125 files changed, 2740 insertions, 969 deletions
diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index 770f717bd250..79c6e5a24456 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -83,22 +83,20 @@ choice
          check out the Linux/Alpha FAQ, accessible on the WWW from
          <http://www.alphalinux.org/>. In summary:
-          Alcor/Alpha-XLT     AS 600
+          Alcor/Alpha-XLT     AS 600, AS 500, XL-300, XL-366
          Alpha-XL            XL-233, XL-266
          AlphaBook1          Alpha laptop
          Avanti              AS 200, AS 205, AS 250, AS 255, AS 300, AS 400
          Cabriolet           AlphaPC64, AlphaPCI64
-          DP264               DP264
+          DP264               DP264 / DS20 / ES40 / DS10 / DS10L
          EB164               EB164 21164 evaluation board
          EB64+               EB64+ 21064 evaluation board
          EB66                EB66 21066 evaluation board
          EB66+               EB66+ 21066 evaluation board
-          Jensen              DECpc 150, DEC 2000 model 300,
+          Jensen              DECpc 150, DEC 2000 models 300, 500
-          DEC 2000 model 500
          LX164               AlphaPC164-LX
          Lynx                AS 2100A
-          Miata               Personal Workstation 433a, 433au, 500a,
+          Miata               Personal Workstation 433/500/600 a/au
-          500au, 600a, or 600au
          Marvel              AlphaServer ES47 / ES80 / GS1280
          Mikasa              AS 1000
          Noname              AXPpci33, UDB (Multia)
@@ -108,9 +106,9 @@ choice
          Ruffian             RPX164-2, AlphaPC164-UX, AlphaPC164-BX
          SX164               AlphaPC164-SX
          Sable               AS 2000, AS 2100
-          Shark               DS 20L
+          Shark               DS 20L
-          Takara              Takara
+          Takara              Takara (OEM)
-          Titan               AlphaServer ES45 / DS25
+          Titan               AlphaServer ES45 / DS25 / DS15
          Wildfire            AlphaServer GS 40/80/160/320
          If you don't know what to do, choose "generic".
@@ -481,6 +479,15 @@ config ALPHA_BROKEN_IRQ_MASK
        depends on ALPHA_GENERIC || ALPHA_PC164
        default y
+config VGA_HOSE
+        bool
+        depends on ALPHA_GENERIC || ALPHA_TITAN || ALPHA_MARVEL || ALPHA_TSUNAMI
+        default y
+        help
+          Support VGA on an arbitrary hose; needed for several platforms
+          which always have multiple hoses, and whose consoles support it.
 config ALPHA_SRM
        bool "Use SRM as bootloader" if ALPHA_CABRIOLET || ALPHA_AVANTI_CH || ALPHA_EB64P || ALPHA_PC164 || ALPHA_TAKARA || ALPHA_EB164 || ALPHA_ALCOR || ALPHA_MIATA || ALPHA_LX164 || ALPHA_SX164 || ALPHA_NAUTILUS || ALPHA_NONAME
        default y if ALPHA_JENSEN || ALPHA_MIKASA || ALPHA_SABLE || ALPHA_LYNX || ALPHA_NORITAKE || ALPHA_DP264 || ALPHA_RAWHIDE || ALPHA_EIGER || ALPHA_WILDFIRE || ALPHA_TITAN || ALPHA_SHARK || ALPHA_MARVEL
@@ -537,10 +544,14 @@ config HAVE_DEC_LOCK
        default y
 config NR_CPUS
-        int "Maximum number of CPUs (2-64)"
+        int "Maximum number of CPUs (2-32)"
-        range 2 64
+        range 2 32
        depends on SMP
-        default "64"
+        default "32" if ALPHA_GENERIC || ALPHA_MARVEL
+        default "4" if !ALPHA_GENERIC && !ALPHA_MARVEL
+        help
+          MARVEL support can handle a maximum of 32 CPUs, all the others
+          with working support have a maximum of 4 CPUs.
 config ARCH_DISCONTIGMEM_ENABLE
        bool "Discontiguous Memory Support (EXPERIMENTAL)"
@@ -644,6 +655,13 @@ source "arch/alpha/oprofile/Kconfig"
 source "arch/alpha/Kconfig.debug"
+# DUMMY_CONSOLE may be defined in drivers/video/console/Kconfig
+# but we also need it if VGA_HOSE is set
+config DUMMY_CONSOLE
+        bool
+        depends on VGA_HOSE
+        default y
 source "security/Kconfig"
 source "crypto/Kconfig"
diff --git a/arch/alpha/boot/tools/mkbb.c b/arch/alpha/boot/tools/mkbb.c
index 23c7190b047c..632a7fd6d7dc 100644
--- a/arch/alpha/boot/tools/mkbb.c
+++ b/arch/alpha/boot/tools/mkbb.c
@@ -81,7 +81,7 @@ typedef union __bootblock {
 #define bootblock_label         __u1.__label
 #define bootblock_checksum      __u2.__checksum
-main(int argc, char ** argv)
+int main(int argc, char ** argv)
 {
    bootblock           bootblock_from_disk;
    bootblock           bootloader_image;
diff --git a/arch/alpha/kernel/console.c b/arch/alpha/kernel/console.c
index f313b34939bb..da711e37fc97 100644
--- a/arch/alpha/kernel/console.c
+++ b/arch/alpha/kernel/console.c
@@ -9,16 +9,20 @@
 #include <linux/init.h>
 #include <linux/tty.h>
 #include <linux/console.h>
+#include <linux/vt.h>
 #include <asm/vga.h>
 #include <asm/machvec.h>
+#include "pci_impl.h"
 #ifdef CONFIG_VGA_HOSE
-/*
+struct pci_controller *pci_vga_hose;
- * Externally-visible vga hose bases
+static struct resource alpha_vga = {
- */
+        .name   = "alpha-vga+",
-unsigned long __vga_hose_io_base = 0;   /* base for default hose */
+        .start  = 0x3C0,
-unsigned long __vga_hose_mem_base = 0;  /* base for default hose */
+        .end    = 0x3DF
+};
 static struct pci_controller * __init 
 default_vga_hose_select(struct pci_controller *h1, struct pci_controller *h2)
@@ -30,36 +34,58 @@ default_vga_hose_select(struct pci_controller *h1, struct pci_controller *h2)
 }
 void __init 
-set_vga_hose(struct pci_controller *hose)
-{
-        if (hose) {
-                __vga_hose_io_base = hose->io_space->start;
-                __vga_hose_mem_base = hose->mem_space->start;
-        }
-}
-void __init 
 locate_and_init_vga(void *(*sel_func)(void *, void *))
 {
        struct pci_controller *hose = NULL;
        struct pci_dev *dev = NULL;
+        /* Default the select function */
        if (!sel_func) sel_func = (void *)default_vga_hose_select;
+        /* Find the console VGA device */
        for(dev=NULL; (dev=pci_get_class(PCI_CLASS_DISPLAY_VGA << 8, dev));) {
-                if (!hose) hose = dev->sysdata;
+                if (!hose)
-                else hose = sel_func(hose, dev->sysdata);
+                        hose = dev->sysdata;
+                else
+                        hose = sel_func(hose, dev->sysdata);
        }
-        /* Did we already inititialize the correct one? */
+        /* Did we already initialize the correct one? Is there one? */
-        if (conswitchp == &vga_con &&
+        if (!hose || (conswitchp == &vga_con && pci_vga_hose == hose))
-            __vga_hose_io_base == hose->io_space->start &&
-            __vga_hose_mem_base == hose->mem_space->start)
                return;
-        /* Set the VGA hose and init the new console */
+        /* Create a new VGA ioport resource WRT the hose it is on. */
-        set_vga_hose(hose);
+        alpha_vga.start += hose->io_space->start;
+        alpha_vga.end += hose->io_space->start;
+        request_resource(hose->io_space, &alpha_vga);
+        /* Set the VGA hose and init the new console. */
+        pci_vga_hose = hose;
        take_over_console(&vga_con, 0, MAX_NR_CONSOLES-1, 1);
 }
+void __init
+find_console_vga_hose(void)
+{
+        u64 *pu64 = (u64 *)((u64)hwrpb + hwrpb->ctbt_offset);
+        if (pu64[7] == 3) {     /* TERM_TYPE == graphics */
+                struct pci_controller *hose;
+                int h = (pu64[30] >> 24) & 0xff;        /* console hose # */
+                /*
+                 * Our hose numbering DOES match the console's, so find
+                 * the right one...
+                 */
+                for (hose = hose_head; hose; hose = hose->next) {
+                        if (hose->index == h) break;
+                }
+                if (hose) {
+                        printk("Console graphics on hose %d\n", h);
+                        pci_vga_hose = hose;
+                }
+        }
+}
 #endif
diff --git a/arch/alpha/kernel/core_marvel.c b/arch/alpha/kernel/core_marvel.c
index 7f6a98455e74..f10d2eddd2c3 100644
--- a/arch/alpha/kernel/core_marvel.c
+++ b/arch/alpha/kernel/core_marvel.c
@@ -25,6 +25,7 @@
 #include <asm/pgalloc.h>
 #include <asm/tlbflush.h>
 #include <asm/rtc.h>
+#include <asm/vga.h>
 #include "proto.h"
 #include "pci_impl.h"
@@ -367,9 +368,8 @@ marvel_io7_present(gct6_node *node)
 }
 static void __init
-marvel_init_vga_hose(void)
+marvel_find_console_vga_hose(void)
 {
-#ifdef CONFIG_VGA_HOSE
        u64 *pu64 = (u64 *)((u64)hwrpb + hwrpb->ctbt_offset);
        if (pu64[7] == 3) {     /* TERM_TYPE == graphics */
@@ -403,7 +403,6 @@ marvel_init_vga_hose(void)
                        pci_vga_hose = hose;
                }
        }
-#endif /* CONFIG_VGA_HOSE */
 }
 gct6_search_struct gct_wanted_node_list[] = {
@@ -459,7 +458,7 @@ marvel_init_arch(void)
                marvel_init_io7(io7);
        /* Check for graphic console location (if any).  */
-        marvel_init_vga_hose();
+        marvel_find_console_vga_hose();
 }
 void
@@ -684,9 +683,6 @@ __marvel_rtc_io(u8 b, unsigned long addr, int write)
 /*
 * IO map support.
 */
-#define __marvel_is_mem_vga(a)  (((a) >= 0xa0000) && ((a) <= 0xc0000))
 void __iomem *
 marvel_ioremap(unsigned long addr, unsigned long size)
 {
@@ -698,13 +694,9 @@ marvel_ioremap(unsigned long addr, unsigned long size)
        unsigned long pfn;
        /*
-         * Adjust the addr.
+         * Adjust the address.
         */ 
-#ifdef CONFIG_VGA_HOSE
+        FIXUP_MEMADDR_VGA(addr);
-        if (pci_vga_hose && __marvel_is_mem_vga(addr)) {
-                addr += pci_vga_hose->mem_space->start;
-        }
-#endif
        /*
         * Find the hose.
@@ -781,7 +773,9 @@ marvel_ioremap(unsigned long addr, unsigned long size)
                return (void __iomem *) vaddr;
        }
-        return NULL;
+        /* Assume it was already a reasonable address */
+        vaddr = baddr + hose->mem_space->start;
+        return (void __iomem *) vaddr;
 }
 void
@@ -803,21 +797,12 @@ marvel_is_mmio(const volatile void __iomem *xaddr)
                return (addr & 0xFF000000UL) == 0;
 }
-#define __marvel_is_port_vga(a) \
-  (((a) >= 0x3b0) && ((a) < 0x3e0) && ((a) != 0x3b3) && ((a) != 0x3d3))
 #define __marvel_is_port_kbd(a) (((a) == 0x60) || ((a) == 0x64))
 #define __marvel_is_port_rtc(a) (((a) == 0x70) || ((a) == 0x71))
 void __iomem *marvel_ioportmap (unsigned long addr)
 {
-        if (__marvel_is_port_rtc (addr) || __marvel_is_port_kbd(addr))
+        FIXUP_IOADDR_VGA(addr);
-                ;
-#ifdef CONFIG_VGA_HOSE
-        else if (__marvel_is_port_vga (addr) && pci_vga_hose)
-                addr += pci_vga_hose->io_space->start;
-#endif
-        else
-                return NULL;
        return (void __iomem *)addr;
 }
@@ -829,8 +814,14 @@ marvel_ioread8(void __iomem *xaddr)
                return 0;
        else if (__marvel_is_port_rtc(addr))
                return __marvel_rtc_io(0, addr, 0);
-        else
+        else if (marvel_is_ioaddr(addr))
                return __kernel_ldbu(*(vucp)addr);
+        else
+                /* this should catch other legacy addresses
+                   that would normally fail on MARVEL,
+                   because there really is nothing there...
+                */
+                return ~0;
 }
 void
@@ -841,7 +832,7 @@ marvel_iowrite8(u8 b, void __iomem *xaddr)
                return;
        else if (__marvel_is_port_rtc(addr)) 
                __marvel_rtc_io(b, addr, 1);
-        else
+        else if (marvel_is_ioaddr(addr))
                __kernel_stb(b, *(vucp)addr);
 }
diff --git a/arch/alpha/kernel/core_titan.c b/arch/alpha/kernel/core_titan.c
index 3662fef7db9a..819326627b96 100644
--- a/arch/alpha/kernel/core_titan.c
+++ b/arch/alpha/kernel/core_titan.c
@@ -21,6 +21,7 @@
 #include <asm/smp.h>
 #include <asm/pgalloc.h>
 #include <asm/tlbflush.h>
+#include <asm/vga.h>
 #include "proto.h"
 #include "pci_impl.h"
@@ -35,6 +36,11 @@ struct
 } saved_config[4] __attribute__((common));
 /*
+ * Is PChip 1 present? No need to query it more than once.
+ */
+static int titan_pchip1_present;
+/*
 * BIOS32-style PCI interface:
 */
@@ -344,43 +350,17 @@ titan_init_one_pachip_port(titan_pachip_port *port, int index)
 static void __init
 titan_init_pachips(titan_pachip *pachip0, titan_pachip *pachip1)
 {
-        int pchip1_present = TITAN_cchip->csc.csr & 1L<<14;
+        titan_pchip1_present = TITAN_cchip->csc.csr & 1L<<14;
        /* Init the ports in hose order... */
        titan_init_one_pachip_port(&pachip0->g_port, 0);        /* hose 0 */
-        if (pchip1_present)
+        if (titan_pchip1_present)
                titan_init_one_pachip_port(&pachip1->g_port, 1);/* hose 1 */
        titan_init_one_pachip_port(&pachip0->a_port, 2);        /* hose 2 */
-        if (pchip1_present)
+        if (titan_pchip1_present)
                titan_init_one_pachip_port(&pachip1->a_port, 3);/* hose 3 */
 }
-static void __init
-titan_init_vga_hose(void)
-{
-#ifdef CONFIG_VGA_HOSE
-        u64 *pu64 = (u64 *)((u64)hwrpb + hwrpb->ctbt_offset);
-        if (pu64[7] == 3) {     /* TERM_TYPE == graphics */
-                struct pci_controller *hose;
-                int h = (pu64[30] >> 24) & 0xff;        /* console hose # */
-                /*
-                 * Our hose numbering matches the console's, so just find
-                 * the right one...
-                 */
-                for (hose = hose_head; hose; hose = hose->next) {
-                        if (hose->index == h) break;
-                }
-                if (hose) {
-                        printk("Console graphics on hose %d\n", hose->index);
-                        pci_vga_hose = hose;
-                }
-        }
-#endif /* CONFIG_VGA_HOSE */
-}
 void __init
 titan_init_arch(void)
 {
@@ -406,6 +386,7 @@ titan_init_arch(void)
        /* With multiple PCI busses, we play with I/O as physical addrs.  */
        ioport_resource.end = ~0UL;
+        iomem_resource.end = ~0UL;
        /* PCI DMA Direct Mapping is 1GB at 2GB.  */
        __direct_map_base = 0x80000000;
@@ -415,7 +396,7 @@ titan_init_arch(void)
        titan_init_pachips(TITAN_pachip0, TITAN_pachip1);
        /* Check for graphic console location (if any).  */
-        titan_init_vga_hose();
+        find_console_vga_hose();
 }
 static void
@@ -441,9 +422,7 @@ titan_kill_one_pachip_port(titan_pachip_port *port, int index)
 static void
 titan_kill_pachips(titan_pachip *pachip0, titan_pachip *pachip1)
 {
-        int pchip1_present = TITAN_cchip->csc.csr & 1L<<14;
+        if (titan_pchip1_present) {
-        if (pchip1_present) {
                titan_kill_one_pachip_port(&pachip1->g_port, 1);
                titan_kill_one_pachip_port(&pachip1->a_port, 3);
        }
@@ -463,6 +442,14 @@ titan_kill_arch(int mode)
 */
 void __iomem *
+titan_ioportmap(unsigned long addr)
+{
+        FIXUP_IOADDR_VGA(addr);
+        return (void __iomem *)(addr + TITAN_IO_BIAS);
+}
+void __iomem *
 titan_ioremap(unsigned long addr, unsigned long size)
 {
        int h = (addr & TITAN_HOSE_MASK) >> TITAN_HOSE_SHIFT;
@@ -475,14 +462,12 @@ titan_ioremap(unsigned long addr, unsigned long size)
        unsigned long pfn;
        /*
-         * Adjust the addr.
+         * Adjust the address and hose, if necessary.
         */ 
-#ifdef CONFIG_VGA_HOSE
+        if (pci_vga_hose && __is_mem_vga(addr)) {
-        if (pci_vga_hose && __titan_is_mem_vga(addr)) {
                h = pci_vga_hose->index;
                addr += pci_vga_hose->mem_space->start;
        }
-#endif
        /*
         * Find the hose.
@@ -521,8 +506,10 @@ titan_ioremap(unsigned long addr, unsigned long size)
                 * Map it
                 */
                area = get_vm_area(size, VM_IOREMAP);
-                if (!area)
+                if (!area) {
+                        printk("ioremap failed... no vm_area...\n");
                        return NULL;
+                }
                ptes = hose->sg_pci->ptes;
                for (vaddr = (unsigned long)area->addr; 
@@ -539,7 +526,7 @@ titan_ioremap(unsigned long addr, unsigned long size)
                        if (__alpha_remap_area_pages(vaddr,
                                                     pfn << PAGE_SHIFT, 
                                                     PAGE_SIZE, 0)) {
-                                printk("FAILED to map...\n");
+                                printk("FAILED to remap_area_pages...\n");
                                vfree(area->addr);
                                return NULL;
                        }
@@ -551,7 +538,8 @@ titan_ioremap(unsigned long addr, unsigned long size)
                return (void __iomem *) vaddr;
        }
-        return NULL;
+        /* Assume a legacy (read: VGA) address, and return appropriately. */
+        return (void __iomem *)(addr + TITAN_MEM_BIAS);
 }
 void
@@ -574,6 +562,7 @@ titan_is_mmio(const volatile void __iomem *xaddr)
 }
 #ifndef CONFIG_ALPHA_GENERIC
+EXPORT_SYMBOL(titan_ioportmap);
 EXPORT_SYMBOL(titan_ioremap);
 EXPORT_SYMBOL(titan_iounmap);
 EXPORT_SYMBOL(titan_is_mmio);
@@ -750,6 +739,7 @@ titan_agp_info(void)
        if (titan_query_agp(port))
                hosenum = 2;
        if (hosenum < 0 && 
+            titan_pchip1_present &&
            titan_query_agp(port = &TITAN_pachip1->a_port)) 
                hosenum = 3;
        
diff --git a/arch/alpha/kernel/core_tsunami.c b/arch/alpha/kernel/core_tsunami.c
index ce623c6e55e1..ef91e09590d4 100644
--- a/arch/alpha/kernel/core_tsunami.c
+++ b/arch/alpha/kernel/core_tsunami.c
@@ -19,6 +19,7 @@
 #include <asm/ptrace.h>
 #include <asm/smp.h>
+#include <asm/vga.h>
 #include "proto.h"
 #include "pci_impl.h"
@@ -349,6 +350,26 @@ tsunami_init_one_pchip(tsunami_pchip *pchip, int index)
        tsunami_pci_tbi(hose, 0, -1);
 }
+void __iomem *
+tsunami_ioportmap(unsigned long addr)
+{
+        FIXUP_IOADDR_VGA(addr);
+        return (void __iomem *)(addr + TSUNAMI_IO_BIAS);
+}
+void __iomem *
+tsunami_ioremap(unsigned long addr, unsigned long size)
+{
+        FIXUP_MEMADDR_VGA(addr);
+        return (void __iomem *)(addr + TSUNAMI_MEM_BIAS);
+}
+#ifndef CONFIG_ALPHA_GENERIC
+EXPORT_SYMBOL(tsunami_ioportmap);
+EXPORT_SYMBOL(tsunami_ioremap);
+#endif
 void __init
 tsunami_init_arch(void)
 {
@@ -393,6 +414,9 @@ tsunami_init_arch(void)
        tsunami_init_one_pchip(TSUNAMI_pchip0, 0);
        if (TSUNAMI_cchip->csc.csr & 1L<<14)
                tsunami_init_one_pchip(TSUNAMI_pchip1, 1);
+        /* Check for graphic console location (if any).  */
+        find_console_vga_hose();
 }
 static void
diff --git a/arch/alpha/kernel/entry.S b/arch/alpha/kernel/entry.S
index c95e95e1ab04..debc8f03886c 100644
--- a/arch/alpha/kernel/entry.S
+++ b/arch/alpha/kernel/entry.S
@@ -391,11 +391,10 @@ $work_resched:
        bne     $2, $work_resched
 $work_notifysig:
-        mov     $sp, $17
+        mov     $sp, $16
        br      $1, do_switch_stack
-        mov     $5, $21
+        mov     $sp, $17
-        mov     $sp, $18
+        mov     $5, $18
-        mov     $31, $16
        jsr     $26, do_notify_resume
        bsr     $1, undo_switch_stack
        br      restore_all
diff --git a/arch/alpha/kernel/pci_iommu.c b/arch/alpha/kernel/pci_iommu.c
index 6e7d1fe6e935..28c84e55feb9 100644
--- a/arch/alpha/kernel/pci_iommu.c
+++ b/arch/alpha/kernel/pci_iommu.c
@@ -7,6 +7,7 @@
 #include <linux/pci.h>
 #include <linux/slab.h>
 #include <linux/bootmem.h>
+#include <linux/log2.h>
 #include <asm/io.h>
 #include <asm/hwrpb.h>
@@ -53,7 +54,7 @@ size_for_memory(unsigned long max)
 {
        unsigned long mem = max_low_pfn << PAGE_SHIFT;
        if (mem < max)
-                max = 1UL << ceil_log2(mem);
+                max = roundup_pow_of_two(mem);
        return max;
 }
diff --git a/arch/alpha/kernel/proto.h b/arch/alpha/kernel/proto.h
index 95912ecc65e1..708d5ca87782 100644
--- a/arch/alpha/kernel/proto.h
+++ b/arch/alpha/kernel/proto.h
@@ -108,6 +108,15 @@ extern int wildfire_cpuid_to_nid(int);
 extern unsigned long wildfire_node_mem_start(int);
 extern unsigned long wildfire_node_mem_size(int);
+/* console.c */
+#ifdef CONFIG_VGA_HOSE
+extern void find_console_vga_hose(void);
+extern void locate_and_init_vga(void *(*)(void *, void *));
+#else
+static inline void find_console_vga_hose(void) { }
+static inline void locate_and_init_vga(void *(*sel_func)(void *, void *)) { }
+#endif
 /* setup.c */
 extern unsigned long srm_hae;
 extern int boot_cpuid;
diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c
index 915f26345c45..bd5e68cd61e8 100644
--- a/arch/alpha/kernel/setup.c
+++ b/arch/alpha/kernel/setup.c
@@ -43,6 +43,7 @@
 #include <linux/notifier.h>
 #include <asm/setup.h>
 #include <asm/io.h>
+#include <linux/log2.h>
 extern struct atomic_notifier_head panic_notifier_list;
 static int alpha_panic_event(struct notifier_block *, unsigned long, void *);
@@ -1303,7 +1304,7 @@ external_cache_probe(int minsize, int width)
        long size = minsize, maxsize = MAX_BCACHE_SIZE * 2;
        if (maxsize > (max_low_pfn + 1) << PAGE_SHIFT)
-                maxsize = 1 << (floor_log2(max_low_pfn + 1) + PAGE_SHIFT);
+                maxsize = 1 << (ilog2(max_low_pfn + 1) + PAGE_SHIFT);
        /* Get the first block cached. */
        read_mem_block(__va(0), stride, size);
diff --git a/arch/alpha/kernel/signal.c b/arch/alpha/kernel/signal.c
index 7f64aa767d5a..410af4f3140e 100644
--- a/arch/alpha/kernel/signal.c
+++ b/arch/alpha/kernel/signal.c
@@ -32,8 +32,8 @@
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 asmlinkage void ret_from_sys_call(void);
-static int do_signal(sigset_t *, struct pt_regs *, struct switch_stack *,
+static void do_signal(struct pt_regs *, struct switch_stack *,
-                     unsigned long, unsigned long);
+                      unsigned long, unsigned long);
 /*
@@ -146,11 +146,9 @@ sys_rt_sigaction(int sig, const struct sigaction __user *act,
 asmlinkage int
 do_sigsuspend(old_sigset_t mask, struct pt_regs *regs, struct switch_stack *sw)
 {
-        sigset_t oldset;
        mask &= _BLOCKABLE;
        spin_lock_irq(&current->sighand->siglock);
-        oldset = current->blocked;
+        current->saved_sigmask = current->blocked;
        siginitset(&current->blocked, mask);
        recalc_sigpending();
        spin_unlock_irq(&current->sighand->siglock);
@@ -160,19 +158,17 @@ do_sigsuspend(old_sigset_t mask, struct pt_regs *regs, struct switch_stack *sw)
        regs->r0 = EINTR;
        regs->r19 = 1;
-        while (1) {
+        current->state = TASK_INTERRUPTIBLE;
-                current->state = TASK_INTERRUPTIBLE;
+        schedule();
-                schedule();
+        set_thread_flag(TIF_RESTORE_SIGMASK);
-                if (do_signal(&oldset, regs, sw, 0, 0))
+        return -ERESTARTNOHAND;
-                        return -EINTR;
-        }
 }
 asmlinkage int
 do_rt_sigsuspend(sigset_t __user *uset, size_t sigsetsize,
                 struct pt_regs *regs, struct switch_stack *sw)
 {
-        sigset_t oldset, set;
+        sigset_t set;
        /* XXX: Don't preclude handling different sized sigset_t's.  */
        if (sigsetsize != sizeof(sigset_t))
@@ -182,7 +178,7 @@ do_rt_sigsuspend(sigset_t __user *uset, size_t sigsetsize,
        sigdelsetmask(&set, ~_BLOCKABLE);
        spin_lock_irq(&current->sighand->siglock);
-        oldset = current->blocked;
+        current->saved_sigmask = current->blocked;
        current->blocked = set;
        recalc_sigpending();
        spin_unlock_irq(&current->sighand->siglock);
@@ -192,12 +188,10 @@ do_rt_sigsuspend(sigset_t __user *uset, size_t sigsetsize,
        regs->r0 = EINTR;
        regs->r19 = 1;
-        while (1) {
+        current->state = TASK_INTERRUPTIBLE;
-                current->state = TASK_INTERRUPTIBLE;
+        schedule();
-                schedule();
+        set_thread_flag(TIF_RESTORE_SIGMASK);
-                if (do_signal(&oldset, regs, sw, 0, 0))
+        return -ERESTARTNOHAND;
-                        return -EINTR;
-        }
 }
 asmlinkage int
@@ -436,7 +430,7 @@ setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs,
        return err;
 }
-static void
+static int
 setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
            struct pt_regs *regs, struct switch_stack * sw)
 {
@@ -481,13 +475,14 @@ setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
                current->comm, current->pid, frame, regs->pc, regs->r26);
 #endif
-        return;
+        return 0;
 give_sigsegv:
        force_sigsegv(sig, current);
+        return -EFAULT;
 }
-static void
+static int
 setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
               sigset_t *set, struct pt_regs *regs, struct switch_stack * sw)
 {
@@ -543,34 +538,38 @@ setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
                current->comm, current->pid, frame, regs->pc, regs->r26);
 #endif
-        return;
+        return 0;
 give_sigsegv:
        force_sigsegv(sig, current);
+        return -EFAULT;
 }
 /*
 * OK, we're invoking a handler.
