREQ_ETERNAL;
+	policy->cur = g5_cpu_freqs[g5_query_freq()].frequency;
+	cpufreq_frequency_table_get_attr(g5_cpu_freqs, policy->cpu);
+
+	return cpufreq_frequency_table_cpuinfo(policy,
+		g5_cpu_freqs);
+}
+
+
+static struct cpufreq_driver g5_cpufreq_driver = {
+	.name		= "powermac",
+	.owner		= THIS_MODULE,
+	.flags		= CPUFREQ_CONST_LOOPS,
+	.init		= g5_cpufreq_cpu_init,
+	.verify		= g5_cpufreq_verify,
+	.target		= g5_cpufreq_target,
+	.get		= g5_cpufreq_get_speed,
+	.attr 		= g5_cpu_freqs_attr,
+};
+
+
+static int __init g5_cpufreq_init(void)
+{
+	struct device_node *cpunode;
+	unsigned int psize, ssize;
+	struct smu_sdbp_header *shdr;
+	unsigned long max_freq;
+	u32 *valp;
+	int rc = -ENODEV;
+
+	/* Look for CPU and SMU nodes */
+	cpunode = of_find_node_by_type(NULL, "cpu");
+	if (!cpunode) {
+		DBG("No CPU node !\n");
+		return -ENODEV;
+	}
+
+	/* Check 970FX for now */
+	valp = (u32 *)get_property(cpunode, "cpu-version", NULL);
+	if (!valp) {
+		DBG("No cpu-version property !\n");
+		goto bail_noprops;
+	}
+	if (((*valp) >> 16) != 0x3c) {
+		DBG("Wrong CPU version: %08x\n", *valp);
+		goto bail_noprops;
+	}
+
+	/* Look for the powertune data in the device-tree */
+	g5_pmode_data = (u32 *)get_property(cpunode, "power-mode-data",&psize);
+	if (!g5_pmode_data) {
+		DBG("No power-mode-data !\n");
+		goto bail_noprops;
+	}
+	g5_pmode_max = psize / sizeof(u32) - 1;
+
+	/* Look for the FVT table */
+	shdr = smu_get_sdb_partition(SMU_SDB_FVT_ID, NULL);
+	if (!shdr)
+		goto bail_noprops;
+	g5_fvt_table = (struct smu_sdbp_fvt *)&shdr[1];
+	ssize = (shdr->len * sizeof(u32)) - sizeof(struct smu_sdbp_header);
+	g5_fvt_count = ssize / sizeof(struct smu_sdbp_fvt);
+	g5_fvt_cur = 0;
+
+	/* Sanity checking */
+	if (g5_fvt_count < 1 || g5_pmode_max < 1)
+		goto bail_noprops;
+
+	/*
+	 * From what I see, clock-frequency is always the maximal frequency.
+	 * The current driver can not slew sysclk yet, so we really only deal
+	 * with powertune steps for now. We also only implement full freq and
+	 * half freq in this version. So far, I haven't yet seen a machine
+	 * supporting anything else.
+	 */
+	valp = (u32 *)get_property(cpunode, "clock-frequency", NULL);
+	if (!valp)
+		return -ENODEV;
+	max_freq = (*valp)/1000;
+	g5_cpu_freqs[0].frequency = max_freq;
+	g5_cpu_freqs[1].frequency = max_freq/2;
+
+	/* Check current frequency */
+	g5_pmode_cur = g5_query_freq();
+	if (g5_pmode_cur > 1)
+		/* We don't support anything but 1:1 and 1:2, fixup ... */
+		g5_pmode_cur = 1;
+
+	/* Force apply current frequency to make sure everything is in
+	 * sync (voltage is right for example). Firmware may leave us with
+	 * a strange setting ...
+	 */
+	g5_switch_freq(g5_pmode_cur);
+
+	printk(KERN_INFO "Registering G5 CPU frequency driver\n");
+	printk(KERN_INFO "Low: %d Mhz, High: %d Mhz, Cur: %d MHz\n",
+		g5_cpu_freqs[1].frequency/1000,
+		g5_cpu_freqs[0].frequency/1000,
+		g5_cpu_freqs[g5_pmode_cur].frequency/1000);
+
+	rc = cpufreq_register_driver(&g5_cpufreq_driver);
+
+	/* We keep the CPU node on hold... hopefully, Apple G5 don't have
+	 * hotplug CPU with a dynamic device-tree ...
+	 */
+	return rc;
+
+ bail_noprops:
+	of_node_put(cpunode);
+
+	return rc;
+}
+
+module_init(g5_cpufreq_init);
+
+
+MODULE_LICENSE("GPL");
diff --git a/arch/powerpc/platforms/powermac/setup.c b/arch/powerpc/platforms/powermac/setup.c
index 80b58c1ec41..7acb0546671 100644
--- a/arch/powerpc/platforms/powermac/setup.c
+++ b/arch/powerpc/platforms/powermac/setup.c
@@ -193,18 +193,6 @@ static void pmac_show_cpuinfo(struct seq_file *m)
 		   pmac_newworld ? "NewWorld" : "OldWorld");
 }
 