 */
-static inline void
+static inline int
 handle_signal(int sig, struct k_sigaction *ka, siginfo_t *info,
              sigset_t *oldset, struct pt_regs * regs, struct switch_stack *sw)
 {
+        int ret;
        if (ka->sa.sa_flags & SA_SIGINFO)
-                setup_rt_frame(sig, ka, info, oldset, regs, sw);
+                ret = setup_rt_frame(sig, ka, info, oldset, regs, sw);
        else
-                setup_frame(sig, ka, oldset, regs, sw);
+                ret = setup_frame(sig, ka, oldset, regs, sw);
-        if (ka->sa.sa_flags & SA_RESETHAND)
+        if (ret == 0) {
-                ka->sa.sa_handler = SIG_DFL;
+                spin_lock_irq(&current->sighand->siglock);
+                sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
+                if (!(ka->sa.sa_flags & SA_NODEFER)) 
+                        sigaddset(&current->blocked,sig);
+                recalc_sigpending();
+                spin_unlock_irq(&current->sighand->siglock);
+        }
-        spin_lock_irq(&current->sighand->siglock);
+        return ret;
-        sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
-        if (!(ka->sa.sa_flags & SA_NODEFER)) 
-                sigaddset(&current->blocked,sig);
-        recalc_sigpending();
-        spin_unlock_irq(&current->sighand->siglock);
 }
 static inline void
@@ -611,30 +610,42 @@ syscall_restart(unsigned long r0, unsigned long r19,
 * restart. "r0" is also used as an indicator whether we can restart at
 * all (if we get here from anything but a syscall return, it will be 0)
 */
-static int
+static void
-do_signal(sigset_t *oldset, struct pt_regs * regs, struct switch_stack * sw,
+do_signal(struct pt_regs * regs, struct switch_stack * sw,
          unsigned long r0, unsigned long r19)
 {
        siginfo_t info;
        int signr;
        unsigned long single_stepping = ptrace_cancel_bpt(current);
        struct k_sigaction ka;
+        sigset_t *oldset;
-        if (!oldset)
+        if (test_thread_flag(TIF_RESTORE_SIGMASK))
+                oldset = &current->saved_sigmask;
+        else
                oldset = &current->blocked;
        /* This lets the debugger run, ... */
        signr = get_signal_to_deliver(&info, &ka, regs, NULL);
        /* ... so re-check the single stepping. */
        single_stepping |= ptrace_cancel_bpt(current);
        if (signr > 0) {
                /* Whee!  Actually deliver the signal.  */
-                if (r0) syscall_restart(r0, r19, regs, &ka);
+                if (r0)
-                handle_signal(signr, &ka, &info, oldset, regs, sw);
+                        syscall_restart(r0, r19, regs, &ka);
+                if (handle_signal(signr, &ka, &info, oldset, regs, sw) == 0) {
+                        /* A signal was successfully delivered, and the
+                           saved sigmask was stored on the signal frame,
+                           and will be restored by sigreturn.  So we can
+                           simply clear the restore sigmask flag.  */
+                        if (test_thread_flag(TIF_RESTORE_SIGMASK))
+                                clear_thread_flag(TIF_RESTORE_SIGMASK);
+                }
                if (single_stepping) 
                        ptrace_set_bpt(current); /* re-set bpt */
-                return 1;
+                return;
        }
        if (r0) {
@@ -654,17 +665,22 @@ do_signal(sigset_t *oldset, struct pt_regs * regs, struct switch_stack * sw,
                        break;
                }
        }
+        /* If there's no signal to deliver, we just restore the saved mask.  */
+        if (test_thread_flag(TIF_RESTORE_SIGMASK)) {
+                clear_thread_flag(TIF_RESTORE_SIGMASK);
+                sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
+        }
        if (single_stepping)
                ptrace_set_bpt(current);        /* re-set breakpoint */
-        return 0;
 }
 void
-do_notify_resume(sigset_t *oldset, struct pt_regs *regs,
+do_notify_resume(struct pt_regs *regs, struct switch_stack *sw,
-                 struct switch_stack *sw, unsigned long r0,
+                 unsigned long thread_info_flags,
-                 unsigned long r19, unsigned long thread_info_flags)
+                 unsigned long r0, unsigned long r19)
 {
-        if (thread_info_flags & _TIF_SIGPENDING)
+        if (thread_info_flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK))
-                do_signal(oldset, regs, sw, r0, r19);
+                do_signal(regs, sw, r0, r19);
 }
diff --git a/arch/alpha/kernel/sys_dp264.c b/arch/alpha/kernel/sys_dp264.c
index 85d2f933dd07..c71b0fd7a61f 100644
--- a/arch/alpha/kernel/sys_dp264.c
+++ b/arch/alpha/kernel/sys_dp264.c
@@ -543,6 +543,7 @@ dp264_init_pci(void)
 {
        common_init_pci();
        SMC669_Init(0);
+        locate_and_init_vga(NULL);
 }
 static void __init
@@ -551,6 +552,14 @@ monet_init_pci(void)
        common_init_pci();
        SMC669_Init(1);
        es1888_init();
+        locate_and_init_vga(NULL);
+}
+static void __init
+clipper_init_pci(void)
+{
+        common_init_pci();
+        locate_and_init_vga(NULL);
 }
 static void __init
@@ -655,7 +664,7 @@ struct alpha_machine_vector clipper_mv __initmv = {
        .init_arch              = tsunami_init_arch,
        .init_irq               = clipper_init_irq,
        .init_rtc               = common_init_rtc,
-        .init_pci               = common_init_pci,
+        .init_pci               = clipper_init_pci,
        .kill_arch              = tsunami_kill_arch,
        .pci_map_irq            = clipper_map_irq,
        .pci_swizzle            = common_swizzle,
diff --git a/arch/alpha/kernel/sys_marvel.c b/arch/alpha/kernel/sys_marvel.c
index e349f03b830e..0bcb968cb60a 100644
--- a/arch/alpha/kernel/sys_marvel.c
+++ b/arch/alpha/kernel/sys_marvel.c
@@ -22,6 +22,7 @@
 #include <asm/core_marvel.h>
 #include <asm/hwrpb.h>
 #include <asm/tlbflush.h>
+#include <asm/vga.h>
 #include "proto.h"
 #include "err_impl.h"
@@ -412,10 +413,7 @@ marvel_init_pci(void)
        pci_probe_only = 1;
        common_init_pci();
-#ifdef CONFIG_VGA_HOSE
        locate_and_init_vga(NULL);
-#endif
        /* Clear any io7 errors.  */
        for (io7 = NULL; (io7 = marvel_next_io7(io7)) != NULL; ) 
diff --git a/arch/alpha/kernel/sys_titan.c b/arch/alpha/kernel/sys_titan.c
index f009b7bc0943..1d3c1398c428 100644
--- a/arch/alpha/kernel/sys_titan.c
+++ b/arch/alpha/kernel/sys_titan.c
@@ -331,9 +331,7 @@ titan_init_pci(void)
        pci_probe_only = 1;
        common_init_pci();
        SMC669_Init(0);
-#ifdef CONFIG_VGA_HOSE
        locate_and_init_vga(NULL);
-#endif
 }
diff --git a/arch/alpha/kernel/systbls.S b/arch/alpha/kernel/systbls.S
index f6cfe8ce3f96..79de99e32c35 100644
--- a/arch/alpha/kernel/systbls.S
+++ b/arch/alpha/kernel/systbls.S
@@ -465,6 +465,38 @@ sys_call_table:
        .quad sys_inotify_init
        .quad sys_inotify_add_watch             /* 445 */
        .quad sys_inotify_rm_watch
+        .quad sys_fdatasync
+        .quad sys_kexec_load
+        .quad sys_migrate_pages
+        .quad sys_openat                        /* 450 */
+        .quad sys_mkdirat
+        .quad sys_mknodat
+        .quad sys_fchownat
+        .quad sys_futimesat
+        .quad sys_fstatat64                     /* 455 */
+        .quad sys_unlinkat
+        .quad sys_renameat
+        .quad sys_linkat
+        .quad sys_symlinkat
+        .quad sys_readlinkat                    /* 460 */
+        .quad sys_fchmodat
+        .quad sys_faccessat
+        .quad sys_pselect6
+        .quad sys_ppoll
+        .quad sys_unshare                       /* 465 */
+        .quad sys_set_robust_list
+        .quad sys_get_robust_list
+        .quad sys_splice
+        .quad sys_sync_file_range
+        .quad sys_tee                           /* 470 */
+        .quad sys_vmsplice
+        .quad sys_move_pages
+        .quad sys_getcpu
+        .quad sys_epoll_pwait
+        .quad sys_utimensat                     /* 475 */
+        .quad sys_signalfd
+        .quad sys_timerfd
+        .quad sys_eventfd
        .size sys_call_table, . - sys_call_table
        .type sys_call_table, @object
diff --git a/arch/alpha/lib/Makefile b/arch/alpha/lib/Makefile
index ea098f3b629f..266f78e13076 100644
--- a/arch/alpha/lib/Makefile
+++ b/arch/alpha/lib/Makefile
@@ -37,7 +37,8 @@ lib-y =	__divqu.o __remqu.o __divlu.o __remlu.o \
        $(ev6-y)clear_page.o \
        $(ev6-y)copy_page.o \
        fpreg.o \
-        callback_srm.o srm_puts.o srm_printk.o
+        callback_srm.o srm_puts.o srm_printk.o \
+        fls.o
 lib-$(CONFIG_SMP) += dec_and_lock.o
diff --git a/arch/alpha/lib/fls.c b/arch/alpha/lib/fls.c
new file mode 100644
index 000000000000..7ad84ea0acf8
--- /dev/null
+++ b/arch/alpha/lib/fls.c
@@ -0,0 +1,38 @@
+/* 
+ * arch/alpha/lib/fls.c
+ */
+#include <linux/module.h>
+#include <asm/bitops.h>
+/* This is fls(x)-1, except zero is held to zero.  This allows most
+   efficent input into extbl, plus it allows easy handling of fls(0)=0.  */
+const unsigned char __flsm1_tab[256] = 
+{
+  0,
+  0,
+  1, 1,
+  2, 2, 2, 2,
+  3, 3, 3, 3, 3, 3, 3, 3,
+  4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4,
+  5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5,
+  5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5,
+  6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
+  6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
+  6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
+  6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
+  7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+  7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+  7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+  7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+  7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+  7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+  7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+  7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+};
+EXPORT_SYMBOL(__flsm1_tab);
diff --git a/arch/arm/kernel/armksyms.c b/arch/arm/kernel/armksyms.c
index 9179e8220314..f73d62e8ab60 100644
--- a/arch/arm/kernel/armksyms.c
+++ b/arch/arm/kernel/armksyms.c
@@ -57,7 +57,7 @@ extern void fp_enter(void);
 #define EXPORT_SYMBOL_ALIAS(sym,orig)           \
 EXPORT_CRC_ALIAS(sym)                          \
 static const struct kernel_symbol __ksymtab_##sym      \
-  __attribute_used__ __attribute__((section("__ksymtab"))) =    \
+  __used __attribute__((section("__ksymtab"))) =        \
    { (unsigned long)&orig, #sym };
 /*
diff --git a/arch/arm/kernel/stacktrace.c b/arch/arm/kernel/stacktrace.c
index 8b63ad89d0a8..ae31deb2d065 100644
--- a/arch/arm/kernel/stacktrace.c
+++ b/arch/arm/kernel/stacktrace.c
@@ -13,7 +13,7 @@ int walk_stackframe(unsigned long fp, unsigned long low, unsigned long high,
                /*
                 * Check current frame pointer is within bounds
                 */
-                if ((fp - 12) < low || fp + 4 >= high)
+                if (fp < (low + 12) || fp + 4 >= high)
                        break;
                frame = (struct stackframe *)(fp - 12);
diff --git a/arch/arm/mach-at91/board-dk.c b/arch/arm/mach-at91/board-dk.c
index 6043c38c0a9e..af497896a96c 100644
--- a/arch/arm/mach-at91/board-dk.c
+++ b/arch/arm/mach-at91/board-dk.c
@@ -132,7 +132,7 @@ static struct mtd_partition __initdata dk_nand_partition[] = {
        },
 };
-static struct mtd_partition *nand_partitions(int size, int *num_partitions)
+static struct mtd_partition * __init nand_partitions(int size, int *num_partitions)
 {
        *num_partitions = ARRAY_SIZE(dk_nand_partition);
        return dk_nand_partition;
diff --git a/arch/arm/mach-at91/board-kb9202.c b/arch/arm/mach-at91/board-kb9202.c
index 76f6e1e553ea..7d9b1a278fd6 100644
--- a/arch/arm/mach-at91/board-kb9202.c
+++ b/arch/arm/mach-at91/board-kb9202.c
@@ -96,7 +96,7 @@ static struct mtd_partition __initdata kb9202_nand_partition[] = {
        },
 };
-static struct mtd_partition *nand_partitions(int size, int *num_partitions)
+static struct mtd_partition * __init nand_partitions(int size, int *num_partitions)
 {
        *num_partitions = ARRAY_SIZE(kb9202_nand_partition);
        return kb9202_nand_partition;
diff --git a/arch/arm/mach-at91/board-sam9261ek.c b/arch/arm/mach-at91/board-sam9261ek.c
index 1f0c8a400b3a..26ca8ab3f62a 100644
--- a/arch/arm/mach-at91/board-sam9261ek.c
+++ b/arch/arm/mach-at91/board-sam9261ek.c
@@ -178,7 +178,7 @@ static struct mtd_partition __initdata ek_nand_partition[] = {
        },
 };
-static struct mtd_partition *nand_partitions(int size, int *num_partitions)
+static struct mtd_partition * __init nand_partitions(int size, int *num_partitions)
 {
        *num_partitions = ARRAY_SIZE(ek_nand_partition);
        return ek_nand_partition;
diff --git a/arch/arm/mach-at91/board-sam9263ek.c b/arch/arm/mach-at91/board-sam9263ek.c
index f57458559fb6..c164c8e58ae6 100644
--- a/arch/arm/mach-at91/board-sam9263ek.c
+++ b/arch/arm/mach-at91/board-sam9263ek.c
@@ -180,7 +180,7 @@ static struct mtd_partition __initdata ek_nand_partition[] = {
        },
 };
-static struct mtd_partition *nand_partitions(int size, int *num_partitions)
+static struct mtd_partition * __init nand_partitions(int size, int *num_partitions)
 {
        *num_partitions = ARRAY_SIZE(ek_nand_partition);
        return ek_nand_partition;
diff --git a/arch/arm/mach-at91/board-sam9rlek.c b/arch/arm/mach-at91/board-sam9rlek.c
index 30c79aca84d4..9b61320f295a 100644
--- a/arch/arm/mach-at91/board-sam9rlek.c
+++ b/arch/arm/mach-at91/board-sam9rlek.c
@@ -87,7 +87,7 @@ static struct mtd_partition __initdata ek_nand_partition[] = {
        },
 };
-static struct mtd_partition *nand_partitions(int size, int *num_partitions)
+static struct mtd_partition * __init nand_partitions(int size, int *num_partitions)
 {
        *num_partitions = ARRAY_SIZE(ek_nand_partition);
        return ek_nand_partition;
diff --git a/arch/arm/mach-footbridge/cats-pci.c b/arch/arm/mach-footbridge/cats-pci.c
index 4f984fde7375..35eb232a649a 100644
--- a/arch/arm/mach-footbridge/cats-pci.c
+++ b/arch/arm/mach-footbridge/cats-pci.c
@@ -45,7 +45,7 @@ static struct hw_pci cats_pci __initdata = {
        .postinit               = dc21285_postinit,
 };
-static int cats_pci_init(void)
+static int __init cats_pci_init(void)
 {
        if (machine_is_cats())
                pci_common_init(&cats_pci);
diff --git a/arch/arm/mach-imx/generic.c b/arch/arm/mach-imx/generic.c
index 7a7fa51ec62c..1c474cf709ca 100644
--- a/arch/arm/mach-imx/generic.c
+++ b/arch/arm/mach-imx/generic.c
@@ -201,7 +201,6 @@ void __init imx_set_mmc_info(struct imxmmc_platform_data *info)
 {
        imx_mmc_device.dev.platform_data = info;
 }
-EXPORT_SYMBOL(imx_set_mmc_info);
 static struct imxfb_mach_info imx_fb_info;
diff --git a/arch/arm/mach-ixp2000/ixdp2400.c b/arch/arm/mach-ixp2000/ixdp2400.c
index 0fdd03ab36e6..ce7c15c73004 100644
--- a/arch/arm/mach-ixp2000/ixdp2400.c
+++ b/arch/arm/mach-ixp2000/ixdp2400.c
@@ -164,7 +164,7 @@ int __init ixdp2400_pci_init(void)
 subsys_initcall(ixdp2400_pci_init);
-void ixdp2400_init_irq(void)
+void __init ixdp2400_init_irq(void)
 {
        ixdp2x00_init_irq(IXDP2400_CPLD_INT_STAT, IXDP2400_CPLD_INT_MASK, IXDP2400_NR_IRQS);
 }
diff --git a/arch/arm/mach-ixp2000/ixdp2800.c b/arch/arm/mach-ixp2000/ixdp2800.c
index 70d247f09a7e..14f09b80ab77 100644
--- a/arch/arm/mach-ixp2000/ixdp2800.c
+++ b/arch/arm/mach-ixp2000/ixdp2800.c
@@ -279,7 +279,7 @@ int __init ixdp2800_pci_init(void)
 subsys_initcall(ixdp2800_pci_init);
-void ixdp2800_init_irq(void)
+void __init ixdp2800_init_irq(void)
 {
        ixdp2x00_init_irq(IXDP2800_CPLD_INT_STAT, IXDP2800_CPLD_INT_MASK, IXDP2800_NR_IRQS);
 }
diff --git a/arch/arm/mach-ixp2000/ixdp2x00.c b/arch/arm/mach-ixp2000/ixdp2x00.c
index 011065b967b4..73c651e83d92 100644
--- a/arch/arm/mach-ixp2000/ixdp2x00.c
+++ b/arch/arm/mach-ixp2000/ixdp2x00.c
@@ -145,7 +145,7 @@ static struct irq_chip ixdp2x00_cpld_irq_chip = {
        .unmask = ixdp2x00_irq_unmask
 };
-void ixdp2x00_init_irq(volatile unsigned long *stat_reg, volatile unsigned long *mask_reg, unsigned long nr_irqs)
+void __init ixdp2x00_init_irq(volatile unsigned long *stat_reg, volatile unsigned long *mask_reg, unsigned long nr_irqs)
 {
        unsigned int irq;
diff --git a/arch/arm/mach-ixp23xx/ixdp2351.c b/arch/arm/mach-ixp23xx/ixdp2351.c
index 7a86a2516eaa..c41a6b5a0acc 100644
--- a/arch/arm/mach-ixp23xx/ixdp2351.c
+++ b/arch/arm/mach-ixp23xx/ixdp2351.c
@@ -124,7 +124,7 @@ static struct irq_chip ixdp2351_intb_chip = {
        .unmask = ixdp2351_intb_unmask
 };
-void ixdp2351_init_irq(void)
+void __init ixdp2351_init_irq(void)
 {
        int irq;
diff --git a/arch/arm/mach-ixp23xx/pci.c b/arch/arm/mach-ixp23xx/pci.c
index ac7d43d23c28..227f808dc0ec 100644
--- a/arch/arm/mach-ixp23xx/pci.c
+++ b/arch/arm/mach-ixp23xx/pci.c
@@ -284,7 +284,7 @@ int ixp23xx_pci_setup(int nr, struct pci_sys_data *sys)
        return 1;
 }
-void ixp23xx_pci_slave_init(void)
+void __init ixp23xx_pci_slave_init(void)
 {
        ixp23xx_pci_common_init();
 }
diff --git a/arch/arm/mach-ixp23xx/roadrunner.c b/arch/arm/mach-ixp23xx/roadrunner.c
index d06e21b70de5..e35644961aa4 100644
--- a/arch/arm/mach-ixp23xx/roadrunner.c
+++ b/arch/arm/mach-ixp23xx/roadrunner.c
@@ -110,7 +110,7 @@ static int __init roadrunner_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
        return NO_IRQ;
 }
-static void roadrunner_pci_preinit(void)
+static void __init roadrunner_pci_preinit(void)
 {
        set_irq_type(IRQ_ROADRUNNER_PCI_INTC, IRQT_LOW);
        set_irq_type(IRQ_ROADRUNNER_PCI_INTD, IRQT_LOW);
diff --git a/arch/arm/mach-ixp4xx/Kconfig b/arch/arm/mach-ixp4xx/Kconfig
index 9715ef506c24..060909870b50 100644
--- a/arch/arm/mach-ixp4xx/Kconfig
+++ b/arch/arm/mach-ixp4xx/Kconfig
@@ -104,9 +104,6 @@ config MACH_DSMG600
          DSM-G600 RevA device. For more information on this platform,
          see http://www.nslu2-linux.org/wiki/DSMG600/HomePage
-#
-# Avila and IXDP share the same source for now. Will change in future
-#
 config  ARCH_IXDP4XX
        bool
        depends on ARCH_IXDP425 || MACH_IXDP465 || MACH_KIXRP435
diff --git a/arch/arm/mach-ixp4xx/common.c b/arch/arm/mach-ixp4xx/common.c
index 64685da1462d..8112f726ffa0 100644
--- a/arch/arm/mach-ixp4xx/common.c
+++ b/arch/arm/mach-ixp4xx/common.c
@@ -283,7 +283,7 @@ static struct irqaction ixp4xx_timer_irq = {
        .handler        = ixp4xx_timer_interrupt,
 };
-static void __init ixp4xx_timer_init(void)
+void __init ixp4xx_timer_init(void)
 {
        /* Reset/disable counter */
        *IXP4XX_OSRT1 = 0;
diff --git a/arch/arm/mach-ixp4xx/coyote-pci.c b/arch/arm/mach-ixp4xx/coyote-pci.c
index 7bc94f3def1c..ad2e5b97966e 100644
--- a/arch/arm/mach-ixp4xx/coyote-pci.c
+++ b/arch/arm/mach-ixp4xx/coyote-pci.c
@@ -25,10 +25,6 @@
 #include <asm/mach/pci.h>
-extern void ixp4xx_pci_preinit(void);
-extern int ixp4xx_setup(int nr, struct pci_sys_data *sys);
-extern struct pci_bus *ixp4xx_scan_bus(int nr, struct pci_sys_data *sys);
 void __init coyote_pci_preinit(void)
 {
        set_irq_type(IRQ_COYOTE_PCI_SLOT0, IRQT_LOW);
diff --git a/arch/arm/mach-ixp4xx/dsmg600-setup.c b/arch/arm/mach-ixp4xx/dsmg600-setup.c
index 1caff65e22cc..1e75e105c4f7 100644
--- a/arch/arm/mach-ixp4xx/dsmg600-setup.c
+++ b/arch/arm/mach-ixp4xx/dsmg600-setup.c
@@ -18,6 +18,7 @@
 #include <asm/mach-types.h>
 #include <asm/mach/arch.h>
 #include <asm/mach/flash.h>
+#include <asm/mach/time.h>
 static struct flash_platform_data dsmg600_flash_data = {
        .map_name               = "cfi_probe",
@@ -128,6 +129,19 @@ static void dsmg600_power_off(void)
        gpio_line_set(DSMG600_PO_GPIO, IXP4XX_GPIO_HIGH);
 }
+static void __init dsmg600_timer_init(void)
+{
+    /* The xtal on this machine is non-standard. */
+    ixp4xx_timer_freq = DSMG600_FREQ;
+    /* Call standard timer_init function. */
+    ixp4xx_timer_init();
+}
+static struct sys_timer dsmg600_timer = {
+    .init   = dsmg600_timer_init,
+};
 static void __init dsmg600_init(void)
 {
        ixp4xx_sys_init();
@@ -155,21 +169,13 @@ static void __init dsmg600_init(void)
 #endif
 }
-static void __init dsmg600_fixup(struct machine_desc *desc,
-                struct tag *tags, char **cmdline, struct meminfo *mi)
-{
-       /* The xtal on this machine is non-standard. */
-        ixp4xx_timer_freq = DSMG600_FREQ;
-}
 MACHINE_START(DSMG600, "D-Link DSM-G600 RevA")
        /* Maintainer: www.nslu2-linux.org */
        .phys_io        = IXP4XX_PERIPHERAL_BASE_PHYS,
        .io_pg_offst    = ((IXP4XX_PERIPHERAL_BASE_VIRT) >> 18) & 0xFFFC,
        .boot_params    = 0x00000100,
-        .fixup          = dsmg600_fixup,
        .map_io         = ixp4xx_map_io,
        .init_irq       = ixp4xx_init_irq,
-        .timer          = &ixp4xx_timer,
+        .timer          = &dsmg600_timer,
        .init_machine   = dsmg600_init,
 MACHINE_END
diff --git a/arch/arm/mach-ixp4xx/ixdpg425-pci.c b/arch/arm/mach-ixp4xx/ixdpg425-pci.c
index 509a95a692a4..d1e75b7dc3b1 100644
--- a/arch/arm/mach-ixp4xx/ixdpg425-pci.c
+++ b/arch/arm/mach-ixp4xx/ixdpg425-pci.c
@@ -23,10 +23,6 @@
 #include <asm/mach/pci.h>
-extern void ixp4xx_pci_preinit(void);
-extern int ixp4xx_setup(int nr, struct pci_sys_data *sys);
-extern struct pci_bus *ixp4xx_scan_bus(int nr, struct pci_sys_data *sys);
 void __init ixdpg425_pci_preinit(void)
 {
        set_irq_type(IRQ_IXP4XX_GPIO6, IRQT_LOW);
diff --git a/arch/arm/mach-ixp4xx/nas100d-setup.c b/arch/arm/mach-ixp4xx/nas100d-setup.c
index 9a31444d9214..78a17413ceca 100644
--- a/arch/arm/mach-ixp4xx/nas100d-setup.c
+++ b/arch/arm/mach-ixp4xx/nas100d-setup.c
@@ -155,7 +155,8 @@ static void __init nas100d_init(void)
        pm_power_off = nas100d_power_off;
-        /* This is only useful on a modified machine, but it is valuable
+        /*
+         * This is only useful on a modified machine, but it is valuable
         * to have it first in order to see debug messages, and so that
         * it does *not* get removed if platform_add_devices fails!
         */
diff --git a/arch/arm/mach-ixp4xx/nslu2-setup.c b/arch/arm/mach-ixp4xx/nslu2-setup.c
index 162c266e5f8f..9bf8ccbcaccf 100644
--- a/arch/arm/mach-ixp4xx/nslu2-setup.c
+++ b/arch/arm/mach-ixp4xx/nslu2-setup.c
@@ -22,6 +22,7 @@
 #include <asm/mach-types.h>
 #include <asm/mach/arch.h>
 #include <asm/mach/flash.h>
+#include <asm/mach/time.h>
 static struct flash_platform_data nslu2_flash_data = {
        .map_name               = "cfi_probe",
@@ -49,26 +50,26 @@ static struct ixp4xx_i2c_pins nslu2_i2c_gpio_pins = {
 static struct resource nslu2_led_resources[] = {
        {
                .name           = "ready",  /* green led */
-                .start          = NSLU2_LED_GRN,
+                .start          = NSLU2_LED_GRN_GPIO,
-                .end            = NSLU2_LED_GRN,
+                .end            = NSLU2_LED_GRN_GPIO,
                .flags          = IXP4XX_GPIO_HIGH,
        },
        {
                .name           = "status", /* red led */
-                .start          = NSLU2_LED_RED,
+                .start          = NSLU2_LED_RED_GPIO,
-                .end            = NSLU2_LED_RED,
+                .end            = NSLU2_LED_RED_GPIO,
                .flags          = IXP4XX_GPIO_HIGH,
        },
        {
                .name           = "disk-1",
-                .start          = NSLU2_LED_DISK1,
+                .start          = NSLU2_LED_DISK1_GPIO,
-                .end            = NSLU2_LED_DISK1,
+                .end            = NSLU2_LED_DISK1_GPIO,
                .flags          = IXP4XX_GPIO_LOW,
        },
        {
                .name           = "disk-2",
-                .start          = NSLU2_LED_DISK2,
+                .start          = NSLU2_LED_DISK2_GPIO,
-                .end            = NSLU2_LED_DISK2,
+                .end            = NSLU2_LED_DISK2_GPIO,
                .flags          = IXP4XX_GPIO_LOW,
        },
 };
@@ -157,10 +158,21 @@ static void nslu2_power_off(void)
        gpio_line_set(NSLU2_PO_GPIO, IXP4XX_GPIO_HIGH);
 }
-static void __init nslu2_init(void)
+static void __init nslu2_timer_init(void)
 {
-        ixp4xx_timer_freq = NSLU2_FREQ;
+    /* The xtal on this machine is non-standard. */
+    ixp4xx_timer_freq = NSLU2_FREQ;
+    /* Call standard timer_init function. */
+    ixp4xx_timer_init();
+}
+static struct sys_timer nslu2_timer = {
+    .init   = nslu2_timer_init,
+};
+static void __init nslu2_init(void)
+{
        ixp4xx_sys_init();
        nslu2_flash_resource.start = IXP4XX_EXP_BUS_BASE(0);
@@ -169,7 +181,8 @@ static void __init nslu2_init(void)
        pm_power_off = nslu2_power_off;
-        /* This is only useful on a modified machine, but it is valuable
+        /*
+         * This is only useful on a modified machine, but it is valuable
         * to have it first in order to see debug messages, and so that
         * it does *not* get removed if platform_add_devices fails!
         */
@@ -185,6 +198,6 @@ MACHINE_START(NSLU2, "Linksys NSLU2")
        .boot_params    = 0x00000100,
        .map_io         = ixp4xx_map_io,
        .init_irq       = ixp4xx_init_irq,
-        .timer          = &ixp4xx_timer,
+        .timer          = &nslu2_timer,
        .init_machine   = nslu2_init,
 MACHINE_END
diff --git a/arch/arm/mach-s3c2410/bast.h b/arch/arm/mach-s3c2410/bast.h
deleted file mode 100644
index e98543742eb9..000000000000
--- a/arch/arm/mach-s3c2410/bast.h
+++ /dev/null
@@ -1,2 +0,0 @@
-/* linux/arch/arm/mach-s3c2410/bast.h
-extern void bast_init_irq(void);
diff --git a/arch/arm/mach-s3c2410/mach-amlm5900.c b/arch/arm/mach-s3c2410/mach-amlm5900.c
index bc308ceb91c3..435adcce6482 100644
--- a/arch/arm/mach-s3c2410/mach-amlm5900.c
+++ b/arch/arm/mach-s3c2410/mach-amlm5900.c
@@ -160,7 +160,7 @@ static struct platform_device *amlm5900_devices[] __initdata = {
 #endif
 };
-void __init amlm5900_map_io(void)
+static void __init amlm5900_map_io(void)
 {
        s3c24xx_init_io(amlm5900_iodesc, ARRAY_SIZE(amlm5900_iodesc));
        s3c24xx_init_clocks(0);
diff --git a/arch/arm/mach-s3c2412/s3c2412.c b/arch/arm/mach-s3c2412/s3c2412.c
index c602aa39f9c4..782b5814ced2 100644
--- a/arch/arm/mach-s3c2412/s3c2412.c
+++ b/arch/arm/mach-s3c2412/s3c2412.c
@@ -16,6 +16,7 @@
 #include <linux/list.h>
 #include <linux/timer.h>
 #include <linux/init.h>
+#include <linux/delay.h>
 #include <linux/sysdev.h>
 #include <linux/serial_core.h>
 #include <linux/platform_device.h>
@@ -29,6 +30,7 @@
 #include <asm/io.h>
 #include <asm/irq.h>
+#include <asm/arch/reset.h>
 #include <asm/arch/idle.h>
 #include <asm/arch/regs-clock.h>
@@ -38,6 +40,7 @@
 #include <asm/arch/regs-gpioj.h>
 #include <asm/arch/regs-dsc.h>
 #include <asm/arch/regs-spi.h>
+#include <asm/arch/regs-s3c2412.h>
 #include <asm/plat-s3c24xx/s3c2412.h>
 #include <asm/plat-s3c24xx/cpu.h>
@@ -106,6 +109,23 @@ static void s3c2412_idle(void)
        cpu_do_idle();
 }
+static void s3c2412_hard_reset(void)
+{
+        /* errata "Watch-dog/Software Reset Problem" specifies that
+         * this reset must be done with the SYSCLK sourced from
+         * EXTCLK instead of FOUT to avoid a glitch in the reset
+         * mechanism.
+         *
+         * See the watchdog section of the S3C2412 manual for more
+         * information on this fix.
+         */
+        __raw_writel(0x00, S3C2412_CLKSRC);
+        __raw_writel(S3C2412_SWRST_RESET, S3C2412_SWRST);
+        mdelay(1);
+}
 /* s3c2412_map_io
 *
 * register the standard cpu IO areas, and any passed in from the
@@ -122,6 +142,10 @@ void __init s3c2412_map_io(struct map_desc *mach_desc, int mach_size)
        s3c24xx_idle = s3c2412_idle;
+        /* set custom reset hook */
+        s3c24xx_reset_hook = s3c2412_hard_reset;
        /* register our io-tables */
        iotable_init(s3c2412_iodesc, ARRAY_SIZE(s3c2412_iodesc));
diff --git a/arch/arm/mach-s3c2443/clock.c b/arch/arm/mach-s3c2443/clock.c
index 5955efb5de8d..58402948c47c 100644
--- a/arch/arm/mach-s3c2443/clock.c
+++ b/arch/arm/mach-s3c2443/clock.c
@@ -394,7 +394,7 @@ static int s3c2443_setrate_usbhost(struct clk *clk, unsigned long rate)
        return 0;
 }
-struct clk clk_usb_bus_host = {
+static struct clk clk_usb_bus_host = {
        .name           = "usb-bus-host-parent",
        .id             = -1,
        .parent         = &clk_esysclk,
@@ -758,7 +758,6 @@ static struct clk init_clocks[] = {
                .parent         = &clk_h,
                .enable         = s3c2443_clkcon_enable_h,
                .ctrlbit        = S3C2443_HCLKCON_CFC,
-                .ctrlbit        = S3C2443_HCLKCON_HSMMC,
        }, {
                .name           = "ssmc",
                .id             = -1,
diff --git a/arch/arm/mach-sa1100/neponset.c b/arch/arm/mach-sa1100/neponset.c
index d7c038a0256b..4cbf9468f654 100644
--- a/arch/arm/mach-sa1100/neponset.c
+++ b/arch/arm/mach-sa1100/neponset.c
@@ -139,12 +139,12 @@ static u_int neponset_get_mctrl(struct uart_port *port)
        return ret;
 }
-static struct sa1100_port_fns neponset_port_fns __initdata = {
+static struct sa1100_port_fns neponset_port_fns __devinitdata = {
        .set_mctrl      = neponset_set_mctrl,
        .get_mctrl      = neponset_get_mctrl,
 };
-static int neponset_probe(struct platform_device *dev)
+static int __devinit neponset_probe(struct platform_device *dev)
 {
        sa1100_register_uart_fns(&neponset_port_fns);
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index 5f472a8b406a..e7904bc92c73 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -379,7 +379,7 @@ config CPU_V7
        select CPU_CP15_MMU
        select CPU_HAS_ASID
        select CPU_COPY_V6 if MMU
-        select CPU_TLB_V6 if MMU
+        select CPU_TLB_V7 if MMU
 # Figure out what processor architecture version we should be using.
 # This defines the compiler instruction set which depends on the machine type.
@@ -498,6 +498,9 @@ config CPU_TLB_V4WBI
 config CPU_TLB_V6
        bool
+config CPU_TLB_V7
+        bool
 endif
 config CPU_HAS_ASID
diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile
index b5bd335ff14a..762702765fc3 100644
--- a/arch/arm/mm/Makefile
+++ b/arch/arm/mm/Makefile
@@ -46,6 +46,7 @@ obj-$(CONFIG_CPU_TLB_V4WT)	+= tlb-v4.o
 obj-$(CONFIG_CPU_TLB_V4WB)      += tlb-v4wb.o
 obj-$(CONFIG_CPU_TLB_V4WBI)     += tlb-v4wbi.o
 obj-$(CONFIG_CPU_TLB_V6)        += tlb-v6.o
+obj-$(CONFIG_CPU_TLB_V7)        += tlb-v7.o
 obj-$(CONFIG_CPU_ARM610)        += proc-arm6_7.o
 obj-$(CONFIG_CPU_ARM710)        += proc-arm6_7.o
diff --git a/arch/arm/mm/alignment.c b/arch/arm/mm/alignment.c
index 36440c899583..074b7cb07743 100644
--- a/arch/arm/mm/alignment.c
+++ b/arch/arm/mm/alignment.c
@@ -630,7 +630,7 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
        fs = get_fs();
        set_fs(KERNEL_DS);
-        if thumb_mode(regs) {
+        if (thumb_mode(regs)) {
                fault = __get_user(tinstr, (u16 *)(instrptr & ~1));
                if (!(fault))
                        instr = thumb2arm(tinstr);
diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S
index dd823dd4a374..718f4782ee8b 100644
--- a/arch/arm/mm/proc-v7.S
+++ b/arch/arm/mm/proc-v7.S
@@ -256,7 +256,7 @@ __v7_proc_info:
        .long   HWCAP_SWP|HWCAP_HALF|HWCAP_THUMB|HWCAP_FAST_MULT|HWCAP_EDSP
        .long   cpu_v7_name
        .long   v7_processor_functions
-        .long   v6wbi_tlb_fns
+        .long   v7wbi_tlb_fns
        .long   v6_user_fns
        .long   v7_cache_fns
        .size   __v7_proc_info, . - __v7_proc_info
diff --git a/arch/arm/mm/tlb-v7.S b/arch/arm/mm/tlb-v7.S
new file mode 100644
index 000000000000..b56dda8052f7
--- /dev/null
+++ b/arch/arm/mm/tlb-v7.S
@@ -0,0 +1,88 @@
+/*
+ *  linux/arch/arm/mm/tlb-v7.S
+ *
+ *  Copyright (C) 1997-2002 Russell King
+ *  Modified for ARMv7 by Catalin Marinas
+ *
+ * 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.
+ *
+ *  ARM architecture version 6 TLB handling functions.
+ *  These assume a split I/D TLB.
+ */
+#include <linux/linkage.h>
+#include <asm/asm-offsets.h>
+#include <asm/page.h>
+#include <asm/tlbflush.h>
+#include "proc-macros.S"
+/*
+ *      v7wbi_flush_user_tlb_range(start, end, vma)
+ *
+ *      Invalidate a range of TLB entries in the specified address space.
+ *
+ *      - start - start address (may not be aligned)
+ *      - end   - end address (exclusive, may not be aligned)
+ *      - vma   - vma_struct describing address range
+ *
+ *      It is assumed that:
+ *      - the "Invalidate single entry" instruction will invalidate
+ *        both the I and the D TLBs on Harvard-style TLBs
+ */
+ENTRY(v7wbi_flush_user_tlb_range)
+        vma_vm_mm r3, r2                        @ get vma->vm_mm
+        mmid    r3, r3                          @ get vm_mm->context.id
+        dsb
+        mov     r0, r0, lsr #PAGE_SHIFT         @ align address
+        mov     r1, r1, lsr #PAGE_SHIFT
+        asid    r3, r3                          @ mask ASID
+        orr     r0, r3, r0, lsl #PAGE_SHIFT     @ Create initial MVA
+        mov     r1, r1, lsl #PAGE_SHIFT
+        vma_vm_flags r2, r2                     @ get vma->vm_flags
+1:
+        mcr     p15, 0, r0, c8, c6, 1           @ TLB invalidate D MVA (was 1)
+        tst     r2, #VM_EXEC                    @ Executable area ?
+        mcrne   p15, 0, r0, c8, c5, 1           @ TLB invalidate I MVA (was 1)
+        add     r0, r0, #PAGE_SZ
+        cmp     r0, r1
+        blo     1b
+        mov     ip, #0
+        mcr     p15, 0, ip, c7, c5, 6           @ flush BTAC/BTB
+        dsb
+        mov     pc, lr
+/*
+ *      v7wbi_flush_kern_tlb_range(start,end)
+ *
+ *      Invalidate a range of kernel TLB entries
+ *
+ *      - start - start address (may not be aligned)
+ *      - end   - end address (exclusive, may not be aligned)
+ */
+ENTRY(v7wbi_flush_kern_tlb_range)
+        dsb
+        mov     r0, r0, lsr #PAGE_SHIFT         @ align address
+        mov     r1, r1, lsr #PAGE_SHIFT
+        mov     r0, r0, lsl #PAGE_SHIFT
+        mov     r1, r1, lsl #PAGE_SHIFT
+1:
+        mcr     p15, 0, r0, c8, c6, 1           @ TLB invalidate D MVA
+        mcr     p15, 0, r0, c8, c5, 1           @ TLB invalidate I MVA
+        add     r0, r0, #PAGE_SZ
+        cmp     r0, r1
+        blo     1b
+        mov     r2, #0
+        mcr     p15, 0, r2, c7, c5, 6           @ flush BTAC/BTB
+        dsb
+        isb
+        mov     pc, lr
+        .section ".text.init", #alloc, #execinstr
+        .type   v7wbi_tlb_fns, #object
+ENTRY(v7wbi_tlb_fns)
+        .long   v7wbi_flush_user_tlb_range
+        .long   v7wbi_flush_kern_tlb_range
+        .long   v6wbi_tlb_flags
+        .size   v7wbi_tlb_fns, . - v7wbi_tlb_fns
diff --git a/arch/arm/nwfpe/softfloat.h b/arch/arm/nwfpe/softfloat.h
index 0a3067452cd2..260fe29d73f5 100644
--- a/arch/arm/nwfpe/softfloat.h
+++ b/arch/arm/nwfpe/softfloat.h
@@ -273,4 +273,7 @@ static inline flag float64_lt_nocheck(float64 a, float64 b)
 extern flag float32_is_nan( float32 a );
 extern flag float64_is_nan( float64 a );
+extern int32 float64_to_uint32( struct roundingData *roundData, float64 a );
+extern int32 float64_to_uint32_round_to_zero( float64 a );
 #endif
diff --git a/arch/arm/oprofile/op_model_mpcore.c b/arch/arm/oprofile/op_model_mpcore.c
index 898500718249..7791da791f5f 100644
--- a/arch/arm/oprofile/op_model_mpcore.c
+++ b/arch/arm/oprofile/op_model_mpcore.c
@@ -257,8 +257,13 @@ static void em_stop(void)
 */
 static void em_route_irq(int irq, unsigned int cpu)
 {
-        irq_desc[irq].affinity = cpumask_of_cpu(cpu);
+        struct irq_desc *desc = irq_desc + irq;
-        irq_desc[irq].chip->set_affinity(irq, cpumask_of_cpu(cpu));
+        cpumask_t mask = cpumask_of_cpu(cpu);
+        spin_lock_irq(&desc->lock);
+        desc->affinity = mask;
+        desc->chip->set_affinity(irq, mask);
+        spin_unlock_irq(&desc->lock);
 }
 static int em_setup(void)
diff --git a/arch/h8300/kernel/sys_h8300.c b/arch/h8300/kernel/sys_h8300.c
index 11ba75a05220..de7688cfd573 100644
--- a/arch/h8300/kernel/sys_h8300.c
+++ b/arch/h8300/kernel/sys_h8300.c
@@ -288,9 +288,9 @@ asmlinkage void syscall_print(void *dummy,...)