-static void pmac_show_percpuinfo(struct seq_file *m, int i)
-{
-#ifdef CONFIG_CPU_FREQ_PMAC
-	extern unsigned int pmac_get_one_cpufreq(int i);
-	unsigned int freq = pmac_get_one_cpufreq(i);
-	if (freq != 0) {
-		seq_printf(m, "clock\t\t: %dMHz\n", freq/1000);
-		return;
-	}
-#endif /* CONFIG_CPU_FREQ_PMAC */
-}
-
 #ifndef CONFIG_ADB_CUDA
 int find_via_cuda(void)
 {
@@ -767,7 +755,6 @@ struct machdep_calls __initdata pmac_md = {
 	.setup_arch		= pmac_setup_arch,
 	.init_early		= pmac_init_early,
 	.show_cpuinfo		= pmac_show_cpuinfo,
-	.show_percpuinfo	= pmac_show_percpuinfo,
 	.init_IRQ		= pmac_pic_init,
 	.get_irq		= mpic_get_irq,	/* changed later */
 	.pcibios_fixup		= pmac_pcibios_fixup,
diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c
index 513e2723149..fcc50bfd43f 100644
--- a/arch/powerpc/platforms/pseries/iommu.c
+++ b/arch/powerpc/platforms/pseries/iommu.c
@@ -37,7 +37,6 @@
 #include <asm/io.h>
 #include <asm/prom.h>
 #include <asm/rtas.h>
-#include <asm/ppcdebug.h>
 #include <asm/iommu.h>
 #include <asm/pci-bridge.h>
 #include <asm/machdep.h>
@@ -47,6 +46,7 @@
 #include <asm/firmware.h>
 #include <asm/tce.h>
 #include <asm/ppc-pci.h>
+#include <asm/udbg.h>
 
 #include "plpar_wrappers.h"
 
diff --git a/arch/powerpc/platforms/pseries/lpar.c b/arch/powerpc/platforms/pseries/lpar.c
index e384a5a9179..a50e5f3f396 100644
--- a/arch/powerpc/platforms/pseries/lpar.c
+++ b/arch/powerpc/platforms/pseries/lpar.c
@@ -19,7 +19,7 @@
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
  */
 
-#define DEBUG
+#undef DEBUG_LOW
 
 #include <linux/config.h>
 #include <linux/kernel.h>
@@ -31,20 +31,21 @@
 #include <asm/machdep.h>
 #include <asm/abs_addr.h>
 #include <asm/mmu_context.h>
-#include <asm/ppcdebug.h>
 #include <asm/iommu.h>
 #include <asm/tlbflush.h>
 #include <asm/tlb.h>
 #include <asm/prom.h>
 #include <asm/abs_addr.h>
 #include <asm/cputable.h>
+#include <asm/udbg.h>
+#include <asm/smp.h>
 
 #include "plpar_wrappers.h"
 
-#ifdef DEBUG
-#define DBG(fmt...) udbg_printf(fmt)
+#ifdef DEBUG_LOW
+#define DBG_LOW(fmt...) do { udbg_printf(fmt); } while(0)
 #else
-#define DBG(fmt...)
+#define DBG_LOW(fmt...) do { } while(0)
 #endif
 