 int kernel_execve(const char *filename, char *const argv[], char *const envp[])
 {
        register long res __asm__("er0");
+        register char *const *_c __asm__("er3") = envp;
+        register char *const *_b __asm__("er2") = argv;
        register const char * _a __asm__("er1") = filename;
-        register void *_b __asm__("er2") = argv;
-        register void *_c __asm__("er3") = envp;
        __asm__ __volatile__ ("mov.l %1,er0\n\t"
                        "trapa  #0\n\t"
                        : "=r" (res)
diff --git a/arch/h8300/kernel/traps.c b/arch/h8300/kernel/traps.c
index 300e3279ca5a..f97183011c2c 100644
--- a/arch/h8300/kernel/traps.c
+++ b/arch/h8300/kernel/traps.c
@@ -136,7 +136,7 @@ void show_stack(struct task_struct *task, unsigned long *esp)
        printk("\nCall Trace:");
        i = 0;
        stack = esp;
-        while (((unsigned long)stack & (THREAD_SIZE - 1)) == 0) {
+        while (((unsigned long)stack & (THREAD_SIZE - 1)) != 0) {
                addr = *stack++;
                /*
                 * If the address is either in the text segment of the
diff --git a/arch/i386/kernel/microcode.c b/arch/i386/kernel/microcode.c
index 83f825f2e2d7..d865d041bea1 100644
--- a/arch/i386/kernel/microcode.c
+++ b/arch/i386/kernel/microcode.c
@@ -478,7 +478,7 @@ static int __init microcode_dev_init (void)
        return 0;
 }
-static void __exit microcode_dev_exit (void)
+static void microcode_dev_exit (void)
 {
        misc_deregister(&microcode_dev);
 }
diff --git a/arch/i386/kernel/reboot.c b/arch/i386/kernel/reboot.c
index 50dfc65319cd..5513f8d5b5be 100644
--- a/arch/i386/kernel/reboot.c
+++ b/arch/i386/kernel/reboot.c
@@ -89,6 +89,14 @@ static int __init set_bios_reboot(struct dmi_system_id *d)
 }
 static struct dmi_system_id __initdata reboot_dmi_table[] = {
+        {       /* Handle problems with rebooting on Dell E520's */
+                .callback = set_bios_reboot,
+                .ident = "Dell E520",
+                .matches = {
+                        DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+                        DMI_MATCH(DMI_PRODUCT_NAME, "Dell DM061"),
+                },
+        },
        {       /* Handle problems with rebooting on Dell 1300's */
                .callback = set_bios_reboot,
                .ident = "Dell PowerEdge 1300",
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index 08f07a74a9d3..88baed1e7e83 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -943,10 +943,9 @@ exit:
 static void smp_tune_scheduling(void)
 {
-        unsigned long cachesize;       /* kB   */
        if (cpu_khz) {
-                cachesize = boot_cpu_data.x86_cache_size;
+                /* cache size in kB */
+                long cachesize = boot_cpu_data.x86_cache_size;
                if (cachesize > 0)
                        max_cache_size = cachesize * 1024;
diff --git a/arch/i386/kernel/vmi.c b/arch/i386/kernel/vmi.c
index c8726c424b35..c12720d7cbc5 100644
--- a/arch/i386/kernel/vmi.c
+++ b/arch/i386/kernel/vmi.c
@@ -27,6 +27,7 @@
 #include <linux/bootmem.h>
 #include <linux/mm.h>
 #include <linux/highmem.h>
+#include <linux/sched.h>
 #include <asm/vmi.h>
 #include <asm/io.h>
 #include <asm/fixmap.h>
diff --git a/arch/i386/oprofile/nmi_int.c b/arch/i386/oprofile/nmi_int.c
index a7c0783b269a..11b7a51566a8 100644
--- a/arch/i386/oprofile/nmi_int.c
+++ b/arch/i386/oprofile/nmi_int.c
@@ -154,7 +154,7 @@ static int allocate_msrs(void)
        size_t counters_size = sizeof(struct op_msr) * model->num_counters;
        int i;
-        for_each_online_cpu(i) {
+        for_each_possible_cpu(i) {
                cpu_msrs[i].counters = kmalloc(counters_size, GFP_KERNEL);
                if (!cpu_msrs[i].counters) {
                        success = 0;
@@ -211,8 +211,14 @@ static int nmi_setup(void)
        /* Assume saved/restored counters are the same on all CPUs */
        model->fill_in_addresses(&cpu_msrs[0]);
        for_each_possible_cpu (cpu) {
-                if (cpu != 0)
+                if (cpu != 0) {
-                        cpu_msrs[cpu] = cpu_msrs[0];
+                        memcpy(cpu_msrs[cpu].counters, cpu_msrs[0].counters,
+                                sizeof(struct op_msr) * model->num_counters);
+                        memcpy(cpu_msrs[cpu].controls, cpu_msrs[0].controls,
+                                sizeof(struct op_msr) * model->num_controls);
+                }
        }
        on_each_cpu(nmi_save_registers, NULL, 0, 1);
        on_each_cpu(nmi_cpu_setup, NULL, 0, 1);
diff --git a/arch/i386/pci/fixup.c b/arch/i386/pci/fixup.c
index b62eafb997bc..b95b42950ed4 100644
--- a/arch/i386/pci/fixup.c
+++ b/arch/i386/pci/fixup.c
@@ -436,3 +436,14 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5530_LEGACY,
                        pci_early_fixup_cyrix_5530);
 DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5530_LEGACY,
                        pci_early_fixup_cyrix_5530);
+/*
+ * Siemens Nixdorf AG FSC Multiprocessor Interrupt Controller:
+ * prevent update of the BAR0, which doesn't look like a normal BAR.
+ */
+static void __devinit pci_siemens_interrupt_controller(struct pci_dev *dev)
+{
+        dev->resource[0].flags |= IORESOURCE_PCI_FIXED;
+}
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_SIEMENS, 0x0015,
+                          pci_siemens_interrupt_controller);
diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig
index b8536c7c0877..85cdd23b0447 100644
--- a/arch/m68k/Kconfig
+++ b/arch/m68k/Kconfig
@@ -355,8 +355,9 @@ config RMW_INSNS
          adventurous.
 config SINGLE_MEMORY_CHUNK
-        bool "Use one physical chunk of memory only"
+        bool "Use one physical chunk of memory only" if ADVANCED && !SUN3
-        depends on ADVANCED && !SUN3
+        default y if SUN3
+        select NEED_MULTIPLE_NODES
        help
          Ignore all but the first contiguous chunk of physical memory for VM
          purposes.  This will save a few bytes kernel size and may speed up
@@ -377,6 +378,14 @@ config 060_WRITETHROUGH
          is hardwired on.  The 53c710 SCSI driver is known to suffer from
          this problem.
+config ARCH_DISCONTIGMEM_ENABLE
+        def_bool !SINGLE_MEMORY_CHUNK
+config NODES_SHIFT
+        int
+        default "3"
+        depends on !SINGLE_MEMORY_CHUNK
 source "mm/Kconfig"
 endmenu
diff --git a/arch/m68k/Makefile b/arch/m68k/Makefile
index c20831a7e1a9..aa383a5ea7ac 100644
--- a/arch/m68k/Makefile
+++ b/arch/m68k/Makefile
@@ -19,6 +19,7 @@ COMPILE_ARCH = $(shell uname -m)
 # override top level makefile
 AS += -m68020
 LDFLAGS := -m m68kelf
+LDFLAGS_MODULE += -T $(srctree)/arch/m68k/kernel/module.lds
 ifneq ($(COMPILE_ARCH),$(ARCH))
        # prefix for cross-compiling binaries
        CROSS_COMPILE = m68k-linux-gnu-
diff --git a/arch/m68k/kernel/Makefile b/arch/m68k/kernel/Makefile
index 0b68ab8d63d1..a806208c7fb5 100644
--- a/arch/m68k/kernel/Makefile
+++ b/arch/m68k/kernel/Makefile
@@ -9,13 +9,12 @@ else
 endif
 extra-y += vmlinux.lds
-obj-y   := entry.o process.o traps.o ints.o signal.o ptrace.o \
+obj-y   := entry.o process.o traps.o ints.o signal.o ptrace.o module.o \
           sys_m68k.o time.o semaphore.o setup.o m68k_ksyms.o devres.o
 devres-y = ../../../kernel/irq/devres.o
 obj-$(CONFIG_PCI)       += bios32.o
-obj-$(CONFIG_MODULES)   += module.o
 obj-y$(CONFIG_MMU_SUN3) += dma.o        # no, it's not a typo
 EXTRA_AFLAGS := -traditional
diff --git a/arch/m68k/kernel/module.c b/arch/m68k/kernel/module.c
index 3b1a2ff61ddc..774862bc6977 100644
--- a/arch/m68k/kernel/module.c
+++ b/arch/m68k/kernel/module.c
@@ -1,3 +1,9 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file COPYING in the main directory of this archive
+ * for more details.
+ */
 #include <linux/moduleloader.h>
 #include <linux/elf.h>
 #include <linux/vmalloc.h>
@@ -11,6 +17,8 @@
 #define DEBUGP(fmt...)
 #endif
+#ifdef CONFIG_MODULES
 void *module_alloc(unsigned long size)
 {
        if (size == 0)
@@ -118,11 +126,32 @@ int apply_relocate_add(Elf32_Shdr *sechdrs,
 int module_finalize(const Elf_Ehdr *hdr,
                    const Elf_Shdr *sechdrs,
-                    struct module *me)
+                    struct module *mod)
 {
+        module_fixup(mod, mod->arch.fixup_start, mod->arch.fixup_end);
        return 0;
 }
 void module_arch_cleanup(struct module *mod)
 {
 }
+#endif /* CONFIG_MODULES */
+void module_fixup(struct module *mod, struct m68k_fixup_info *start,
+                  struct m68k_fixup_info *end)
+{
+        struct m68k_fixup_info *fixup;
+        for (fixup = start; fixup < end; fixup++) {
+                switch (fixup->type) {
+                case m68k_fixup_memoffset:
+                        *(u32 *)fixup->addr = m68k_memoffset;
+                        break;
+                case m68k_fixup_vnode_shift:
+                        *(u16 *)fixup->addr += m68k_virt_to_node_shift;
+                        break;
+                }
+        }
+}
diff --git a/arch/m68k/kernel/module.lds b/arch/m68k/kernel/module.lds
new file mode 100644
index 000000000000..fda94fa38243
--- /dev/null
+++ b/arch/m68k/kernel/module.lds
@@ -0,0 +1,7 @@
+SECTIONS {
+        .m68k_fixup : {
+                __start_fixup = .;
+                *(.m68k_fixup)
+                __stop_fixup = .;
+        }
+}
diff --git a/arch/m68k/kernel/setup.c b/arch/m68k/kernel/setup.c
index 610319356691..215c7bd43924 100644
--- a/arch/m68k/kernel/setup.c
+++ b/arch/m68k/kernel/setup.c
@@ -60,14 +60,12 @@ extern unsigned long availmem;
 int m68k_num_memory;
 int m68k_realnum_memory;
 EXPORT_SYMBOL(m68k_realnum_memory);
-#ifdef CONFIG_SINGLE_MEMORY_CHUNK
 unsigned long m68k_memoffset;
 EXPORT_SYMBOL(m68k_memoffset);
-#endif
 struct mem_info m68k_memory[NUM_MEMINFO];
 EXPORT_SYMBOL(m68k_memory);
-static struct mem_info m68k_ramdisk;
+struct mem_info m68k_ramdisk;
 static char m68k_command_line[CL_SIZE];
@@ -208,9 +206,6 @@ static void __init m68k_parse_bootinfo(const struct bi_record *record)
 void __init setup_arch(char **cmdline_p)
 {
        extern int _etext, _edata, _end;
-#ifndef CONFIG_SUN3
-        unsigned long endmem, startmem;
-#endif
        int i;
        /* The bootinfo is located right after the kernel bss */
@@ -320,30 +315,16 @@ void __init setup_arch(char **cmdline_p)
                panic("No configuration setup");
        }
-#ifndef CONFIG_SUN3
+        paging_init();
-        startmem= m68k_memory[0].addr;
-        endmem = startmem + m68k_memory[0].size;
-        high_memory = (void *)PAGE_OFFSET;
-        for (i = 0; i < m68k_num_memory; i++) {
-                m68k_memory[i].size &= MASK_256K;
-                if (m68k_memory[i].addr < startmem)
-                        startmem = m68k_memory[i].addr;
-                if (m68k_memory[i].addr+m68k_memory[i].size > endmem)
-                        endmem = m68k_memory[i].addr+m68k_memory[i].size;
-                high_memory += m68k_memory[i].size;
-        }
-        availmem += init_bootmem_node(NODE_DATA(0), availmem >> PAGE_SHIFT,
-                                      startmem >> PAGE_SHIFT, endmem >> PAGE_SHIFT);
-        for (i = 0; i < m68k_num_memory; i++)
-                free_bootmem(m68k_memory[i].addr, m68k_memory[i].size);
-        reserve_bootmem(m68k_memory[0].addr, availmem - m68k_memory[0].addr);
+#ifndef CONFIG_SUN3
+        for (i = 1; i < m68k_num_memory; i++)
+                free_bootmem_node(NODE_DATA(i), m68k_memory[i].addr,
+                                  m68k_memory[i].size);
 #ifdef CONFIG_BLK_DEV_INITRD
        if (m68k_ramdisk.size) {
-                reserve_bootmem(m68k_ramdisk.addr, m68k_ramdisk.size);
+                reserve_bootmem_node(__virt_to_node(phys_to_virt(m68k_ramdisk.addr)),
+                                     m68k_ramdisk.addr, m68k_ramdisk.size);
                initrd_start = (unsigned long)phys_to_virt(m68k_ramdisk.addr);
                initrd_end = initrd_start + m68k_ramdisk.size;
                printk("initrd: %08lx - %08lx\n", initrd_start, initrd_end);
@@ -362,8 +343,6 @@ void __init setup_arch(char **cmdline_p)
 #endif /* !CONFIG_SUN3 */
-        paging_init();
 /* set ISA defs early as possible */
 #if defined(CONFIG_ISA) && defined(MULTI_ISA)
 #if defined(CONFIG_Q40)
diff --git a/arch/m68k/kernel/vmlinux-std.lds b/arch/m68k/kernel/vmlinux-std.lds
index 78f139226a1b..40f02b128f22 100644
--- a/arch/m68k/kernel/vmlinux-std.lds
+++ b/arch/m68k/kernel/vmlinux-std.lds
@@ -60,6 +60,11 @@ SECTIONS
  __con_initcall_start = .;
  .con_initcall.init : { *(.con_initcall.init) }
  __con_initcall_end = .;
+  .m68k_fixup : {
+        __start_fixup = .;
+        *(.m68k_fixup)
+        __stop_fixup = .;
+  }
  SECURITY_INIT
 #ifdef CONFIG_BLK_DEV_INITRD
  . = ALIGN(8192);
diff --git a/arch/m68k/kernel/vmlinux-sun3.lds b/arch/m68k/kernel/vmlinux-sun3.lds
index c8999b2db23b..f06425b6d206 100644
--- a/arch/m68k/kernel/vmlinux-sun3.lds
+++ b/arch/m68k/kernel/vmlinux-sun3.lds
@@ -54,6 +54,11 @@ __init_begin = .;
        __con_initcall_start = .;
        .con_initcall.init : { *(.con_initcall.init) }
        __con_initcall_end = .;
+        .m68k_fixup : {
+                __start_fixup = .;
+                *(.m68k_fixup)
+                __stop_fixup = .;
+        }
        SECURITY_INIT
 #ifdef CONFIG_BLK_DEV_INITRD
        . = ALIGN(8192);
diff --git a/arch/m68k/mac/debug.c b/arch/m68k/mac/debug.c
index 7a5bed5bdc57..e8a57138b4a6 100644
--- a/arch/m68k/mac/debug.c
+++ b/arch/m68k/mac/debug.c
@@ -71,7 +71,7 @@ void mac_debugging_short(int pos, short num)
        /* calculate current offset */
        pengoffset = (unsigned char *)mac_videobase +
-                (150+line*2) * mac_rowbytes) + 80 * peng;
+                (150+line*2) * mac_rowbytes + 80 * peng;
        pptr = pengoffset;
diff --git a/arch/m68k/mm/init.c b/arch/m68k/mm/init.c
index ab90213e5c54..f1de19e1dde6 100644
--- a/arch/m68k/mm/init.c
+++ b/arch/m68k/mm/init.c
@@ -7,6 +7,7 @@
 *  to motorola.c and sun3mmu.c
 */
+#include <linux/module.h>
 #include <linux/signal.h>
 #include <linux/sched.h>
 #include <linux/mm.h>
@@ -31,6 +32,37 @@
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
+static bootmem_data_t __initdata bootmem_data[MAX_NUMNODES];
+pg_data_t pg_data_map[MAX_NUMNODES];
+EXPORT_SYMBOL(pg_data_map);
+int m68k_virt_to_node_shift;
+#ifndef CONFIG_SINGLE_MEMORY_CHUNK
+pg_data_t *pg_data_table[65];
+EXPORT_SYMBOL(pg_data_table);
+#endif
+void m68k_setup_node(int node)
+{
+#ifndef CONFIG_SINGLE_MEMORY_CHUNK
+        struct mem_info *info = m68k_memory + node;
+        int i, end;
+        i = (unsigned long)phys_to_virt(info->addr) >> __virt_to_node_shift();
+        end = (unsigned long)phys_to_virt(info->addr + info->size - 1) >> __virt_to_node_shift();
+        for (; i <= end; i++) {
+                if (pg_data_table[i])
+                        printk("overlap at %u for chunk %u\n", i, node);
+                pg_data_table[i] = pg_data_map + node;
+        }
+#endif
+        pg_data_map[node].bdata = bootmem_data + node;
+        node_set_online(node);
+}
 /*
 * ZERO_PAGE is a special page that is used for zero-initialized
 * data and COW.
@@ -40,52 +72,51 @@ void *empty_zero_page;
 void show_mem(void)
 {
-    unsigned long i;
+        pg_data_t *pgdat;
-    int free = 0, total = 0, reserved = 0, shared = 0;
+        int free = 0, total = 0, reserved = 0, shared = 0;
-    int cached = 0;
+        int cached = 0;
+        int i;
-    printk("\nMem-info:\n");
-    show_free_areas();
+        printk("\nMem-info:\n");
-    printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
+        show_free_areas();
-    i = max_mapnr;
+        printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
-    while (i-- > 0) {
+        for_each_online_pgdat(pgdat) {
-        total++;
+                for (i = 0; i < pgdat->node_spanned_pages; i++) {
-        if (PageReserved(mem_map+i))
+                        struct page *page = pgdat->node_mem_map + i;
-            reserved++;
+                        total++;
-        else if (PageSwapCache(mem_map+i))
+                        if (PageReserved(page))
-            cached++;
+                                reserved++;
-        else if (!page_count(mem_map+i))
+                        else if (PageSwapCache(page))
-            free++;
+                                cached++;
-        else
+                        else if (!page_count(page))
-            shared += page_count(mem_map+i) - 1;
+                                free++;
-    }
+                        else
-    printk("%d pages of RAM\n",total);
+                                shared += page_count(page) - 1;
-    printk("%d free pages\n",free);
+                }
-    printk("%d reserved pages\n",reserved);
+        }
-    printk("%d pages shared\n",shared);
+        printk("%d pages of RAM\n",total);
-    printk("%d pages swap cached\n",cached);
+        printk("%d free pages\n",free);
+        printk("%d reserved pages\n",reserved);
+        printk("%d pages shared\n",shared);
+        printk("%d pages swap cached\n",cached);
 }
 extern void init_pointer_table(unsigned long ptable);
 /* References to section boundaries */
-extern char _text, _etext, _edata, __bss_start, _end;
+extern char _text[], _etext[];
-extern char __init_begin, __init_end;
+extern char __init_begin[], __init_end[];
 extern pmd_t *zero_pgtable;
 void __init mem_init(void)
 {
+        pg_data_t *pgdat;
        int codepages = 0;
        int datapages = 0;
        int initpages = 0;
-        unsigned long tmp;
-#ifndef CONFIG_SUN3
        int i;
-#endif
-        max_mapnr = num_physpages = (((unsigned long)high_memory - PAGE_OFFSET) >> PAGE_SHIFT);
 #ifdef CONFIG_ATARI
        if (MACH_IS_ATARI)
@@ -93,19 +124,25 @@ void __init mem_init(void)
 #endif
        /* this will put all memory onto the freelists */
-        totalram_pages = free_all_bootmem();
+        totalram_pages = num_physpages = 0;
+        for_each_online_pgdat(pgdat) {
-        for (tmp = PAGE_OFFSET ; tmp < (unsigned long)high_memory; tmp += PAGE_SIZE) {
+                num_physpages += pgdat->node_present_pages;
-                if (PageReserved(virt_to_page(tmp))) {
-                        if (tmp >= (unsigned long)&_text
+                totalram_pages += free_all_bootmem_node(pgdat);
-                            && tmp < (unsigned long)&_etext)
+                for (i = 0; i < pgdat->node_spanned_pages; i++) {
+                        struct page *page = pgdat->node_mem_map + i;
+                        char *addr = page_to_virt(page);
+                        if (!PageReserved(page))
+                                continue;
+                        if (addr >= _text &&
+                            addr < _etext)
                                codepages++;
-                        else if (tmp >= (unsigned long) &__init_begin
+                        else if (addr >= __init_begin &&
-                                 && tmp < (unsigned long) &__init_end)
+                                 addr < __init_end)
                                initpages++;
                        else
                                datapages++;
-                        continue;
                }
        }
@@ -124,7 +161,7 @@ void __init mem_init(void)
        printk("Memory: %luk/%luk available (%dk kernel code, %dk data, %dk init)\n",
               (unsigned long)nr_free_pages() << (PAGE_SHIFT-10),
-               max_mapnr << (PAGE_SHIFT-10),
+               totalram_pages << (PAGE_SHIFT-10),
               codepages << (PAGE_SHIFT-10),
               datapages << (PAGE_SHIFT-10),
               initpages << (PAGE_SHIFT-10));
diff --git a/arch/m68k/mm/memory.c b/arch/m68k/mm/memory.c
index 13c0b4ad01eb..b7473525b431 100644
--- a/arch/m68k/mm/memory.c
+++ b/arch/m68k/mm/memory.c
@@ -127,67 +127,6 @@ int free_pointer_table (pmd_t *ptable)
        return 0;
 }
-#ifdef DEBUG_INVALID_PTOV
-int mm_inv_cnt = 5;
-#endif
-#ifndef CONFIG_SINGLE_MEMORY_CHUNK
-/*
- * The following two routines map from a physical address to a kernel
- * virtual address and vice versa.
- */
-unsigned long mm_vtop(unsigned long vaddr)
-{
-        int i=0;
-        unsigned long voff = (unsigned long)vaddr - PAGE_OFFSET;
-        do {
-                if (voff < m68k_memory[i].size) {
-#ifdef DEBUGPV
-                        printk ("VTOP(%p)=%lx\n", vaddr,
-                                m68k_memory[i].addr + voff);
-#endif
-                        return m68k_memory[i].addr + voff;
-                }
-                voff -= m68k_memory[i].size;
-        } while (++i < m68k_num_memory);
-        /* As a special case allow `__pa(high_memory)'.  */
-        if (voff == 0)
-                return m68k_memory[i-1].addr + m68k_memory[i-1].size;
-        return -1;
-}
-EXPORT_SYMBOL(mm_vtop);
-unsigned long mm_ptov (unsigned long paddr)
-{
-        int i = 0;
-        unsigned long poff, voff = PAGE_OFFSET;
-        do {
-                poff = paddr - m68k_memory[i].addr;
-                if (poff < m68k_memory[i].size) {
-#ifdef DEBUGPV
-                        printk ("PTOV(%lx)=%lx\n", paddr, poff + voff);
-#endif
-                        return poff + voff;
-                }
-                voff += m68k_memory[i].size;
-        } while (++i < m68k_num_memory);
-#ifdef DEBUG_INVALID_PTOV
-        if (mm_inv_cnt > 0) {
-                mm_inv_cnt--;
-                printk("Invalid use of phys_to_virt(0x%lx) at 0x%p!\n",
-                        paddr, __builtin_return_address(0));
-        }
-#endif
-        return -1;
-}
-EXPORT_SYMBOL(mm_ptov);
-#endif
 /* invalidate page in both caches */
 static inline void clear040(unsigned long paddr)
 {
@@ -354,15 +293,3 @@ void cache_push (unsigned long paddr, int len)
 }
 EXPORT_SYMBOL(cache_push);
-#ifndef CONFIG_SINGLE_MEMORY_CHUNK
-int mm_end_of_chunk (unsigned long addr, int len)
-{
-        int i;
-        for (i = 0; i < m68k_num_memory; i++)
-                if (m68k_memory[i].addr + m68k_memory[i].size == addr + len)
-                        return 1;
-        return 0;
-}
-EXPORT_SYMBOL(mm_end_of_chunk);
-#endif
diff --git a/arch/m68k/mm/motorola.c b/arch/m68k/mm/motorola.c
index afcccdc6ad45..7d571a2b44dd 100644
--- a/arch/m68k/mm/motorola.c
+++ b/arch/m68k/mm/motorola.c
@@ -43,6 +43,11 @@ unsigned long mm_cachebits;
 EXPORT_SYMBOL(mm_cachebits);
 #endif
+/* size of memory already mapped in head.S */
+#define INIT_MAPPED_SIZE        (4UL<<20)
+extern unsigned long availmem;
 static pte_t * __init kernel_page_table(void)
 {
        pte_t *ptablep;
@@ -98,19 +103,20 @@ static pmd_t * __init kernel_ptr_table(void)
        return last_pgtable;
 }
-static unsigned long __init
+static void __init map_node(int node)
-map_chunk (unsigned long addr, long size)
 {
 #define PTRTREESIZE (256*1024)
 #define ROOTTREESIZE (32*1024*1024)
-        static unsigned long virtaddr = PAGE_OFFSET;
+        unsigned long physaddr, virtaddr, size;
-        unsigned long physaddr;
        pgd_t *pgd_dir;
        pmd_t *pmd_dir;
        pte_t *pte_dir;
-        physaddr = (addr | m68k_supervisor_cachemode |
+        size = m68k_memory[node].size;
-                    _PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_DIRTY);
+        physaddr = m68k_memory[node].addr;
+        virtaddr = (unsigned long)phys_to_virt(physaddr);
+        physaddr |= m68k_supervisor_cachemode |
+                    _PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_DIRTY;
        if (CPU_IS_040_OR_060)
                physaddr |= _PAGE_GLOBAL040;
@@ -190,8 +196,6 @@ map_chunk (unsigned long addr, long size)
 #ifdef DEBUG
        printk("\n");
 #endif
-        return virtaddr;
 }
 /*
@@ -200,15 +204,16 @@ map_chunk (unsigned long addr, long size)
 */
 void __init paging_init(void)
 {
-        int chunk;
-        unsigned long mem_avail = 0;
        unsigned long zones_size[MAX_NR_ZONES] = { 0, };
+        unsigned long min_addr, max_addr;
+        unsigned long addr, size, end;
+        int i;
 #ifdef DEBUG
        {
                extern unsigned long availmem;
-                printk ("start of paging_init (%p, %lx, %lx, %lx)\n",
+                printk ("start of paging_init (%p, %lx)\n",
-                        kernel_pg_dir, availmem, start_mem, end_mem);
+                        kernel_pg_dir, availmem);
        }
 #endif
@@ -222,24 +227,62 @@ void __init paging_init(void)
                        pgprot_val(protection_map[i]) |= _PAGE_CACHE040;
        }
+        min_addr = m68k_memory[0].addr;
+        max_addr = min_addr + m68k_memory[0].size;
+        for (i = 1; i < m68k_num_memory;) {
+                if (m68k_memory[i].addr < min_addr) {
+                        printk("Ignoring memory chunk at 0x%lx:0x%lx before the first chunk\n",
+                                m68k_memory[i].addr, m68k_memory[i].size);
+                        printk("Fix your bootloader or use a memfile to make use of this area!\n");
+                        m68k_num_memory--;
+                        memmove(m68k_memory + i, m68k_memory + i + 1,
+                                (m68k_num_memory - i) * sizeof(struct mem_info));
+                        continue;
+                }
+                addr = m68k_memory[i].addr + m68k_memory[i].size;
+                if (addr > max_addr)
+                        max_addr = addr;
+                i++;
+        }
+        m68k_memoffset = min_addr - PAGE_OFFSET;
+        m68k_virt_to_node_shift = fls(max_addr - min_addr - 1) - 6;
+        module_fixup(NULL, __start_fixup, __stop_fixup);
+        flush_icache();
+        high_memory = phys_to_virt(max_addr);
+        min_low_pfn = availmem >> PAGE_SHIFT;
+        max_low_pfn = max_addr >> PAGE_SHIFT;
+        for (i = 0; i < m68k_num_memory; i++) {
+                addr = m68k_memory[i].addr;
+                end = addr + m68k_memory[i].size;
+                m68k_setup_node(i);
+                availmem = PAGE_ALIGN(availmem);
+                availmem += init_bootmem_node(NODE_DATA(i),
+                                              availmem >> PAGE_SHIFT,
+                                              addr >> PAGE_SHIFT,
+                                              end >> PAGE_SHIFT);
+        }
        /*
         * Map the physical memory available into the kernel virtual
-         * address space.  It may allocate some memory for page
+         * address space. First initialize the bootmem allocator with
-         * tables and thus modify availmem.
+         * the memory we already mapped, so map_node() has something
+         * to allocate.
         */
+        addr = m68k_memory[0].addr;
+        size = m68k_memory[0].size;
+        free_bootmem_node(NODE_DATA(0), availmem, min(INIT_MAPPED_SIZE, size) - (availmem - addr));
+        map_node(0);
+        if (size > INIT_MAPPED_SIZE)
+                free_bootmem_node(NODE_DATA(0), addr + INIT_MAPPED_SIZE, size - INIT_MAPPED_SIZE);
-        for (chunk = 0; chunk < m68k_num_memory; chunk++) {
+        for (i = 1; i < m68k_num_memory; i++)
-                mem_avail = map_chunk (m68k_memory[chunk].addr,
+                map_node(i);
-                                       m68k_memory[chunk].size);
-        }
        flush_tlb_all();
-#ifdef DEBUG
-        printk ("memory available is %ldKB\n", mem_avail >> 10);
-        printk ("start_mem is %#lx\nvirtual_end is %#lx\n",
-                start_mem, end_mem);
-#endif
        /*
         * initialize the bad page table and bad page to point
@@ -256,14 +299,11 @@ void __init paging_init(void)
 #ifdef DEBUG
        printk ("before free_area_init\n");
 #endif
-        zones_size[ZONE_DMA] = (mach_max_dma_address < (unsigned long)high_memory ?
+        for (i = 0; i < m68k_num_memory; i++) {
-                                (mach_max_dma_address+1) : (unsigned long)high_memory);
+                zones_size[ZONE_DMA] = m68k_memory[i].size >> PAGE_SHIFT;
-        zones_size[ZONE_NORMAL] = (unsigned long)high_memory - zones_size[0];
+                free_area_init_node(i, pg_data_map + i, zones_size,
+                                    m68k_memory[i].addr >> PAGE_SHIFT, NULL);
-        zones_size[ZONE_DMA] = (zones_size[ZONE_DMA] - PAGE_OFFSET) >> PAGE_SHIFT;
+        }
-        zones_size[ZONE_NORMAL] >>= PAGE_SHIFT;
-        free_area_init(zones_size);
 }
 extern char __init_begin, __init_end;
diff --git a/arch/m68k/sun3/config.c b/arch/m68k/sun3/config.c
index 4851b8437a87..c0fbd278fbb1 100644
--- a/arch/m68k/sun3/config.c
+++ b/arch/m68k/sun3/config.c
@@ -21,6 +21,7 @@
 #include <asm/contregs.h>
 #include <asm/movs.h>
 #include <asm/pgtable.h>
+#include <asm/pgalloc.h>
 #include <asm/sun3-head.h>
 #include <asm/sun3mmu.h>
 #include <asm/rtc.h>
@@ -127,6 +128,7 @@ void __init sun3_bootmem_alloc(unsigned long memory_start, unsigned long memory_
        high_memory = (void *)memory_end;
        availmem = memory_start;
+        m68k_setup_node(0);
        availmem += init_bootmem_node(NODE_DATA(0), start_page, 0, num_pages);
        availmem = (availmem + (PAGE_SIZE-1)) & PAGE_MASK;
diff --git a/arch/mips/jmr3927/rbhma3100/kgdb_io.c b/arch/mips/jmr3927/rbhma3100/kgdb_io.c
index 2604f2c9a96e..342579cfdc01 100644
--- a/arch/mips/jmr3927/rbhma3100/kgdb_io.c
+++ b/arch/mips/jmr3927/rbhma3100/kgdb_io.c
@@ -36,7 +36,7 @@
 #define TIMEOUT       0xffffff
 static int remoteDebugInitialized = 0;
-static void debugInit(int baud)
+static void debugInit(int baud);
 int putDebugChar(unsigned char c)
 {
diff --git a/arch/mips/pci/pci-ocelot.c b/arch/mips/pci/pci-ocelot.c
index 7f94f26d35ae..1421d34535ef 100644
--- a/arch/mips/pci/pci-ocelot.c
+++ b/arch/mips/pci/pci-ocelot.c
@@ -71,19 +71,19 @@ static inline void pci0WriteConfigReg(unsigned int offset, unsigned int data)
 }
 static struct resource ocelot_mem_resource = {
-        start   = GT_PCI_MEM_BASE;
+        .start  = GT_PCI_MEM_BASE,
-        end     = GT_PCI_MEM_BASE + GT_PCI_MEM_BASE - 1;
+        .end    = GT_PCI_MEM_BASE + GT_PCI_MEM_BASE - 1,
 };
 static struct resource ocelot_io_resource = {
-        start   = GT_PCI_IO_BASE;
+        .start  = GT_PCI_IO_BASE,
-        end     = GT_PCI_IO_BASE + GT_PCI_IO_SIZE - 1;
+        .end    = GT_PCI_IO_BASE + GT_PCI_IO_SIZE - 1,
 };
 static struct pci_controller ocelot_pci_controller = {
-        .pci_ops        = gt64xxx_pci0_ops;
+        .pci_ops        = gt64xxx_pci0_ops,
-        .mem_resource   = &ocelot_mem_resource;
+        .mem_resource   = &ocelot_mem_resource,
-        .io_resource    = &ocelot_io_resource;
+        .io_resource    = &ocelot_io_resource,
 };
 static int __init ocelot_pcibios_init(void)
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 56d3c0dcd2b8..5eaeafd30bdf 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -118,6 +118,7 @@ config GENERIC_BUG
        depends on BUG
 config SYS_SUPPORTS_APM_EMULATION
+        default y if PMAC_APM_EMU
        bool
 config DEFAULT_UIMAGE
diff --git a/arch/powerpc/boot/dts/lite5200.dts b/arch/powerpc/boot/dts/lite5200.dts
index eae68ab1177f..d29308fe4c24 100644
--- a/arch/powerpc/boot/dts/lite5200.dts
+++ b/arch/powerpc/boot/dts/lite5200.dts
@@ -67,7 +67,7 @@
                        interrupt-controller;
                        #interrupt-cells = <3>;
                        device_type = "interrupt-controller";
-                        compatible = "mpc5200_pic";
+                        compatible = "mpc5200-pic";
                        reg = <500 80>;
                        built-in;
                };
diff --git a/arch/powerpc/boot/dts/lite5200b.dts b/arch/powerpc/boot/dts/lite5200b.dts
index 5185625a9419..f242531f0451 100644
--- a/arch/powerpc/boot/dts/lite5200b.dts
+++ b/arch/powerpc/boot/dts/lite5200b.dts
@@ -67,7 +67,7 @@
                        interrupt-controller;
                        #interrupt-cells = <3>;
                        device_type = "interrupt-controller";
-                        compatible = "mpc5200b-pic\0mpc5200_pic";
+                        compatible = "mpc5200b-pic\0mpc5200-pic";
                        reg = <500 80>;
                        built-in;
                };
diff --git a/arch/powerpc/mm/pgtable_32.c b/arch/powerpc/mm/pgtable_32.c
index d8232b7a08f7..f6ae1a57d652 100644
--- a/arch/powerpc/mm/pgtable_32.c
+++ b/arch/powerpc/mm/pgtable_32.c
@@ -93,7 +93,7 @@ void pgd_free(pgd_t *pgd)
        free_pages((unsigned long)pgd, PGDIR_ORDER);
 }
-pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address)
+__init_refok pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address)
 {
        pte_t *pte;
        extern int mem_init_done;
diff --git a/arch/powerpc/platforms/chrp/pegasos_eth.c b/arch/powerpc/platforms/chrp/pegasos_eth.c
index 71045677559a..5bcc58d9a4dd 100644
--- a/arch/powerpc/platforms/chrp/pegasos_eth.c
+++ b/arch/powerpc/platforms/chrp/pegasos_eth.c
@@ -169,7 +169,7 @@ static int Enable_SRAM(void)
 /***********/
 /***********/
-int mv643xx_eth_add_pds(void)
+static int __init mv643xx_eth_add_pds(void)
 {
        int ret = 0;
        static struct pci_device_id pci_marvell_mv64360[] = {
diff --git a/arch/powerpc/platforms/powermac/setup.c b/arch/powerpc/platforms/powermac/setup.c
index a410bc76a8a8..07b1c4ec428d 100644
--- a/arch/powerpc/platforms/powermac/setup.c
+++ b/arch/powerpc/platforms/powermac/setup.c
@@ -384,7 +384,7 @@ int boot_part;
 static dev_t boot_dev;
 #ifdef CONFIG_SCSI
-void __init note_scsi_host(struct device_node *node, void *host)
+void note_scsi_host(struct device_node *node, void *host)
 {
        int l;
        char *p;
diff --git a/arch/powerpc/sysdev/qe_lib/Kconfig b/arch/powerpc/sysdev/qe_lib/Kconfig
index 887739f3badc..f611d344a126 100644
--- a/arch/powerpc/sysdev/qe_lib/Kconfig
+++ b/arch/powerpc/sysdev/qe_lib/Kconfig
@@ -5,15 +5,13 @@
 config UCC_SLOW
        bool
        default n
-        select UCC
        help
          This option provides qe_lib support to UCC slow
          protocols: UART, BISYNC, QMC
 config UCC_FAST
        bool
-        default n
+        default y if UCC_GETH
-        select UCC
        help
          This option provides qe_lib support to UCC fast
          protocols: HDLC, Ethernet, ATM, transparent
diff --git a/arch/ppc/kernel/entry.S b/arch/ppc/kernel/entry.S
index ab64256110bd..fba7ca17a67e 100644
--- a/arch/ppc/kernel/entry.S
+++ b/arch/ppc/kernel/entry.S
@@ -596,7 +596,11 @@ fast_exception_return:
        mr      r12,r4          /* restart at exc_exit_restart */
        b       2b
-        .comm   fee_restarts,4
+        .section .bss
+        .align  2
+fee_restarts:
+        .space  4
+        .previous
 /* aargh, a nonrecoverable interrupt, panic */
 /* aargh, we don't know which trap this is */
@@ -851,7 +855,11 @@ load_dbcr0:
        mtspr   SPRN_DBSR,r11   /* clear all pending debug events */
        blr
-        .comm   global_dbcr0,8
+        .section .bss
+        .align  4
+global_dbcr0:
+        .space  8
+        .previous
 #endif /* !(CONFIG_4xx || CONFIG_BOOKE) */
 do_work:                        /* r10 contains MSR_KERNEL here */
@@ -926,4 +934,8 @@ END_FTR_SECTION_IFSET(CPU_FTR_601)
        /* shouldn't return */
        b       4b
-        .comm   ee_restarts,4
+        .section .bss
+        .align  2
+ee_restarts:
+        .space  4
+        .previous
diff --git a/arch/ppc/kernel/ppc_ksyms.c b/arch/ppc/kernel/ppc_ksyms.c
index 4ad499605d05..a4165209ac7c 100644
--- a/arch/ppc/kernel/ppc_ksyms.c
+++ b/arch/ppc/kernel/ppc_ksyms.c
@@ -40,7 +40,6 @@
 #include <asm/time.h>
 #include <asm/cputable.h>
 #include <asm/btext.h>
-#include <asm/div64.h>
 #include <asm/xmon.h>
 #include <asm/signal.h>
 #include <asm/dcr.h>
@@ -93,7 +92,6 @@ EXPORT_SYMBOL(strncpy);
 EXPORT_SYMBOL(strcat);
 EXPORT_SYMBOL(strlen);
 EXPORT_SYMBOL(strcmp);
-EXPORT_SYMBOL(__div64_32);
 EXPORT_SYMBOL(csum_partial);
 EXPORT_SYMBOL(csum_partial_copy_generic);
diff --git a/arch/ppc/mm/hashtable.S b/arch/ppc/mm/hashtable.S
index e756942e65c4..5f364dc50154 100644
--- a/arch/ppc/mm/hashtable.S
+++ b/arch/ppc/mm/hashtable.S
@@ -30,7 +30,11 @@
 #include <asm/asm-offsets.h>
 #ifdef CONFIG_SMP
-        .comm   mmu_hash_lock,4
+        .section .bss
+        .align  2
+        .globl mmu_hash_lock
+mmu_hash_lock:
+        .space  4
 #endif /* CONFIG_SMP */
 /*
@@ -461,9 +465,17 @@ found_slot:
        sync            /* make sure pte updates get to memory */
        blr
-        .comm   next_slot,4
+        .section .bss
-        .comm   primary_pteg_full,4
+        .align  2
-        .comm   htab_hash_searches,4
+next_slot:
+        .space  4
+        .globl primary_pteg_full
+primary_pteg_full:
+        .space  4
+        .globl htab_hash_searches
+htab_hash_searches:
+        .space  4
+        .previous
 /*
 * Flush the entry for a particular page from the hash table.
diff --git a/arch/ppc/mm/pgtable.c b/arch/ppc/mm/pgtable.c
index c023b7298809..35ebb6395ae3 100644
--- a/arch/ppc/mm/pgtable.c
+++ b/arch/ppc/mm/pgtable.c
@@ -92,7 +92,7 @@ void pgd_free(pgd_t *pgd)
        free_pages((unsigned long)pgd, PGDIR_ORDER);
 }
-pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address)
+__init_refok pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address)
 {
        pte_t *pte;
        extern int mem_init_done;
diff --git a/arch/s390/hypfs/hypfs_diag.c b/arch/s390/hypfs/hypfs_diag.c
index 2782cf9da5b4..b9a1ce1f28e4 100644
--- a/arch/s390/hypfs/hypfs_diag.c
+++ b/arch/s390/hypfs/hypfs_diag.c
@@ -481,9 +481,17 @@ out:
 /* Diagnose 224 functions */
-static void diag224(void *ptr)
+static int diag224(void *ptr)
 {
-        asm volatile("diag %0,%1,0x224" : :"d" (0), "d"(ptr) : "memory");
+        int rc = -ENOTSUPP;
+        asm volatile(
+                "       diag    %1,%2,0x224\n"
+                "0:     lhi     %0,0x0\n"
+                "1:\n"
+                EX_TABLE(0b,1b)
+                : "+d" (rc) :"d" (0), "d" (ptr) : "memory");
+        return rc;
 }
 static int diag224_get_name_table(void)
@@ -492,7 +500,10 @@ static int diag224_get_name_table(void)
        diag224_cpu_names = kmalloc(PAGE_SIZE, GFP_KERNEL | GFP_DMA);
        if (!diag224_cpu_names)
                return -ENOMEM;
-        diag224(diag224_cpu_names);
+        if (diag224(diag224_cpu_names)) {
+                kfree(diag224_cpu_names);
+                return -ENOTSUPP;
+        }
        EBCASC(diag224_cpu_names + 16, (*diag224_cpu_names + 1) * 16);
        return 0;
 }
diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c
index dca6eaf82c80..1b2f5ce45320 100644
--- a/arch/s390/kernel/debug.c
+++ b/arch/s390/kernel/debug.c
@@ -163,7 +163,7 @@ unsigned int debug_feature_version = __DEBUG_FEATURE_VERSION;
 static debug_info_t *debug_area_first = NULL;
 static debug_info_t *debug_area_last = NULL;
-static DECLARE_MUTEX(debug_lock);
+static DEFINE_MUTEX(debug_mutex);
 static int initialized;
@@ -576,7 +576,7 @@ debug_input(struct file *file, const char __user *user_buf, size_t length,
        int rc = 0;
        file_private_info_t *p_info;
-        down(&debug_lock);
+        mutex_lock(&debug_mutex);
        p_info = ((file_private_info_t *) file->private_data);
        if (p_info->view->input_proc)
                rc = p_info->view->input_proc(p_info->debug_info_org,
@@ -584,7 +584,7 @@ debug_input(struct file *file, const char __user *user_buf, size_t length,
                                              length, offset);
        else
                rc = -EPERM;
-        up(&debug_lock);
+        mutex_unlock(&debug_mutex);
        return rc;              /* number of input characters */
 }
@@ -602,7 +602,7 @@ debug_open(struct inode *inode, struct file *file)
        file_private_info_t *p_info;
        debug_info_t *debug_info, *debug_info_snapshot;
-        down(&debug_lock);
+        mutex_lock(&debug_mutex);
        debug_info = file->f_path.dentry->d_inode->i_private;
        /* find debug view */
        for (i = 0; i < DEBUG_MAX_VIEWS; i++) {
@@ -653,7 +653,7 @@ found:
        file->private_data = p_info;
        debug_info_get(debug_info);
 out:
-        up(&debug_lock);
+        mutex_unlock(&debug_mutex);
        return rc;
 }
@@ -688,7 +688,7 @@ debug_register (char *name, int pages_per_area, int nr_areas, int buf_size)
        if (!initialized)
                BUG();
-        down(&debug_lock);
+        mutex_lock(&debug_mutex);
        /* create new debug_info */
@@ -702,7 +702,7 @@ out:
        if (!rc){
                printk(KERN_ERR "debug: debug_register failed for %s\n",name);
        }
-        up(&debug_lock);
+        mutex_unlock(&debug_mutex);
        return rc;
 }
@@ -716,9 +716,9 @@ debug_unregister(debug_info_t * id)
 {
        if (!id)
                goto out;
-        down(&debug_lock);
+        mutex_lock(&debug_mutex);
        debug_info_put(id);
-        up(&debug_lock);
+        mutex_unlock(&debug_mutex);
 out:
        return;
@@ -1054,11 +1054,11 @@ __init debug_init(void)
        int rc = 0;
        s390dbf_sysctl_header = register_sysctl_table(s390dbf_dir_table);
-        down(&debug_lock);
+        mutex_lock(&debug_mutex);
        debug_debugfs_root_entry = debugfs_create_dir(DEBUG_DIR_ROOT,NULL);
        printk(KERN_INFO "debug: Initialization complete\n");
        initialized = 1;
-        up(&debug_lock);
+        mutex_unlock(&debug_mutex);
        return rc;
 }
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 6bfb0889eb10..51d6309e7f3b 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -102,7 +102,7 @@ static struct resource data_resource = {
 /*
 * cpu_init() initializes state that is per-CPU.
 */
-void __devinit cpu_init (void)
+void __cpuinit cpu_init(void)
 {
        int addr = hard_smp_processor_id();
@@ -915,7 +915,7 @@ setup_arch(char **cmdline_p)
        setup_zfcpdump(console_devno);
 }
-void print_cpu_info(struct cpuinfo_S390 *cpuinfo)
+void __cpuinit print_cpu_info(struct cpuinfo_S390 *cpuinfo)
 {
   printk("cpu %d "
 #ifdef CONFIG_SMP
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 09f028a3266b..8ff2feaf9b00 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -492,7 +492,7 @@ static unsigned int __init smp_count_cpus(void)
 /*
 *      Activate a secondary processor.