 /* in pSeries_hvCall.S */
@@ -276,8 +277,9 @@ void vpa_init(int cpu)
 }
 
 long pSeries_lpar_hpte_insert(unsigned long hpte_group,
-			      unsigned long va, unsigned long prpn,
-			      unsigned long vflags, unsigned long rflags)
+ 			      unsigned long va, unsigned long pa,
+ 			      unsigned long rflags, unsigned long vflags,
+ 			      int psize)
 {
 	unsigned long lpar_rc;
 	unsigned long flags;
@@ -285,11 +287,28 @@ long pSeries_lpar_hpte_insert(unsigned long hpte_group,
 	unsigned long hpte_v, hpte_r;
 	unsigned long dummy0, dummy1;
 
-	hpte_v = ((va >> 23) << HPTE_V_AVPN_SHIFT) | vflags | HPTE_V_VALID;
-	if (vflags & HPTE_V_LARGE)
-		hpte_v &= ~(1UL << HPTE_V_AVPN_SHIFT);
-
-	hpte_r = (prpn << HPTE_R_RPN_SHIFT) | rflags;
+	if (!(vflags & HPTE_V_BOLTED))
+		DBG_LOW("hpte_insert(group=%lx, va=%016lx, pa=%016lx, "
+			"rflags=%lx, vflags=%lx, psize=%d)\n",
+		hpte_group, va, pa, rflags, vflags, psize);
+
+ 	hpte_v = hpte_encode_v(va, psize) | vflags | HPTE_V_VALID;
+	hpte_r = hpte_encode_r(pa, psize) | rflags;
+
+	if (!(vflags & HPTE_V_BOLTED))
+		DBG_LOW(" hpte_v=%016lx, hpte_r=%016lx\n", hpte_v, hpte_r);
+
+#if 1
+	{
+		int i;
+		for (i=0;i<8;i++) {
+			unsigned long w0, w1;
+			plpar_pte_read(0, hpte_group, &w0, &w1);
+			BUG_ON (HPTE_V_COMPARE(hpte_v, w0)
+				&& (w0 & HPTE_V_VALID));
+		}
+	}
+#endif
 
 	/* Now fill in the actual HPTE */
 	/* Set CEC cookie to 0         */
@@ -299,23 +318,30 @@ long pSeries_lpar_hpte_insert(unsigned long hpte_group,
 	/* Exact = 0                   */
 	flags = 0;
 
-	/* XXX why is this here? - Anton */
+	/* Make pHyp happy */
 	if (rflags & (_PAGE_GUARDED|_PAGE_NO_CACHE))
 		hpte_r &= ~_PAGE_COHERENT;
 
 	lpar_rc = plpar_hcall(H_ENTER, flags, hpte_group, hpte_v,
 			      hpte_r, &slot, &dummy0, &dummy1);
-
-	if (unlikely(lpar_rc == H_PTEG_Full))
+	if (unlikely(lpar_rc == H_PTEG_Full)) {
+		if (!(vflags & HPTE_V_BOLTED))
+			DBG_LOW(" full\n");
 		return -1;
+	}
 
 	/*
 	 * Since we try and ioremap PHBs we don't own, the pte insert
 	 * will fail. However we must catch the failure in hash_page
 	 * or we will loop forever, so return -2 in this case.
 	 */
-	if (unlikely(lpar_rc != H_Success))
+	if (unlikely(lpar_rc != H_Success)) {
+		if (!(vflags & HPTE_V_BOLTED))
+			DBG_LOW(" lpar err %d\n", lpar_rc);
 		return -2;
+	}
+	if (!(vflags & HPTE_V_BOLTED))
+		DBG_LOW(" -> slot: %d\n", slot & 7);
 