 */
-int __devinit start_secondary(void *cpuvoid)
+int __cpuinit start_secondary(void *cpuvoid)
 {
        /* Setup the cpu */
        cpu_init();
@@ -741,7 +741,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
                        smp_create_idle(cpu);
 }
-void __devinit smp_prepare_boot_cpu(void)
+void __init smp_prepare_boot_cpu(void)
 {
        BUG_ON(smp_processor_id() != 0);
@@ -750,7 +750,7 @@ void __devinit smp_prepare_boot_cpu(void)
        current_set[0] = current;
 }
-void smp_cpus_done(unsigned int max_cpus)
+void __init smp_cpus_done(unsigned int max_cpus)
 {
        cpu_present_map = cpu_possible_map;
 }
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 7b1122417050..883b03b040c4 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -39,7 +39,7 @@ cflags-$(CONFIG_CPU_SH2A)		:= -m2a $(call cc-option,-m2a-nofpu,)
 cflags-$(CONFIG_CPU_SH3)                := -m3
 cflags-$(CONFIG_CPU_SH4)                := -m4 \
        $(call cc-option,-mno-implicit-fp,-m4-nofpu)
-cflags-$(CONFIG_CPU_SH4A)               := -m4a $(call cc-option,-m4a-nofpu,)
+cflags-$(CONFIG_CPU_SH4A)               := $(call cc-option,-m4a,) $(call cc-option,-m4a-nofpu,)
 cflags-$(CONFIG_CPU_BIG_ENDIAN)         += -mb
 cflags-$(CONFIG_CPU_LITTLE_ENDIAN)      += -ml
diff --git a/arch/sh/drivers/dma/dma-api.c b/arch/sh/drivers/dma/dma-api.c
index 8057a27a1bc6..cf8e11994330 100644
--- a/arch/sh/drivers/dma/dma-api.c
+++ b/arch/sh/drivers/dma/dma-api.c
@@ -16,6 +16,7 @@
 #include <linux/list.h>
 #include <linux/platform_device.h>
 #include <linux/mm.h>
+#include <linux/sched.h>
 #include <asm/dma.h>
 DEFINE_SPINLOCK(dma_spin_lock);
diff --git a/arch/sh/kernel/cf-enabler.c b/arch/sh/kernel/cf-enabler.c
index 849a9e191391..ebc73b85094a 100644
--- a/arch/sh/kernel/cf-enabler.c
+++ b/arch/sh/kernel/cf-enabler.c
@@ -12,6 +12,7 @@
 #include <linux/init.h>
 #include <linux/mm.h>
 #include <linux/vmalloc.h>
+#include <linux/interrupt.h>
 #include <asm/io.h>
 #include <asm/irq.h>
@@ -149,6 +150,11 @@ static int __init cf_init_se(void)
        ctrl_outb(0x42, PA_MRSHPC_MW2 + 0x200);
        return 0;
 }
+#else
+static int __init cf_init_se(void)
+{
+        return -1;
+}
 #endif
 int __init cf_init(void)
diff --git a/arch/sh/kernel/cpu/sh3/entry.S b/arch/sh/kernel/cpu/sh3/entry.S
index 832c0b4a1e6c..659cc081e5e7 100644
--- a/arch/sh/kernel/cpu/sh3/entry.S
+++ b/arch/sh/kernel/cpu/sh3/entry.S
@@ -320,6 +320,7 @@ skip_restore:
        .align  2
 5:      .long   0x00001000      ! DSP
+6:      .long   in_nmi
 7:      .long   0x30000000
 ! common exception handler
diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index 8cd04904c77a..fab2eb07196b 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -12,6 +12,7 @@
 */
 #include <linux/init.h>
 #include <linux/io.h>
+#include <linux/smp.h>
 #include <asm/processor.h>
 #include <asm/cache.h>
diff --git a/arch/sh/kernel/smp.c b/arch/sh/kernel/smp.c
index dbebaddcfe39..283e1425ced5 100644
--- a/arch/sh/kernel/smp.c
+++ b/arch/sh/kernel/smp.c
@@ -10,6 +10,8 @@
 * Free Software Foundation; either version 2 of the License, or (at your
 * option) any later version.
 */
+#include <linux/err.h>
 #include <linux/cache.h>
 #include <linux/cpumask.h>
 #include <linux/delay.h>
diff --git a/arch/sh/kernel/timers/timer.c b/arch/sh/kernel/timers/timer.c
index a6bcc913d25e..4e7e747d1b69 100644
--- a/arch/sh/kernel/timers/timer.c
+++ b/arch/sh/kernel/timers/timer.c
@@ -13,7 +13,7 @@
 #include <linux/string.h>
 #include <asm/timer.h>
-static struct sys_timer *sys_timers[] __initdata = {
+static struct sys_timer *sys_timers[] = {
 #ifdef CONFIG_SH_TMU
        &tmu_timer,
 #endif
@@ -26,7 +26,7 @@ static struct sys_timer *sys_timers[] __initdata = {
        NULL,
 };
-static char timer_override[10] __initdata;
+static char timer_override[10];
 static int __init timer_setup(char *str)
 {
        if (str)
@@ -53,4 +53,3 @@ struct sys_timer *get_sys_timer(void)
        return NULL;
 }
diff --git a/arch/sh/kernel/vsyscall/vsyscall.c b/arch/sh/kernel/vsyscall/vsyscall.c
index e146bafcd14f..2aa9438361bc 100644
--- a/arch/sh/kernel/vsyscall/vsyscall.c
+++ b/arch/sh/kernel/vsyscall/vsyscall.c
@@ -17,6 +17,7 @@
 #include <linux/gfp.h>
 #include <linux/module.h>
 #include <linux/elf.h>
+#include <linux/sched.h>
 /*
 * Should the kernel map a VDSO page into processes and pass its
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index bd992c0048f0..fbcc00c6c06e 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -178,6 +178,13 @@ config ARCH_HAS_ILOG2_U64
        bool
        default n
+config EMULATED_CMPXCHG
+        bool
+        default y
+        help
+          Sparc32 does not have a CAS instruction like sparc64. cmpxchg()
+          is emulated, and therefore it is not completely atomic.
 config SUN_PM
        bool
        default y
diff --git a/arch/sparc/kernel/time.c b/arch/sparc/kernel/time.c
index f1401b57ccc7..7b4612da74a6 100644
--- a/arch/sparc/kernel/time.c
+++ b/arch/sparc/kernel/time.c
@@ -148,7 +148,7 @@ irqreturn_t timer_interrupt(int irq, void *dev_id)
 }
 /* Kick start a stopped clock (procedure from the Sun NVRAM/hostid FAQ). */
-static void __init kick_start_clock(void)
+static void __devinit kick_start_clock(void)
 {
        struct mostek48t02 *regs = (struct mostek48t02 *)mstk48t02_regs;
        unsigned char sec;
@@ -223,7 +223,7 @@ static __inline__ int has_low_battery(void)
        return (data1 == data2);        /* Was the write blocked? */
 }
-static void __init mostek_set_system_time(void)
+static void __devinit mostek_set_system_time(void)
 {
        unsigned int year, mon, day, hour, min, sec;
        struct mostek48t02 *mregs;
diff --git a/arch/sparc/lib/atomic32.c b/arch/sparc/lib/atomic32.c
index 559335f4917d..cbddeb38ffda 100644
--- a/arch/sparc/lib/atomic32.c
+++ b/arch/sparc/lib/atomic32.c
@@ -2,6 +2,7 @@
 * atomic32.c: 32-bit atomic_t implementation
 *
 * Copyright (C) 2004 Keith M Wesolowski
+ * Copyright (C) 2007 Kyle McMartin
 * 
 * Based on asm-parisc/atomic.h Copyright (C) 2000 Philipp Rumpf
 */
@@ -117,3 +118,17 @@ unsigned long ___change_bit(unsigned long *addr, unsigned long mask)
        return old & mask;
 }
 EXPORT_SYMBOL(___change_bit);
+unsigned long __cmpxchg_u32(volatile u32 *ptr, u32 old, u32 new)
+{
+        unsigned long flags;
+        u32 prev;
+        spin_lock_irqsave(ATOMIC_HASH(ptr), flags);
+        if ((prev = *ptr) == old)
+                *ptr = new;
+        spin_unlock_irqrestore(ATOMIC_HASH(ptr), flags);
+        return (unsigned long)prev;
+}
+EXPORT_SYMBOL(__cmpxchg_u32);
diff --git a/arch/sparc64/Kconfig b/arch/sparc64/Kconfig
index 831781cab271..bd00f89eed1e 100644
--- a/arch/sparc64/Kconfig
+++ b/arch/sparc64/Kconfig
@@ -147,10 +147,10 @@ config SMP
          If you don't know what to do here, say N.
 config NR_CPUS
-        int "Maximum number of CPUs (2-64)"
+        int "Maximum number of CPUs (2-1024)"
-        range 2 64
+        range 2 1024
        depends on SMP
-        default "32"
+        default "64"
 source "drivers/cpufreq/Kconfig"
diff --git a/arch/sparc64/kernel/Makefile b/arch/sparc64/kernel/Makefile
index c749dccacc32..d8d19093d12f 100644
--- a/arch/sparc64/kernel/Makefile
+++ b/arch/sparc64/kernel/Makefile
@@ -8,11 +8,11 @@ EXTRA_CFLAGS := -Werror
 extra-y         := head.o init_task.o vmlinux.lds
 obj-y           := process.o setup.o cpu.o idprom.o \
-                   traps.o devices.o auxio.o una_asm.o \
+                   traps.o auxio.o una_asm.o \
                   irq.o ptrace.o time.o sys_sparc.o signal.o \
                   unaligned.o central.o pci.o starfire.o semaphore.o \
                   power.o sbus.o iommu_common.o sparc64_ksyms.o chmc.o \
-                   visemul.o prom.o of_device.o hvapi.o
+                   visemul.o prom.o of_device.o hvapi.o sstate.o mdesc.o
 obj-$(CONFIG_STACKTRACE) += stacktrace.o
 obj-$(CONFIG_PCI)        += ebus.o isa.o pci_common.o pci_iommu.o \
diff --git a/arch/sparc64/kernel/devices.c b/arch/sparc64/kernel/devices.c
deleted file mode 100644
index 0e03c8e218cd..000000000000
--- a/arch/sparc64/kernel/devices.c
+++ /dev/null
@@ -1,196 +0,0 @@
-/* devices.c: Initial scan of the prom device tree for important
- *            Sparc device nodes which we need to find.
- *
- * Copyright (C) 1996 David S. Miller (davem@caip.rutgers.edu)
- */
-#include <linux/kernel.h>
-#include <linux/threads.h>
-#include <linux/init.h>
-#include <linux/ioport.h>
-#include <linux/string.h>
-#include <linux/spinlock.h>
-#include <linux/errno.h>
-#include <linux/bootmem.h>
-#include <asm/page.h>
-#include <asm/oplib.h>
-#include <asm/system.h>
-#include <asm/smp.h>
-#include <asm/spitfire.h>
-#include <asm/timer.h>
-#include <asm/cpudata.h>
-/* Used to synchronize accesses to NatSemi SUPER I/O chip configure
- * operations in asm/ns87303.h
- */
-DEFINE_SPINLOCK(ns87303_lock);
-extern void cpu_probe(void);
-extern void central_probe(void);
-static const char *cpu_mid_prop(void)
-{
-        if (tlb_type == spitfire)
-                return "upa-portid";
-        return "portid";
-}
-static int get_cpu_mid(struct device_node *dp)
-{
-        struct property *prop;
-        if (tlb_type == hypervisor) {
-                struct linux_prom64_registers *reg;
-                int len;
-                prop = of_find_property(dp, "cpuid", &len);
-                if (prop && len == 4)
-                        return *(int *) prop->value;
-                prop = of_find_property(dp, "reg", NULL);
-                reg = prop->value;
-                return (reg[0].phys_addr >> 32) & 0x0fffffffUL;
-        } else {
-                const char *prop_name = cpu_mid_prop();
-                prop = of_find_property(dp, prop_name, NULL);
-                if (prop)
-                        return *(int *) prop->value;
-                return 0;
-        }
-}
-static int check_cpu_node(struct device_node *dp, int *cur_inst,
-                          int (*compare)(struct device_node *, int, void *),
-                          void *compare_arg,
-                          struct device_node **dev_node, int *mid)
-{
-        if (!compare(dp, *cur_inst, compare_arg)) {
-                if (dev_node)
-                        *dev_node = dp;
-                if (mid)
-                        *mid = get_cpu_mid(dp);
-                return 0;
-        }
-        (*cur_inst)++;
-        return -ENODEV;
-}
-static int __cpu_find_by(int (*compare)(struct device_node *, int, void *),
-                         void *compare_arg,
-                         struct device_node **dev_node, int *mid)
-{
-        struct device_node *dp;
-        int cur_inst;
-        cur_inst = 0;
-        for_each_node_by_type(dp, "cpu") {
-                int err = check_cpu_node(dp, &cur_inst,
-                                         compare, compare_arg,
-                                         dev_node, mid);
-                if (err == 0)
-                        return 0;
-        }
-        return -ENODEV;
-}
-static int cpu_instance_compare(struct device_node *dp, int instance, void *_arg)
-{
-        int desired_instance = (int) (long) _arg;
-        if (instance == desired_instance)
-                return 0;
-        return -ENODEV;
-}
-int cpu_find_by_instance(int instance, struct device_node **dev_node, int *mid)
-{
-        return __cpu_find_by(cpu_instance_compare, (void *)(long)instance,
-                             dev_node, mid);
-}
-static int cpu_mid_compare(struct device_node *dp, int instance, void *_arg)
-{
-        int desired_mid = (int) (long) _arg;
-        int this_mid;
-        this_mid = get_cpu_mid(dp);
-        if (this_mid == desired_mid)
-                return 0;
-        return -ENODEV;
-}
-int cpu_find_by_mid(int mid, struct device_node **dev_node)
-{
-        return __cpu_find_by(cpu_mid_compare, (void *)(long)mid,
-                             dev_node, NULL);
-}
-void __init device_scan(void)
-{
-        /* FIX ME FAST... -DaveM */
-        ioport_resource.end = 0xffffffffffffffffUL;
-        prom_printf("Booting Linux...\n");
-#ifndef CONFIG_SMP
-        {
-                struct device_node *dp;
-                int err, def;
-                err = cpu_find_by_instance(0, &dp, NULL);
-                if (err) {
-                        prom_printf("No cpu nodes, cannot continue\n");
-                        prom_halt();
-                }
-                cpu_data(0).clock_tick =
-                        of_getintprop_default(dp, "clock-frequency", 0);
-                def = ((tlb_type == hypervisor) ?
-                       (8 * 1024) :
-                       (16 * 1024));
-                cpu_data(0).dcache_size = of_getintprop_default(dp,
-                                                                "dcache-size",
-                                                                def);
-                def = 32;
-                cpu_data(0).dcache_line_size =
-                        of_getintprop_default(dp, "dcache-line-size", def);
-                def = 16 * 1024;
-                cpu_data(0).icache_size = of_getintprop_default(dp,
-                                                                "icache-size",
-                                                                def);
-                def = 32;
-                cpu_data(0).icache_line_size =
-                        of_getintprop_default(dp, "icache-line-size", def);
-                def = ((tlb_type == hypervisor) ?
-                       (3 * 1024 * 1024) :
-                       (4 * 1024 * 1024));
-                cpu_data(0).ecache_size = of_getintprop_default(dp,
-                                                                "ecache-size",
-                                                                def);
-                def = 64;
-                cpu_data(0).ecache_line_size =
-                        of_getintprop_default(dp, "ecache-line-size", def);
-                printk("CPU[0]: Caches "
-                       "D[sz(%d):line_sz(%d)] "
-                       "I[sz(%d):line_sz(%d)] "
-                       "E[sz(%d):line_sz(%d)]\n",
-                       cpu_data(0).dcache_size, cpu_data(0).dcache_line_size,
-                       cpu_data(0).icache_size, cpu_data(0).icache_line_size,
-                       cpu_data(0).ecache_size, cpu_data(0).ecache_line_size);
-        }
-#endif
-        central_probe();
-        cpu_probe();
-}
diff --git a/arch/sparc64/kernel/entry.S b/arch/sparc64/kernel/entry.S
index 732b77cb71f8..ed712e0b3372 100644
--- a/arch/sparc64/kernel/entry.S
+++ b/arch/sparc64/kernel/entry.S
@@ -1725,96 +1725,142 @@ real_hard_smp_processor_id:
         * returns %o0: sysino
         */
        .globl  sun4v_devino_to_sysino
+        .type   sun4v_devino_to_sysino,#function
 sun4v_devino_to_sysino:
        mov     HV_FAST_INTR_DEVINO2SYSINO, %o5
        ta      HV_FAST_TRAP
        retl
         mov    %o1, %o0
+        .size   sun4v_devino_to_sysino, .-sun4v_devino_to_sysino
        /* %o0: sysino
         *
         * returns %o0: intr_enabled (HV_INTR_{DISABLED,ENABLED})
         */
        .globl  sun4v_intr_getenabled
+        .type   sun4v_intr_getenabled,#function
 sun4v_intr_getenabled:
        mov     HV_FAST_INTR_GETENABLED, %o5
        ta      HV_FAST_TRAP
        retl
         mov    %o1, %o0
+        .size   sun4v_intr_getenabled, .-sun4v_intr_getenabled
        /* %o0: sysino
         * %o1: intr_enabled (HV_INTR_{DISABLED,ENABLED})
         */
        .globl  sun4v_intr_setenabled
+        .type   sun4v_intr_setenabled,#function
 sun4v_intr_setenabled:
        mov     HV_FAST_INTR_SETENABLED, %o5
        ta      HV_FAST_TRAP
        retl
         nop
+        .size   sun4v_intr_setenabled, .-sun4v_intr_setenabled
        /* %o0: sysino
         *
         * returns %o0: intr_state (HV_INTR_STATE_*)
         */
        .globl  sun4v_intr_getstate
+        .type   sun4v_intr_getstate,#function
 sun4v_intr_getstate:
        mov     HV_FAST_INTR_GETSTATE, %o5
        ta      HV_FAST_TRAP
        retl
         mov    %o1, %o0
+        .size   sun4v_intr_getstate, .-sun4v_intr_getstate
        /* %o0: sysino
         * %o1: intr_state (HV_INTR_STATE_*)
         */
        .globl  sun4v_intr_setstate
+        .type   sun4v_intr_setstate,#function
 sun4v_intr_setstate:
        mov     HV_FAST_INTR_SETSTATE, %o5
        ta      HV_FAST_TRAP
        retl
         nop
+        .size   sun4v_intr_setstate, .-sun4v_intr_setstate
        /* %o0: sysino
         *
         * returns %o0: cpuid
         */
        .globl  sun4v_intr_gettarget
+        .type   sun4v_intr_gettarget,#function
 sun4v_intr_gettarget:
        mov     HV_FAST_INTR_GETTARGET, %o5
        ta      HV_FAST_TRAP
        retl
         mov    %o1, %o0
+        .size   sun4v_intr_gettarget, .-sun4v_intr_gettarget
        /* %o0: sysino
         * %o1: cpuid
         */
        .globl  sun4v_intr_settarget
+        .type   sun4v_intr_settarget,#function
 sun4v_intr_settarget:
        mov     HV_FAST_INTR_SETTARGET, %o5
        ta      HV_FAST_TRAP
        retl
         nop
+        .size   sun4v_intr_settarget, .-sun4v_intr_settarget
-        /* %o0: type
+        /* %o0: cpuid
-         * %o1: queue paddr
+         * %o1: pc
-         * %o2: num queue entries
+         * %o2: rtba
+         * %o3: arg0
         *
         * returns %o0: status
         */
-        .globl  sun4v_cpu_qconf
+        .globl  sun4v_cpu_start
-sun4v_cpu_qconf:
+        .type   sun4v_cpu_start,#function
-        mov     HV_FAST_CPU_QCONF, %o5
+sun4v_cpu_start:
+        mov     HV_FAST_CPU_START, %o5
        ta      HV_FAST_TRAP
        retl
         nop
+        .size   sun4v_cpu_start, .-sun4v_cpu_start
-        /* returns %o0: status
+        /* %o0: cpuid
+         *
+         * returns %o0: status
         */
+        .globl  sun4v_cpu_stop
+        .type   sun4v_cpu_stop,#function
+sun4v_cpu_stop:
+        mov     HV_FAST_CPU_STOP, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_cpu_stop, .-sun4v_cpu_stop
+        /* returns %o0: status  */
        .globl  sun4v_cpu_yield
+        .type   sun4v_cpu_yield, #function
 sun4v_cpu_yield:
        mov     HV_FAST_CPU_YIELD, %o5
        ta      HV_FAST_TRAP
        retl
         nop
+        .size   sun4v_cpu_yield, .-sun4v_cpu_yield
+        /* %o0: type
+         * %o1: queue paddr
+         * %o2: num queue entries
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_cpu_qconf
+        .type   sun4v_cpu_qconf,#function
+sun4v_cpu_qconf:
+        mov     HV_FAST_CPU_QCONF, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_cpu_qconf, .-sun4v_cpu_qconf
        /* %o0: num cpus in cpu list
         * %o1: cpu list paddr
@@ -1823,11 +1869,13 @@ sun4v_cpu_yield:
         * returns %o0: status
         */
        .globl  sun4v_cpu_mondo_send
+        .type   sun4v_cpu_mondo_send,#function
 sun4v_cpu_mondo_send:
        mov     HV_FAST_CPU_MONDO_SEND, %o5
        ta      HV_FAST_TRAP
        retl
         nop
+        .size   sun4v_cpu_mondo_send, .-sun4v_cpu_mondo_send
        /* %o0: CPU ID
         *
@@ -1835,6 +1883,7 @@ sun4v_cpu_mondo_send:
         *         %o0: cpu state as HV_CPU_STATE_*
         */
        .globl  sun4v_cpu_state
+        .type   sun4v_cpu_state,#function
 sun4v_cpu_state:
        mov     HV_FAST_CPU_STATE, %o5
        ta      HV_FAST_TRAP
@@ -1843,6 +1892,37 @@ sun4v_cpu_state:
        mov     %o1, %o0
 1:      retl
         nop
+        .size   sun4v_cpu_state, .-sun4v_cpu_state
+        /* %o0: virtual address
+         * %o1: must be zero
+         * %o2: TTE
+         * %o3: HV_MMU_* flags
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_mmu_map_perm_addr
+        .type   sun4v_mmu_map_perm_addr,#function
+sun4v_mmu_map_perm_addr:
+        mov     HV_FAST_MMU_MAP_PERM_ADDR, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_mmu_map_perm_addr, .-sun4v_mmu_map_perm_addr
+        /* %o0: number of TSB descriptions
+         * %o1: TSB descriptions real address
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_mmu_tsb_ctx0
+        .type   sun4v_mmu_tsb_ctx0,#function
+sun4v_mmu_tsb_ctx0:
+        mov     HV_FAST_MMU_TSB_CTX0, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_mmu_tsb_ctx0, .-sun4v_mmu_tsb_ctx0
        /* %o0: API group number
         * %o1: pointer to unsigned long major number storage
@@ -1851,6 +1931,7 @@ sun4v_cpu_state:
         * returns %o0: status
         */
        .globl  sun4v_get_version
+        .type   sun4v_get_version,#function
 sun4v_get_version:
        mov     HV_CORE_GET_VER, %o5
        mov     %o1, %o3
@@ -1859,6 +1940,7 @@ sun4v_get_version:
        stx     %o1, [%o3]
        retl
         stx    %o2, [%o4]
+        .size   sun4v_get_version, .-sun4v_get_version
        /* %o0: API group number
         * %o1: desired major number
@@ -1868,18 +1950,49 @@ sun4v_get_version:
         * returns %o0: status
         */
        .globl  sun4v_set_version
+        .type   sun4v_set_version,#function
 sun4v_set_version:
        mov     HV_CORE_SET_VER, %o5
        mov     %o3, %o4
        ta      HV_CORE_TRAP
        retl
         stx    %o1, [%o4]
+        .size   sun4v_set_version, .-sun4v_set_version
+        /* %o0: pointer to unsigned long time
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_tod_get
+        .type   sun4v_tod_get,#function
+sun4v_tod_get:
+        mov     %o0, %o4
+        mov     HV_FAST_TOD_GET, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%o4]
+        retl
+         nop
+        .size   sun4v_tod_get, .-sun4v_tod_get
+        /* %o0: time
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_tod_set
+        .type   sun4v_tod_set,#function
+sun4v_tod_set:
+        mov     HV_FAST_TOD_SET, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_tod_set, .-sun4v_tod_set
        /* %o0: pointer to unsigned long status
         *
         * returns %o0: signed character
         */
        .globl  sun4v_con_getchar
+        .type   sun4v_con_getchar,#function
 sun4v_con_getchar:
        mov     %o0, %o4
        mov     HV_FAST_CONS_GETCHAR, %o5
@@ -1889,17 +2002,20 @@ sun4v_con_getchar:
        stx     %o0, [%o4]
        retl
         sra    %o1, 0, %o0
+        .size   sun4v_con_getchar, .-sun4v_con_getchar
        /* %o0: signed long character
         *
         * returns %o0: status
         */
        .globl  sun4v_con_putchar
+        .type   sun4v_con_putchar,#function
 sun4v_con_putchar:
        mov     HV_FAST_CONS_PUTCHAR, %o5
        ta      HV_FAST_TRAP
        retl
         sra    %o0, 0, %o0
+        .size   sun4v_con_putchar, .-sun4v_con_putchar
        /* %o0: buffer real address
         * %o1: buffer size
@@ -1908,6 +2024,7 @@ sun4v_con_putchar:
         * returns %o0: status
         */
        .globl  sun4v_con_read
+        .type   sun4v_con_read,#function
 sun4v_con_read:
        mov     %o2, %o4
        mov     HV_FAST_CONS_READ, %o5
@@ -1922,6 +2039,7 @@ sun4v_con_read:
        stx     %o1, [%o4]
 1:      retl
         nop
+        .size   sun4v_con_read, .-sun4v_con_read
        /* %o0: buffer real address
         * %o1: buffer size
@@ -1930,6 +2048,7 @@ sun4v_con_read:
         * returns %o0: status
         */
        .globl  sun4v_con_write
+        .type   sun4v_con_write,#function
 sun4v_con_write:
        mov     %o2, %o4
        mov     HV_FAST_CONS_WRITE, %o5
@@ -1937,3 +2056,517 @@ sun4v_con_write:
        stx     %o1, [%o4]
        retl
         nop
+        .size   sun4v_con_write, .-sun4v_con_write
+        /* %o0: soft state
+         * %o1: address of description string
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_mach_set_soft_state
+        .type   sun4v_mach_set_soft_state,#function
+sun4v_mach_set_soft_state:
+        mov     HV_FAST_MACH_SET_SOFT_STATE, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_mach_set_soft_state, .-sun4v_mach_set_soft_state
+        /* %o0: exit code
+         *
+         * Does not return.