 	/* Because of iSeries, we have to pass down the secondary
 	 * bucket bit here as well
@@ -340,10 +366,8 @@ static long pSeries_lpar_hpte_remove(unsigned long hpte_group)
 		/* don't remove a bolted entry */
 		lpar_rc = plpar_pte_remove(H_ANDCOND, hpte_group + slot_offset,
 					   (0x1UL << 4), &dummy1, &dummy2);
-
 		if (lpar_rc == H_Success)
 			return i;
-
 		BUG_ON(lpar_rc != H_Not_Found);
 
 		slot_offset++;
@@ -371,20 +395,28 @@ static void pSeries_lpar_hptab_clear(void)
  * We can probably optimize here and assume the high bits of newpp are
  * already zero.  For now I am paranoid.
  */
-static long pSeries_lpar_hpte_updatepp(unsigned long slot, unsigned long newpp,
-				       unsigned long va, int large, int local)
+static long pSeries_lpar_hpte_updatepp(unsigned long slot,
+				       unsigned long newpp,
+				       unsigned long va,
+				       int psize, int local)
 {
 	unsigned long lpar_rc;
 	unsigned long flags = (newpp & 7) | H_AVPN;
-	unsigned long avpn = va >> 23;
+	unsigned long want_v;
 
-	if (large)
-		avpn &= ~0x1UL;
+	want_v = hpte_encode_v(va, psize);
 
-	lpar_rc = plpar_pte_protect(flags, slot, (avpn << 7));
+	DBG_LOW("    update: avpnv=%016lx, hash=%016lx, f=%x, psize: %d ... ",
+		want_v & HPTE_V_AVPN, slot, flags, psize);
 
-	if (lpar_rc == H_Not_Found)
+	lpar_rc = plpar_pte_protect(flags, slot, want_v & HPTE_V_AVPN);
+
+	if (lpar_rc == H_Not_Found) {
+		DBG_LOW("not found !\n");
 		return -1;
+	}
+
+	DBG_LOW("ok\n");
 
 	BUG_ON(lpar_rc != H_Success);
 
@@ -410,21 +442,22 @@ static unsigned long pSeries_lpar_hpte_getword0(unsigned long slot)
 	return dword0;
 }
 
-static long pSeries_lpar_hpte_find(unsigned long vpn)
+static long pSeries_lpar_hpte_find(unsigned long va, int psize)
 {
 	unsigned long hash;
 	unsigned long i, j;
 	long slot;
-	unsigned long hpte_v;
+	unsigned long want_v, hpte_v;
 
-	hash = hpt_hash(vpn, 0);
+	hash = hpt_hash(va, mmu_psize_defs[psize].shift);
+	want_v = hpte_encode_v(va, psize);
 
 	for (j = 0; j < 2; j++) {
 		slot = (hash & htab_hash_mask) * HPTES_PER_GROUP;
 		for (i = 0; i < HPTES_PER_GROUP; i++) {
 			hpte_v = pSeries_lpar_hpte_getword0(slot);
 
-			if ((HPTE_V_AVPN_VAL(hpte_v) == (vpn >> 11))
+			if (HPTE_V_COMPARE(hpte_v, want_v)
 			    && (hpte_v & HPTE_V_VALID)
 			    && (!!(hpte_v & HPTE_V_SECONDARY) == j)) {
 				/* HPTE matches */
@@ -441,17 +474,15 @@ static long pSeries_lpar_hpte_find(unsigned long vpn)
 } 
 
 static void pSeries_lpar_hpte_updateboltedpp(unsigned long newpp,
-					     unsigned long ea)
+					     unsigned long ea,
+					     int psize)
 {
-	unsigned long lpar_rc;
-	unsigned long vsid, va, vpn, flags;
-	long slot;
+	unsigned long lpar_rc, slot, vsid, va, flags;
 
 	vsid = get_kernel_vsid(ea);
 	va = (vsid << 28) | (ea & 0x0fffffff);
-	vpn = va >> PAGE_SHIFT;
 
-	slot = pSeries_lpar_hpte_find(vpn);
+	slot = pSeries_lpar_hpte_find(va, psize);
 	BUG_ON(slot == -1);
 
 	flags = newpp & 7;
@@ -461,18 +492,18 @@ static void pSeries_lpar_hpte_updateboltedpp(unsigned long newpp,
 }
 
 static void pSeries_lpar_hpte_invalidate(unsigned long slot, unsigned long va,
-					 int large, int local)
+					 int psize, int local)
 {
-	unsigned long avpn = va >> 23;
+	unsigned long want_v;
 	unsigned long lpar_rc;
 	unsigned long dummy1, dummy2;
 
-	if (large)
-		avpn &= ~0x1UL;
-
-	lpar_rc = plpar_pte_remove(H_AVPN, slot, (avpn << 7), &dummy1,
-				   &dummy2);
+	DBG_LOW("    inval : slot=%lx, va=%016lx, psize: %d, local: %d",
+		slot, va, psize, local);
 
+	want_v = hpte_encode_v(va, psize);
+	lpar_rc = plpar_pte_remove(H_AVPN, slot, want_v & HPTE_V_AVPN,
+				   &dummy1, &dummy2);
 	if (lpar_rc == H_Not_Found)
 		return;
 
@@ -494,7 +525,8 @@ void pSeries_lpar_flush_hash_range(unsigned long number, int local)
 		spin_lock_irqsave(&pSeries_lpar_tlbie_lock, flags);
 
 	for (i = 0; i < number; i++)
-		flush_hash_page(batch->vaddr[i], batch->pte[i], local);
+		flush_hash_page(batch->vaddr[i], batch->pte[i],
+				batch->psize, local);
 
 	if (lock_tlbie)
 		spin_unlock_irqrestore(&pSeries_lpar_tlbie_lock, flags);
diff --git a/arch/powerpc/platforms/pseries/plpar_wrappers.h b/arch/powerpc/platforms/pseries/plpar_wrappers.h
index 382f8c5b0e7..3bd1b3e0600 100644
--- a/arch/powerpc/platforms/pseries/plpar_wrappers.h
+++ b/arch/powerpc/platforms/pseries/plpar_wrappers.h
@@ -107,14 +107,4 @@ static inline long plpar_put_term_char(unsigned long termno, unsigned long len,
 			lbuf[1]);
 }
 
-static inline long plpar_set_xdabr(unsigned long address, unsigned long flags)
-{
-	return plpar_hcall_norets(H_SET_XDABR, address, flags);
-}
-
-static inline long plpar_set_dabr(unsigned long val)
-{
-	return plpar_hcall_norets(H_SET_DABR, val);
-}
-
 #endif /* _PSERIES_PLPAR_WRAPPERS_H */
diff --git a/arch/powerpc/platforms/pseries/ras.c b/arch/powerpc/platforms/pseries/ras.c
index 6562ff4b0a8..fbd214d68b0 100644
--- a/arch/powerpc/platforms/pseries/ras.c
+++ b/arch/powerpc/platforms/pseries/ras.c
@@ -48,7 +48,7 @@
 #include <asm/ptrace.h>
 #include <asm/machdep.h>
 #include <asm/rtas.h>
-#include <asm/ppcdebug.h>
+#include <asm/udbg.h>
 
 static unsigned char ras_log_buf[RTAS_ERROR_LOG_MAX];
 static DEFINE_SPINLOCK(ras_log_buf_lock);
diff --git a/arch/powerpc/platforms/pseries/reconfig.c b/arch/powerpc/platforms/pseries/reconfig.c
index 58c61219d08..d7d40033945 100644
--- a/arch/powerpc/platforms/pseries/reconfig.c
+++ b/arch/powerpc/platforms/pseries/reconfig.c
@@ -286,10 +286,8 @@ static struct property *new_property(const char *name, const int length,
 	return new;
 
 cleanup:
-	if (new->name)
-		kfree(new->name);
-	if (new->value)
-		kfree(new->value);
+	kfree(new->name);
+	kfree(new->value);
 	kfree(new);
 	return NULL;
 }
diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c
index 65bee939eec..a093a0d4dd6 100644
--- a/arch/powerpc/platforms/pseries/setup.c
+++ b/arch/powerpc/platforms/pseries/setup.c
@@ -65,6 +65,7 @@
 #include <asm/ppc-pci.h>
 #include <asm/i8259.h>
 #include <asm/udbg.h>
+#include <asm/smp.h>
 