+         */
+        .globl  sun4v_mach_exit
+        .type   sun4v_mach_exit,#function
+sun4v_mach_exit:
+        mov     HV_FAST_MACH_EXIT, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_mach_exit, .-sun4v_mach_exit
+        /* %o0: buffer real address
+         * %o1: buffer length
+         * %o2: pointer to unsigned long real_buf_len
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_mach_desc
+        .type   sun4v_mach_desc,#function
+sun4v_mach_desc:
+        mov     %o2, %o4
+        mov     HV_FAST_MACH_DESC, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%o4]
+        retl
+         nop
+        .size   sun4v_mach_desc, .-sun4v_mach_desc
+        /* %o0: new timeout in milliseconds
+         * %o1: pointer to unsigned long orig_timeout
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_mach_set_watchdog
+        .type   sun4v_mach_set_watchdog,#function
+sun4v_mach_set_watchdog:
+        mov     %o1, %o4
+        mov     HV_FAST_MACH_SET_WATCHDOG, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%o4]
+        retl
+         nop
+        .size   sun4v_mach_set_watchdog, .-sun4v_mach_set_watchdog
+        /* No inputs and does not return.  */
+        .globl  sun4v_mach_sir
+        .type   sun4v_mach_sir,#function
+sun4v_mach_sir:
+        mov     %o1, %o4
+        mov     HV_FAST_MACH_SIR, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%o4]
+        retl
+         nop
+        .size   sun4v_mach_sir, .-sun4v_mach_sir
+        /* %o0: channel
+         * %o1: ra
+         * %o2: num_entries
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_ldc_tx_qconf
+        .type   sun4v_ldc_tx_qconf,#function
+sun4v_ldc_tx_qconf:
+        mov     HV_FAST_LDC_TX_QCONF, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_ldc_tx_qconf, .-sun4v_ldc_tx_qconf
+        /* %o0: channel
+         * %o1: pointer to unsigned long ra
+         * %o2: pointer to unsigned long num_entries
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_ldc_tx_qinfo
+        .type   sun4v_ldc_tx_qinfo,#function
+sun4v_ldc_tx_qinfo:
+        mov     %o1, %g1
+        mov     %o2, %g2
+        mov     HV_FAST_LDC_TX_QINFO, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%g1]
+        stx     %o2, [%g2]
+        retl
+         nop
+        .size   sun4v_ldc_tx_qinfo, .-sun4v_ldc_tx_qinfo
+        /* %o0: channel
+         * %o1: pointer to unsigned long head_off
+         * %o2: pointer to unsigned long tail_off
+         * %o2: pointer to unsigned long chan_state
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_ldc_tx_get_state
+        .type   sun4v_ldc_tx_get_state,#function
+sun4v_ldc_tx_get_state:
+        mov     %o1, %g1
+        mov     %o2, %g2
+        mov     %o3, %g3
+        mov     HV_FAST_LDC_TX_GET_STATE, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%g1]
+        stx     %o2, [%g2]
+        stx     %o3, [%g3]
+        retl
+         nop
+        .size   sun4v_ldc_tx_get_state, .-sun4v_ldc_tx_get_state
+        /* %o0: channel
+         * %o1: tail_off
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_ldc_tx_set_qtail
+        .type   sun4v_ldc_tx_set_qtail,#function
+sun4v_ldc_tx_set_qtail:
+        mov     HV_FAST_LDC_TX_SET_QTAIL, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_ldc_tx_set_qtail, .-sun4v_ldc_tx_set_qtail
+        /* %o0: channel
+         * %o1: ra
+         * %o2: num_entries
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_ldc_rx_qconf
+        .type   sun4v_ldc_rx_qconf,#function
+sun4v_ldc_rx_qconf:
+        mov     HV_FAST_LDC_RX_QCONF, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_ldc_rx_qconf, .-sun4v_ldc_rx_qconf
+        /* %o0: channel
+         * %o1: pointer to unsigned long ra
+         * %o2: pointer to unsigned long num_entries
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_ldc_rx_qinfo
+        .type   sun4v_ldc_rx_qinfo,#function
+sun4v_ldc_rx_qinfo:
+        mov     %o1, %g1
+        mov     %o2, %g2
+        mov     HV_FAST_LDC_RX_QINFO, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%g1]
+        stx     %o2, [%g2]
+        retl
+         nop
+        .size   sun4v_ldc_rx_qinfo, .-sun4v_ldc_rx_qinfo
+        /* %o0: channel
+         * %o1: pointer to unsigned long head_off
+         * %o2: pointer to unsigned long tail_off
+         * %o2: pointer to unsigned long chan_state
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_ldc_rx_get_state
+        .type   sun4v_ldc_rx_get_state,#function
+sun4v_ldc_rx_get_state:
+        mov     %o1, %g1
+        mov     %o2, %g2
+        mov     %o3, %g3
+        mov     HV_FAST_LDC_RX_GET_STATE, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%g1]
+        stx     %o2, [%g2]
+        stx     %o3, [%g3]
+        retl
+         nop
+        .size   sun4v_ldc_rx_get_state, .-sun4v_ldc_rx_get_state
+        /* %o0: channel
+         * %o1: head_off
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_ldc_rx_set_qhead
+        .type   sun4v_ldc_rx_set_qhead,#function
+sun4v_ldc_rx_set_qhead:
+        mov     HV_FAST_LDC_RX_SET_QHEAD, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_ldc_rx_set_qhead, .-sun4v_ldc_rx_set_qhead
+        /* %o0: channel
+         * %o1: ra
+         * %o2: num_entries
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_ldc_set_map_table
+        .type   sun4v_ldc_set_map_table,#function
+sun4v_ldc_set_map_table:
+        mov     HV_FAST_LDC_SET_MAP_TABLE, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_ldc_set_map_table, .-sun4v_ldc_set_map_table
+        /* %o0: channel
+         * %o1: pointer to unsigned long ra
+         * %o2: pointer to unsigned long num_entries
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_ldc_get_map_table
+        .type   sun4v_ldc_get_map_table,#function
+sun4v_ldc_get_map_table:
+        mov     %o1, %g1
+        mov     %o2, %g2
+        mov     HV_FAST_LDC_GET_MAP_TABLE, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%g1]
+        stx     %o2, [%g2]
+        retl
+         nop
+        .size   sun4v_ldc_get_map_table, .-sun4v_ldc_get_map_table
+        /* %o0: channel
+         * %o1: dir_code
+         * %o2: tgt_raddr
+         * %o3: lcl_raddr
+         * %o4: len
+         * %o5: pointer to unsigned long actual_len
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_ldc_copy
+        .type   sun4v_ldc_copy,#function
+sun4v_ldc_copy:
+        mov     %o5, %g1
+        mov     HV_FAST_LDC_COPY, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%g1]
+        retl
+         nop
+        .size   sun4v_ldc_copy, .-sun4v_ldc_copy
+        /* %o0: channel
+         * %o1: cookie
+         * %o2: pointer to unsigned long ra
+         * %o3: pointer to unsigned long perm
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_ldc_mapin
+        .type   sun4v_ldc_mapin,#function
+sun4v_ldc_mapin:
+        mov     %o2, %g1
+        mov     %o3, %g2
+        mov     HV_FAST_LDC_MAPIN, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%g1]
+        stx     %o2, [%g2]
+        retl
+         nop
+        .size   sun4v_ldc_mapin, .-sun4v_ldc_mapin
+        /* %o0: ra
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_ldc_unmap
+        .type   sun4v_ldc_unmap,#function
+sun4v_ldc_unmap:
+        mov     HV_FAST_LDC_UNMAP, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_ldc_unmap, .-sun4v_ldc_unmap
+        /* %o0: cookie
+         * %o1: mte_cookie
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_ldc_revoke
+        .type   sun4v_ldc_revoke,#function
+sun4v_ldc_revoke:
+        mov     HV_FAST_LDC_REVOKE, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_ldc_revoke, .-sun4v_ldc_revoke
+        /* %o0: device handle
+         * %o1: device INO
+         * %o2: pointer to unsigned long cookie
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_vintr_get_cookie
+        .type   sun4v_vintr_get_cookie,#function
+sun4v_vintr_get_cookie:
+        mov     %o2, %g1
+        mov     HV_FAST_VINTR_GET_COOKIE, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%g1]
+        retl
+         nop
+        .size   sun4v_vintr_get_cookie, .-sun4v_vintr_get_cookie
+        /* %o0: device handle
+         * %o1: device INO
+         * %o2: cookie
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_vintr_set_cookie
+        .type   sun4v_vintr_set_cookie,#function
+sun4v_vintr_set_cookie:
+        mov     HV_FAST_VINTR_SET_COOKIE, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_vintr_set_cookie, .-sun4v_vintr_set_cookie
+        /* %o0: device handle
+         * %o1: device INO
+         * %o2: pointer to unsigned long valid_state
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_vintr_get_valid
+        .type   sun4v_vintr_get_valid,#function
+sun4v_vintr_get_valid:
+        mov     %o2, %g1
+        mov     HV_FAST_VINTR_GET_VALID, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%g1]
+        retl
+         nop
+        .size   sun4v_vintr_get_valid, .-sun4v_vintr_get_valid
+        /* %o0: device handle
+         * %o1: device INO
+         * %o2: valid_state
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_vintr_set_valid
+        .type   sun4v_vintr_set_valid,#function
+sun4v_vintr_set_valid:
+        mov     HV_FAST_VINTR_SET_VALID, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_vintr_set_valid, .-sun4v_vintr_set_valid
+        /* %o0: device handle
+         * %o1: device INO
+         * %o2: pointer to unsigned long state
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_vintr_get_state
+        .type   sun4v_vintr_get_state,#function
+sun4v_vintr_get_state:
+        mov     %o2, %g1
+        mov     HV_FAST_VINTR_GET_STATE, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%g1]
+        retl
+         nop
+        .size   sun4v_vintr_get_state, .-sun4v_vintr_get_state
+        /* %o0: device handle
+         * %o1: device INO
+         * %o2: state
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_vintr_set_state
+        .type   sun4v_vintr_set_state,#function
+sun4v_vintr_set_state:
+        mov     HV_FAST_VINTR_SET_STATE, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_vintr_set_state, .-sun4v_vintr_set_state
+        /* %o0: device handle
+         * %o1: device INO
+         * %o2: pointer to unsigned long cpuid
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_vintr_get_target
+        .type   sun4v_vintr_get_target,#function
+sun4v_vintr_get_target:
+        mov     %o2, %g1
+        mov     HV_FAST_VINTR_GET_TARGET, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%g1]
+        retl
+         nop
+        .size   sun4v_vintr_get_target, .-sun4v_vintr_get_target
+        /* %o0: device handle
+         * %o1: device INO
+         * %o2: cpuid
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_vintr_set_target
+        .type   sun4v_vintr_set_target,#function
+sun4v_vintr_set_target:
+        mov     HV_FAST_VINTR_SET_TARGET, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_vintr_set_target, .-sun4v_vintr_set_target
+        /* %o0: NCS sub-function
+         * %o1: sub-function arg real-address
+         * %o2: sub-function arg size
+         *
+         * returns %o0: status
+         */
+        .globl  sun4v_ncs_request
+        .type   sun4v_ncs_request,#function
+sun4v_ncs_request:
+        mov     HV_FAST_NCS_REQUEST, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_ncs_request, .-sun4v_ncs_request
+        .globl  sun4v_scv_send
+        .type   sun4v_scv_send,#function
+sun4v_scv_send:
+        save    %sp, -192, %sp
+        mov     %i0, %o0
+        mov     %i1, %o1
+        mov     %i2, %o2
+        mov     HV_FAST_SVC_SEND, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%i3]
+        ret
+        restore
+        .size   sun4v_scv_send, .-sun4v_scv_send
+        .globl  sun4v_scv_recv
+        .type   sun4v_scv_recv,#function
+sun4v_scv_recv:
+        save    %sp, -192, %sp
+        mov     %i0, %o0
+        mov     %i1, %o1
+        mov     %i2, %o2
+        mov     HV_FAST_SVC_RECV, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%i3]
+        ret
+        restore
+        .size   sun4v_scv_recv, .-sun4v_scv_recv
+        .globl  sun4v_scv_getstatus
+        .type   sun4v_scv_getstatus,#function
+sun4v_scv_getstatus:
+        mov     HV_FAST_SVC_GETSTATUS, %o5
+        mov     %o1, %o4
+        ta      HV_FAST_TRAP
+        stx     %o1, [%o4]
+        retl
+         nop
+        .size   sun4v_scv_getstatus, .-sun4v_scv_getstatus
+        .globl  sun4v_scv_setstatus
+        .type   sun4v_scv_setstatus,#function
+sun4v_scv_setstatus:
+        mov     HV_FAST_SVC_SETSTATUS, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_scv_setstatus, .-sun4v_scv_setstatus
+        .globl  sun4v_scv_clrstatus
+        .type   sun4v_scv_clrstatus,#function
+sun4v_scv_clrstatus:
+        mov     HV_FAST_SVC_CLRSTATUS, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_scv_clrstatus, .-sun4v_scv_clrstatus
diff --git a/arch/sparc64/kernel/head.S b/arch/sparc64/kernel/head.S
index baea10a98196..77259526cb15 100644
--- a/arch/sparc64/kernel/head.S
+++ b/arch/sparc64/kernel/head.S
@@ -523,7 +523,7 @@ tlb_fixup_done:
 #else
        mov     0, %o0
 #endif
-        stb     %o0, [%g6 + TI_CPU]
+        sth     %o0, [%g6 + TI_CPU]
        /* Off we go.... */
        call    start_kernel
@@ -653,33 +653,54 @@ setup_tba:
         restore
 sparc64_boot_end:
-#include "ktlb.S"
-#include "tsb.S"
 #include "etrap.S"
 #include "rtrap.S"
 #include "winfixup.S"
 #include "entry.S"
 #include "sun4v_tlb_miss.S"
 #include "sun4v_ivec.S"
+#include "ktlb.S"
+#include "tsb.S"
 /*
 * The following skip makes sure the trap table in ttable.S is aligned
 * on a 32K boundary as required by the v9 specs for TBA register.
 *
 * We align to a 32K boundary, then we have the 32K kernel TSB,
- * then the 32K aligned trap table.
+ * the 64K kernel 4MB TSB, and then the 32K aligned trap table.
 */
 1:
        .skip   0x4000 + _start - 1b
+! 0x0000000000408000
        .globl  swapper_tsb
 swapper_tsb:
        .skip   (32 * 1024)
-! 0x0000000000408000
+        .globl  swapper_4m_tsb
+swapper_4m_tsb:
+        .skip   (64 * 1024)
+! 0x0000000000420000
+        /* Some care needs to be exercised if you try to move the
+         * location of the trap table relative to other things.  For
+         * one thing there are br* instructions in some of the
+         * trap table entires which branch back to code in ktlb.S
+         * Those instructions can only handle a signed 16-bit
+         * displacement.
+         *
+         * There is a binutils bug (bugzilla #4558) which causes
+         * the relocation overflow checks for such instructions to
+         * not be done correctly.  So bintuils will not notice the
+         * error and will instead write junk into the relocation and
+         * you'll have an unbootable kernel.
+         */
 #include "ttable.S"
+! 0x0000000000428000
 #include "systbls.S"
        .data
diff --git a/arch/sparc64/kernel/hvapi.c b/arch/sparc64/kernel/hvapi.c
index f03ffc829c7a..f34f5d6181ef 100644
--- a/arch/sparc64/kernel/hvapi.c
+++ b/arch/sparc64/kernel/hvapi.c
@@ -9,6 +9,7 @@
 #include <asm/hypervisor.h>
 #include <asm/oplib.h>
+#include <asm/sstate.h>
 /* If the hypervisor indicates that the API setting
 * calls are unsupported, by returning HV_EBADTRAP or
@@ -107,7 +108,7 @@ int sun4v_hvapi_register(unsigned long group, unsigned long major,
                                p->minor = actual_minor;
                                ret = 0;
                        } else if (hv_ret == HV_EBADTRAP ||
-                                   HV_ENOTSUPPORTED) {
+                                   hv_ret == HV_ENOTSUPPORTED) {
                                if (p->flags & FLAG_PRE_API) {
                                        if (major == 1) {
                                                p->major = 1;
@@ -179,6 +180,8 @@ void __init sun4v_hvapi_init(void)
        if (sun4v_hvapi_register(group, major, &minor))
                goto bad;
+        sun4v_sstate_init();
        return;
 bad:
diff --git a/arch/sparc64/kernel/irq.c b/arch/sparc64/kernel/irq.c
index 3edc18e1b818..a36f8dd0c021 100644
--- a/arch/sparc64/kernel/irq.c
+++ b/arch/sparc64/kernel/irq.c
@@ -171,8 +171,6 @@ skip:
        return 0;
 }
-extern unsigned long real_hard_smp_processor_id(void);
 static unsigned int sun4u_compute_tid(unsigned long imap, unsigned long cpuid)
 {
        unsigned int tid;
@@ -694,9 +692,20 @@ void init_irqwork_curcpu(void)
        trap_block[cpu].irq_worklist = 0;
 }
-static void __cpuinit register_one_mondo(unsigned long paddr, unsigned long type)
+/* Please be very careful with register_one_mondo() and
+ * sun4v_register_mondo_queues().
+ *
+ * On SMP this gets invoked from the CPU trampoline before
+ * the cpu has fully taken over the trap table from OBP,
+ * and it's kernel stack + %g6 thread register state is
+ * not fully cooked yet.
+ *
+ * Therefore you cannot make any OBP calls, not even prom_printf,
+ * from these two routines.
+ */
+static void __cpuinit register_one_mondo(unsigned long paddr, unsigned long type, unsigned long qmask)
 {
-        unsigned long num_entries = 128;
+        unsigned long num_entries = (qmask + 1) / 64;
        unsigned long status;
        status = sun4v_cpu_qconf(type, paddr, num_entries);
@@ -711,44 +720,58 @@ static void __cpuinit sun4v_register_mondo_queues(int this_cpu)
 {
        struct trap_per_cpu *tb = &trap_block[this_cpu];
-        register_one_mondo(tb->cpu_mondo_pa, HV_CPU_QUEUE_CPU_MONDO);
+        register_one_mondo(tb->cpu_mondo_pa, HV_CPU_QUEUE_CPU_MONDO,
-        register_one_mondo(tb->dev_mondo_pa, HV_CPU_QUEUE_DEVICE_MONDO);
+                           tb->cpu_mondo_qmask);
-        register_one_mondo(tb->resum_mondo_pa, HV_CPU_QUEUE_RES_ERROR);
+        register_one_mondo(tb->dev_mondo_pa, HV_CPU_QUEUE_DEVICE_MONDO,
-        register_one_mondo(tb->nonresum_mondo_pa, HV_CPU_QUEUE_NONRES_ERROR);
+                           tb->dev_mondo_qmask);
+        register_one_mondo(tb->resum_mondo_pa, HV_CPU_QUEUE_RES_ERROR,
+                           tb->resum_qmask);
+        register_one_mondo(tb->nonresum_mondo_pa, HV_CPU_QUEUE_NONRES_ERROR,
+                           tb->nonresum_qmask);
 }
-static void __cpuinit alloc_one_mondo(unsigned long *pa_ptr, int use_bootmem)
+static void __cpuinit alloc_one_mondo(unsigned long *pa_ptr, unsigned long qmask, int use_bootmem)
 {
-        void *page;
+        unsigned long size = PAGE_ALIGN(qmask + 1);
+        unsigned long order = get_order(size);
+        void *p = NULL;
-        if (use_bootmem)
+        if (use_bootmem) {
-                page = alloc_bootmem_low_pages(PAGE_SIZE);
+                p = __alloc_bootmem_low(size, size, 0);
-        else
+        } else {
-                page = (void *) get_zeroed_page(GFP_ATOMIC);
+                struct page *page = alloc_pages(GFP_ATOMIC | __GFP_ZERO, order);
+                if (page)
+                        p = page_address(page);
+        }
-        if (!page) {
+        if (!p) {
                prom_printf("SUN4V: Error, cannot allocate mondo queue.\n");
                prom_halt();
        }
-        *pa_ptr = __pa(page);
+        *pa_ptr = __pa(p);
 }
-static void __cpuinit alloc_one_kbuf(unsigned long *pa_ptr, int use_bootmem)
+static void __cpuinit alloc_one_kbuf(unsigned long *pa_ptr, unsigned long qmask, int use_bootmem)
 {
-        void *page;
+        unsigned long size = PAGE_ALIGN(qmask + 1);
+        unsigned long order = get_order(size);
+        void *p = NULL;
-        if (use_bootmem)
+        if (use_bootmem) {
-                page = alloc_bootmem_low_pages(PAGE_SIZE);
+                p = __alloc_bootmem_low(size, size, 0);
-        else
+        } else {
-                page = (void *) get_zeroed_page(GFP_ATOMIC);
+                struct page *page = alloc_pages(GFP_ATOMIC | __GFP_ZERO, order);
+                if (page)
+                        p = page_address(page);
+        }
-        if (!page) {
+        if (!p) {
                prom_printf("SUN4V: Error, cannot allocate kbuf page.\n");
                prom_halt();
        }
-        *pa_ptr = __pa(page);
+        *pa_ptr = __pa(p);
 }
 static void __cpuinit init_cpu_send_mondo_info(struct trap_per_cpu *tb, int use_bootmem)
@@ -779,12 +802,12 @@ void __cpuinit sun4v_init_mondo_queues(int use_bootmem, int cpu, int alloc, int
        struct trap_per_cpu *tb = &trap_block[cpu];
        if (alloc) {
-                alloc_one_mondo(&tb->cpu_mondo_pa, use_bootmem);
+                alloc_one_mondo(&tb->cpu_mondo_pa, tb->cpu_mondo_qmask, use_bootmem);
-                alloc_one_mondo(&tb->dev_mondo_pa, use_bootmem);
+                alloc_one_mondo(&tb->dev_mondo_pa, tb->dev_mondo_qmask, use_bootmem);
-                alloc_one_mondo(&tb->resum_mondo_pa, use_bootmem);
+                alloc_one_mondo(&tb->resum_mondo_pa, tb->resum_qmask, use_bootmem);
-                alloc_one_kbuf(&tb->resum_kernel_buf_pa, use_bootmem);
+                alloc_one_kbuf(&tb->resum_kernel_buf_pa, tb->resum_qmask, use_bootmem);
-                alloc_one_mondo(&tb->nonresum_mondo_pa, use_bootmem);
+                alloc_one_mondo(&tb->nonresum_mondo_pa, tb->nonresum_qmask, use_bootmem);
-                alloc_one_kbuf(&tb->nonresum_kernel_buf_pa, use_bootmem);
+                alloc_one_kbuf(&tb->nonresum_kernel_buf_pa, tb->nonresum_qmask, use_bootmem);
                init_cpu_send_mondo_info(tb, use_bootmem);
        }
diff --git a/arch/sparc64/kernel/itlb_miss.S b/arch/sparc64/kernel/itlb_miss.S
index ad46e2024f4b..5a8377b54955 100644
--- a/arch/sparc64/kernel/itlb_miss.S
+++ b/arch/sparc64/kernel/itlb_miss.S
@@ -11,12 +11,12 @@
 /* ITLB ** ICACHE line 2: TSB compare and TLB load      */
        bne,pn  %xcc, tsb_miss_itlb             ! Miss
         mov    FAULT_CODE_ITLB, %g3
-        andcc   %g5, _PAGE_EXEC_4U, %g0         ! Executable?