 #include "plpar_wrappers.h"
 
@@ -353,14 +354,15 @@ static void pSeries_mach_cpu_die(void)
 
 static int pseries_set_dabr(unsigned long dabr)
 {
-	if (firmware_has_feature(FW_FEATURE_XDABR)) {
-		/* We want to catch accesses from kernel and userspace */
-		return plpar_set_xdabr(dabr, H_DABRX_KERNEL | H_DABRX_USER);
-	}
-
-	return plpar_set_dabr(dabr);
+	return plpar_hcall_norets(H_SET_DABR, dabr);
 }
 
+static int pseries_set_xdabr(unsigned long dabr)
+{
+	/* We want to catch accesses from kernel and userspace */
+	return plpar_hcall_norets(H_SET_XDABR, dabr,
+			H_DABRX_KERNEL | H_DABRX_USER);
+}
 
 /*
  * Early initialization.  Relocation is on but do not reference unbolted pages
@@ -396,8 +398,10 @@ static void __init pSeries_init_early(void)
 		DBG("Hello World !\n");
 	}
 
-	if (firmware_has_feature(FW_FEATURE_XDABR | FW_FEATURE_DABR))
+	if (firmware_has_feature(FW_FEATURE_DABR))
 		ppc_md.set_dabr = pseries_set_dabr;
+	else if (firmware_has_feature(FW_FEATURE_XDABR))
+		ppc_md.set_dabr = pseries_set_xdabr;
 
 	iommu_init_early_pSeries();
 
@@ -465,6 +469,7 @@ static inline void dedicated_idle_sleep(unsigned int cpu)
 		 * more.
 		 */
 		clear_thread_flag(TIF_POLLING_NRFLAG);
+		smp_mb__after_clear_bit();
 
 		/*
 		 * SMT dynamic mode. Cede will result in this thread going
@@ -477,6 +482,7 @@ static inline void dedicated_idle_sleep(unsigned int cpu)
 			cede_processor();
 		else
 			local_irq_enable();
+		set_thread_flag(TIF_POLLING_NRFLAG);
 	} else {
 		/*
 		 * Give the HV an opportunity at the processor, since we are
@@ -488,11 +494,11 @@ static inline void dedicated_idle_sleep(unsigned int cpu)
 
 static void pseries_dedicated_idle(void)
 { 
-	long oldval;
 	struct paca_struct *lpaca = get_paca();
 	unsigned int cpu = smp_processor_id();
 	unsigned long start_snooze;
 	unsigned long *smt_snooze_delay = &__get_cpu_var(smt_snooze_delay);
+	set_thread_flag(TIF_POLLING_NRFLAG);
 
 	while (1) {
 		/*
@@ -501,10 +507,7 @@ static void pseries_dedicated_idle(void)
 		 */
 		lpaca->lppaca.idle = 1;
 
-		oldval = test_and_clear_thread_flag(TIF_NEED_RESCHED);
-		if (!oldval) {
-			set_thread_flag(TIF_POLLING_NRFLAG);
-
+		if (!need_resched()) {
 			start_snooze = __get_tb() +
 				*smt_snooze_delay * tb_ticks_per_usec;
 
@@ -527,15 +530,14 @@ static void pseries_dedicated_idle(void)
 			}
 
 			HMT_medium();
-			clear_thread_flag(TIF_POLLING_NRFLAG);
-		} else {
-			set_need_resched();
 		}
 
 		lpaca->lppaca.idle = 0;
 		ppc64_runlatch_on();
 
+		preempt_enable_no_resched();
 		schedule();
+		preempt_disable();
 
 		if (cpu_is_offline(cpu) && system_state == SYSTEM_RUNNING)
 			cpu_die();
@@ -579,7 +581,9 @@ static void pseries_shared_idle(void)
 		lpaca->lppaca.idle = 0;
 		ppc64_runlatch_on();
 
+		preempt_enable_no_resched();
 		schedule();
+		preempt_disable();
 
 		if (cpu_is_offline(cpu) && system_state == SYSTEM_RUNNING)
 			cpu_die();