+        sethi   %hi(_PAGE_EXEC_4U), %g4
+        andcc   %g5, %g4, %g0                   ! Executable?
        be,pn   %xcc, tsb_do_fault
         nop                                    ! Delay slot, fill me
        stxa    %g5, [%g0] ASI_ITLB_DATA_IN     ! Load TLB
        retry                                   ! Trap done
-        nop
 /* ITLB ** ICACHE line 3:                               */
        nop
diff --git a/arch/sparc64/kernel/mdesc.c b/arch/sparc64/kernel/mdesc.c
new file mode 100644
index 000000000000..9246c2cf9574
--- /dev/null
+++ b/arch/sparc64/kernel/mdesc.c
@@ -0,0 +1,619 @@
+/* mdesc.c: Sun4V machine description handling.
+ *
+ * Copyright (C) 2007 David S. Miller <davem@davemloft.net>
+ */
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/bootmem.h>
+#include <linux/log2.h>
+#include <asm/hypervisor.h>
+#include <asm/mdesc.h>
+#include <asm/prom.h>
+#include <asm/oplib.h>
+#include <asm/smp.h>
+/* Unlike the OBP device tree, the machine description is a full-on
+ * DAG.  An arbitrary number of ARCs are possible from one
+ * node to other nodes and thus we can't use the OBP device_node
+ * data structure to represent these nodes inside of the kernel.
+ *
+ * Actually, it isn't even a DAG, because there are back pointers
+ * which create cycles in the graph.
+ *
+ * mdesc_hdr and mdesc_elem describe the layout of the data structure
+ * we get from the Hypervisor.
+ */
+struct mdesc_hdr {
+        u32     version; /* Transport version */
+        u32     node_sz; /* node block size */
+        u32     name_sz; /* name block size */
+        u32     data_sz; /* data block size */
+};
+struct mdesc_elem {
+        u8      tag;
+#define MD_LIST_END     0x00
+#define MD_NODE         0x4e
+#define MD_NODE_END     0x45
+#define MD_NOOP         0x20
+#define MD_PROP_ARC     0x61
+#define MD_PROP_VAL     0x76
+#define MD_PROP_STR     0x73
+#define MD_PROP_DATA    0x64
+        u8      name_len;
+        u16     resv;
+        u32     name_offset;
+        union {
+                struct {
+                        u32     data_len;
+                        u32     data_offset;
+                } data;
+                u64     val;
+        } d;
+};
+static struct mdesc_hdr *main_mdesc;
+static struct mdesc_node *allnodes;
+static struct mdesc_node *allnodes_tail;
+static unsigned int unique_id;
+static struct mdesc_node **mdesc_hash;
+static unsigned int mdesc_hash_size;
+static inline unsigned int node_hashfn(u64 node)
+{
+        return ((unsigned int) (node ^ (node >> 8) ^ (node >> 16)))
+                & (mdesc_hash_size - 1);
+}
+static inline void hash_node(struct mdesc_node *mp)
+{
+        struct mdesc_node **head = &mdesc_hash[node_hashfn(mp->node)];
+        mp->hash_next = *head;
+        *head = mp;
+        if (allnodes_tail) {
+                allnodes_tail->allnodes_next = mp;
+                allnodes_tail = mp;
+        } else {
+                allnodes = allnodes_tail = mp;
+        }
+}
+static struct mdesc_node *find_node(u64 node)
+{
+        struct mdesc_node *mp = mdesc_hash[node_hashfn(node)];
+        while (mp) {
+                if (mp->node == node)
+                        return mp;
+                mp = mp->hash_next;
+        }
+        return NULL;
+}
+struct property *md_find_property(const struct mdesc_node *mp,
+                                  const char *name,
+                                  int *lenp)
+{
+        struct property *pp;
+        for (pp = mp->properties; pp != 0; pp = pp->next) {
+                if (strcasecmp(pp->name, name) == 0) {
+                        if (lenp)
+                                *lenp = pp->length;
+                        break;
+                }
+        }
+        return pp;
+}
+EXPORT_SYMBOL(md_find_property);
+/*
+ * Find a property with a given name for a given node
+ * and return the value.
+ */
+const void *md_get_property(const struct mdesc_node *mp, const char *name,
+                            int *lenp)
+{
+        struct property *pp = md_find_property(mp, name, lenp);
+        return pp ? pp->value : NULL;
+}
+EXPORT_SYMBOL(md_get_property);
+struct mdesc_node *md_find_node_by_name(struct mdesc_node *from,
+                                        const char *name)
+{
+        struct mdesc_node *mp;
+        mp = from ? from->allnodes_next : allnodes;
+        for (; mp != NULL; mp = mp->allnodes_next) {
+                if (strcmp(mp->name, name) == 0)
+                        break;
+        }
+        return mp;
+}
+EXPORT_SYMBOL(md_find_node_by_name);
+static unsigned int mdesc_early_allocated;
+static void * __init mdesc_early_alloc(unsigned long size)
+{
+        void *ret;
+        ret = __alloc_bootmem(size, SMP_CACHE_BYTES, 0UL);
+        if (ret == NULL) {
+                prom_printf("MDESC: alloc of %lu bytes failed.\n", size);
+                prom_halt();
+        }
+        memset(ret, 0, size);
+        mdesc_early_allocated += size;
+        return ret;
+}
+static unsigned int __init count_arcs(struct mdesc_elem *ep)
+{
+        unsigned int ret = 0;
+        ep++;
+        while (ep->tag != MD_NODE_END) {
+                if (ep->tag == MD_PROP_ARC)
+                        ret++;
+                ep++;
+        }
+        return ret;
+}
+static void __init mdesc_node_alloc(u64 node, struct mdesc_elem *ep, const char *names)
+{
+        unsigned int num_arcs = count_arcs(ep);
+        struct mdesc_node *mp;
+        mp = mdesc_early_alloc(sizeof(*mp) +
+                               (num_arcs * sizeof(struct mdesc_arc)));
+        mp->name = names + ep->name_offset;
+        mp->node = node;
+        mp->unique_id = unique_id++;
+        mp->num_arcs = num_arcs;
+        hash_node(mp);
+}
+static inline struct mdesc_elem *node_block(struct mdesc_hdr *mdesc)
+{
+        return (struct mdesc_elem *) (mdesc + 1);
+}
+static inline void *name_block(struct mdesc_hdr *mdesc)
+{
+        return ((void *) node_block(mdesc)) + mdesc->node_sz;
+}
+static inline void *data_block(struct mdesc_hdr *mdesc)
+{
+        return ((void *) name_block(mdesc)) + mdesc->name_sz;
+}
+/* In order to avoid recursion (the graph can be very deep) we use a
+ * two pass algorithm.  First we allocate all the nodes and hash them.
+ * Then we iterate over each node, filling in the arcs and properties.
+ */
+static void __init build_all_nodes(struct mdesc_hdr *mdesc)
+{
+        struct mdesc_elem *start, *ep;
+        struct mdesc_node *mp;
+        const char *names;
+        void *data;
+        u64 last_node;
+        start = ep = node_block(mdesc);
+        last_node = mdesc->node_sz / 16;
+        names = name_block(mdesc);
+        while (1) {
+                u64 node = ep - start;
+                if (ep->tag == MD_LIST_END)
+                        break;
+                if (ep->tag != MD_NODE) {
+                        prom_printf("MDESC: Inconsistent element list.\n");
+                        prom_halt();
+                }
+                mdesc_node_alloc(node, ep, names);
+                if (ep->d.val >= last_node) {
+                        printk("MDESC: Warning, early break out of node scan.\n");
+                        printk("MDESC: Next node [%lu] last_node [%lu].\n",
+                               node, last_node);
+                        break;
+                }
+                ep = start + ep->d.val;
+        }
+        data = data_block(mdesc);
+        for (mp = allnodes; mp; mp = mp->allnodes_next) {
+                struct mdesc_elem *ep = start + mp->node;
+                struct property **link = &mp->properties;
+                unsigned int this_arc = 0;
+                ep++;
+                while (ep->tag != MD_NODE_END) {
+                        switch (ep->tag) {
+                        case MD_PROP_ARC: {
+                                struct mdesc_node *target;
+                                if (this_arc >= mp->num_arcs) {
+                                        prom_printf("MDESC: ARC overrun [%u:%u]\n",
+                                                    this_arc, mp->num_arcs);
+                                        prom_halt();
+                                }
+                                target = find_node(ep->d.val);
+                                if (!target) {
+                                        printk("MDESC: Warning, arc points to "
+                                               "missing node, ignoring.\n");
+                                        break;
+                                }
+                                mp->arcs[this_arc].name =
+                                        (names + ep->name_offset);
+                                mp->arcs[this_arc].arc = target;
+                                this_arc++;
+                                break;
+                        }
+                        case MD_PROP_VAL:
+                        case MD_PROP_STR:
+                        case MD_PROP_DATA: {
+                                struct property *p = mdesc_early_alloc(sizeof(*p));
+                                p->unique_id = unique_id++;
+                                p->name = (char *) names + ep->name_offset;
+                                if (ep->tag == MD_PROP_VAL) {
+                                        p->value = &ep->d.val;
+                                        p->length = 8;
+                                } else {
+                                        p->value = data + ep->d.data.data_offset;
+                                        p->length = ep->d.data.data_len;
+                                }
+                                *link = p;
+                                link = &p->next;
+                                break;
+                        }
+                        case MD_NOOP:
+                                break;
+                        default:
+                                printk("MDESC: Warning, ignoring unknown tag type %02x\n",
+                                       ep->tag);
+                        }
+                        ep++;
+                }
+        }
+}
+static unsigned int __init count_nodes(struct mdesc_hdr *mdesc)
+{
+        struct mdesc_elem *ep = node_block(mdesc);
+        struct mdesc_elem *end;
+        unsigned int cnt = 0;
+        end = ((void *)ep) + mdesc->node_sz;
+        while (ep < end) {
+                if (ep->tag == MD_NODE)
+                        cnt++;
+                ep++;
+        }
+        return cnt;
+}
+static void __init report_platform_properties(void)
+{
+        struct mdesc_node *pn = md_find_node_by_name(NULL, "platform");
+        const char *s;
+        const u64 *v;
+        if (!pn) {
+                prom_printf("No platform node in machine-description.\n");
+                prom_halt();
+        }
+        s = md_get_property(pn, "banner-name", NULL);
+        printk("PLATFORM: banner-name [%s]\n", s);
+        s = md_get_property(pn, "name", NULL);
+        printk("PLATFORM: name [%s]\n", s);
+        v = md_get_property(pn, "hostid", NULL);
+        if (v)
+                printk("PLATFORM: hostid [%08lx]\n", *v);
+        v = md_get_property(pn, "serial#", NULL);
+        if (v)
+                printk("PLATFORM: serial# [%08lx]\n", *v);
+        v = md_get_property(pn, "stick-frequency", NULL);
+        printk("PLATFORM: stick-frequency [%08lx]\n", *v);
+        v = md_get_property(pn, "mac-address", NULL);
+        if (v)
+                printk("PLATFORM: mac-address [%lx]\n", *v);
+        v = md_get_property(pn, "watchdog-resolution", NULL);
+        if (v)
+                printk("PLATFORM: watchdog-resolution [%lu ms]\n", *v);
+        v = md_get_property(pn, "watchdog-max-timeout", NULL);
+        if (v)
+                printk("PLATFORM: watchdog-max-timeout [%lu ms]\n", *v);
+        v = md_get_property(pn, "max-cpus", NULL);
+        if (v)
+                printk("PLATFORM: max-cpus [%lu]\n", *v);
+}
+static int inline find_in_proplist(const char *list, const char *match, int len)
+{
+        while (len > 0) {
+                int l;
+                if (!strcmp(list, match))
+                        return 1;
+                l = strlen(list) + 1;
+                list += l;
+                len -= l;
+        }
+        return 0;
+}
+static void __init fill_in_one_cache(cpuinfo_sparc *c, struct mdesc_node *mp)
+{
+        const u64 *level = md_get_property(mp, "level", NULL);
+        const u64 *size = md_get_property(mp, "size", NULL);
+        const u64 *line_size = md_get_property(mp, "line-size", NULL);
+        const char *type;
+        int type_len;
+        type = md_get_property(mp, "type", &type_len);
+        switch (*level) {
+        case 1:
+                if (find_in_proplist(type, "instn", type_len)) {
+                        c->icache_size = *size;
+                        c->icache_line_size = *line_size;
+                } else if (find_in_proplist(type, "data", type_len)) {
+                        c->dcache_size = *size;
+                        c->dcache_line_size = *line_size;
+                }
+                break;
+        case 2:
+                c->ecache_size = *size;
+                c->ecache_line_size = *line_size;
+                break;
+        default:
+                break;
+        }
+        if (*level == 1) {
+                unsigned int i;
+                for (i = 0; i < mp->num_arcs; i++) {
+                        struct mdesc_node *t = mp->arcs[i].arc;
+                        if (strcmp(mp->arcs[i].name, "fwd"))
+                                continue;
+                        if (!strcmp(t->name, "cache"))
+                                fill_in_one_cache(c, t);
+                }
+        }
+}
+static void __init mark_core_ids(struct mdesc_node *mp, int core_id)
+{
+        unsigned int i;
+        for (i = 0; i < mp->num_arcs; i++) {
+                struct mdesc_node *t = mp->arcs[i].arc;
+                const u64 *id;
+                if (strcmp(mp->arcs[i].name, "back"))
+                        continue;
+                if (!strcmp(t->name, "cpu")) {
+                        id = md_get_property(t, "id", NULL);
+                        if (*id < NR_CPUS)
+                                cpu_data(*id).core_id = core_id;
+                } else {
+                        unsigned int j;
+                        for (j = 0; j < t->num_arcs; j++) {
+                                struct mdesc_node *n = t->arcs[j].arc;
+                                if (strcmp(t->arcs[j].name, "back"))
+                                        continue;
+                                if (strcmp(n->name, "cpu"))
+                                        continue;
+                                id = md_get_property(n, "id", NULL);
+                                if (*id < NR_CPUS)
+                                        cpu_data(*id).core_id = core_id;
+                        }
+                }
+        }
+}
+static void __init set_core_ids(void)
+{
+        struct mdesc_node *mp;
+        int idx;
+        idx = 1;
+        md_for_each_node_by_name(mp, "cache") {
+                const u64 *level = md_get_property(mp, "level", NULL);
+                const char *type;
+                int len;
+                if (*level != 1)
+                        continue;
+                type = md_get_property(mp, "type", &len);
+                if (!find_in_proplist(type, "instn", len))
+                        continue;
+                mark_core_ids(mp, idx);
+                idx++;
+        }
+}
+static void __init get_one_mondo_bits(const u64 *p, unsigned int *mask, unsigned char def)
+{
+        u64 val;
+        if (!p)
+                goto use_default;
+        val = *p;
+        if (!val || val >= 64)
+                goto use_default;
+        *mask = ((1U << val) * 64U) - 1U;
+        return;
+use_default:
+        *mask = ((1U << def) * 64U) - 1U;
+}
+static void __init get_mondo_data(struct mdesc_node *mp, struct trap_per_cpu *tb)
+{
+        const u64 *val;
+        val = md_get_property(mp, "q-cpu-mondo-#bits", NULL);
+        get_one_mondo_bits(val, &tb->cpu_mondo_qmask, 7);
+        val = md_get_property(mp, "q-dev-mondo-#bits", NULL);
+        get_one_mondo_bits(val, &tb->dev_mondo_qmask, 7);
+        val = md_get_property(mp, "q-resumable-#bits", NULL);
+        get_one_mondo_bits(val, &tb->resum_qmask, 6);
+        val = md_get_property(mp, "q-nonresumable-#bits", NULL);
+        get_one_mondo_bits(val, &tb->nonresum_qmask, 2);
+}
+static void __init mdesc_fill_in_cpu_data(void)
+{
+        struct mdesc_node *mp;
+        ncpus_probed = 0;
+        md_for_each_node_by_name(mp, "cpu") {
+                const u64 *id = md_get_property(mp, "id", NULL);
+                const u64 *cfreq = md_get_property(mp, "clock-frequency", NULL);
+                struct trap_per_cpu *tb;
+                cpuinfo_sparc *c;
+                unsigned int i;
+                int cpuid;
+                ncpus_probed++;
+                cpuid = *id;
+#ifdef CONFIG_SMP
+                if (cpuid >= NR_CPUS)
+                        continue;
+#else
+                /* On uniprocessor we only want the values for the
+                 * real physical cpu the kernel booted onto, however
+                 * cpu_data() only has one entry at index 0.
+                 */
+                if (cpuid != real_hard_smp_processor_id())
+                        continue;
+                cpuid = 0;
+#endif
+                c = &cpu_data(cpuid);
+                c->clock_tick = *cfreq;
+                tb = &trap_block[cpuid];
+                get_mondo_data(mp, tb);
+                for (i = 0; i < mp->num_arcs; i++) {
+                        struct mdesc_node *t = mp->arcs[i].arc;
+                        unsigned int j;
+                        if (strcmp(mp->arcs[i].name, "fwd"))
+                                continue;
+                        if (!strcmp(t->name, "cache")) {
+                                fill_in_one_cache(c, t);
+                                continue;
+                        }
+                        for (j = 0; j < t->num_arcs; j++) {
+                                struct mdesc_node *n;
+                                n = t->arcs[j].arc;
+                                if (strcmp(t->arcs[j].name, "fwd"))
+                                        continue;
+                                if (!strcmp(n->name, "cache"))
+                                        fill_in_one_cache(c, n);
+                        }
+                }
+#ifdef CONFIG_SMP
+                cpu_set(cpuid, cpu_present_map);
+                cpu_set(cpuid, phys_cpu_present_map);
+#endif
+                c->core_id = 0;
+        }
+        set_core_ids();
+        smp_fill_in_sib_core_maps();
+}
+void __init sun4v_mdesc_init(void)
+{
+        unsigned long len, real_len, status;
+        (void) sun4v_mach_desc(0UL, 0UL, &len);
+        printk("MDESC: Size is %lu bytes.\n", len);
+        main_mdesc = mdesc_early_alloc(len);
+        status = sun4v_mach_desc(__pa(main_mdesc), len, &real_len);
+        if (status != HV_EOK || real_len > len) {
+                prom_printf("sun4v_mach_desc fails, err(%lu), "
+                            "len(%lu), real_len(%lu)\n",
+                            status, len, real_len);
+                prom_halt();
+        }
+        len = count_nodes(main_mdesc);
+        printk("MDESC: %lu nodes.\n", len);
+        len = roundup_pow_of_two(len);
+        mdesc_hash = mdesc_early_alloc(len * sizeof(struct mdesc_node *));
+        mdesc_hash_size = len;
+        printk("MDESC: Hash size %lu entries.\n", len);
+        build_all_nodes(main_mdesc);
+        printk("MDESC: Built graph with %u bytes of memory.\n",
+               mdesc_early_allocated);
+        report_platform_properties();
+        mdesc_fill_in_cpu_data();
+}
diff --git a/arch/sparc64/kernel/pci.c b/arch/sparc64/kernel/pci.c
index d4c077dc5e85..38a32bc95d22 100644
--- a/arch/sparc64/kernel/pci.c
+++ b/arch/sparc64/kernel/pci.c
@@ -306,6 +306,20 @@ static void __init pci_controller_probe(void)
        pci_controller_scan(pci_controller_init);
 }
+static int ofpci_verbose;
+static int __init ofpci_debug(char *str)
+{
+        int val = 0;
+        get_option(&str, &val);
+        if (val)
+                ofpci_verbose = 1;
+        return 1;
+}
+__setup("ofpci_debug=", ofpci_debug);
 static unsigned long pci_parse_of_flags(u32 addr0)
 {
        unsigned long flags = 0;
@@ -337,7 +351,9 @@ static void pci_parse_of_addrs(struct of_device *op,
        addrs = of_get_property(node, "assigned-addresses", &proplen);
        if (!addrs)
                return;
-        printk("    parse addresses (%d bytes) @ %p\n", proplen, addrs);
+        if (ofpci_verbose)
+                printk("    parse addresses (%d bytes) @ %p\n",
+                       proplen, addrs);
        op_res = &op->resource[0];
        for (; proplen >= 20; proplen -= 20, addrs += 5, op_res++) {
                struct resource *res;
@@ -348,8 +364,9 @@ static void pci_parse_of_addrs(struct of_device *op,
                if (!flags)
                        continue;
                i = addrs[0] & 0xff;
-                printk("  start: %lx, end: %lx, i: %x\n",
+                if (ofpci_verbose)
-                       op_res->start, op_res->end, i);
+                        printk("  start: %lx, end: %lx, i: %x\n",
+                               op_res->start, op_res->end, i);
                if (PCI_BASE_ADDRESS_0 <= i && i <= PCI_BASE_ADDRESS_5) {
                        res = &dev->resource[(i - PCI_BASE_ADDRESS_0) >> 2];
@@ -393,8 +410,9 @@ struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm,
        if (type == NULL)
                type = "";
-        printk("    create device, devfn: %x, type: %s hostcontroller(%d)\n",
+        if (ofpci_verbose)
-               devfn, type, host_controller);
+                printk("    create device, devfn: %x, type: %s\n",
+                       devfn, type);
        dev->bus = bus;
        dev->sysdata = node;
@@ -434,8 +452,9 @@ struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm,
                sprintf(pci_name(dev), "%04x:%02x:%02x.%d", pci_domain_nr(bus),
                        dev->bus->number, PCI_SLOT(devfn), PCI_FUNC(devfn));
        }
-        printk("    class: 0x%x device name: %s\n",
+        if (ofpci_verbose)
-               dev->class, pci_name(dev));
+                printk("    class: 0x%x device name: %s\n",
+                       dev->class, pci_name(dev));
        /* I have seen IDE devices which will not respond to
         * the bmdma simplex check reads if bus mastering is
@@ -469,7 +488,8 @@ struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm,
        }
        pci_parse_of_addrs(sd->op, node, dev);
-        printk("    adding to system ...\n");
+        if (ofpci_verbose)
+                printk("    adding to system ...\n");
        pci_device_add(dev, bus);
@@ -547,7 +567,8 @@ static void __devinit of_scan_pci_bridge(struct pci_pbm_info *pbm,
        unsigned int flags;
        u64 size;
-        printk("of_scan_pci_bridge(%s)\n", node->full_name);
+        if (ofpci_verbose)
+                printk("of_scan_pci_bridge(%s)\n", node->full_name);
        /* parse bus-range property */
        busrange = of_get_property(node, "bus-range", &len);
@@ -632,7 +653,8 @@ static void __devinit of_scan_pci_bridge(struct pci_pbm_info *pbm,
 simba_cont:
        sprintf(bus->name, "PCI Bus %04x:%02x", pci_domain_nr(bus),
                bus->number);
-        printk("    bus name: %s\n", bus->name);
+        if (ofpci_verbose)
+                printk("    bus name: %s\n", bus->name);
        pci_of_scan_bus(pbm, node, bus);
 }
@@ -646,12 +668,14 @@ static void __devinit pci_of_scan_bus(struct pci_pbm_info *pbm,
        int reglen, devfn;
        struct pci_dev *dev;
-        printk("PCI: scan_bus[%s] bus no %d\n",
+        if (ofpci_verbose)
-               node->full_name, bus->number);
+                printk("PCI: scan_bus[%s] bus no %d\n",
+                       node->full_name, bus->number);
        child = NULL;
        while ((child = of_get_next_child(node, child)) != NULL) {
-                printk("  * %s\n", child->full_name);
+                if (ofpci_verbose)
+                        printk("  * %s\n", child->full_name);
                reg = of_get_property(child, "reg", &reglen);
                if (reg == NULL || reglen < 20)
                        continue;
@@ -661,7 +685,9 @@ static void __devinit pci_of_scan_bus(struct pci_pbm_info *pbm,
                dev = of_create_pci_dev(pbm, child, bus, devfn, 0);
                if (!dev)
                        continue;
-                printk("PCI: dev header type: %x\n", dev->hdr_type);
+                if (ofpci_verbose)
+                        printk("PCI: dev header type: %x\n",
+                               dev->hdr_type);
                if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE ||
                    dev->hdr_type == PCI_HEADER_TYPE_CARDBUS)
diff --git a/arch/sparc64/kernel/pci_sabre.c b/arch/sparc64/kernel/pci_sabre.c
index e2377796de89..323d6c278518 100644
--- a/arch/sparc64/kernel/pci_sabre.c
+++ b/arch/sparc64/kernel/pci_sabre.c
@@ -762,9 +762,10 @@ void sabre_init(struct device_node *dp, char *model_name)
                        /* Of course, Sun has to encode things a thousand
                         * different ways, inconsistently.
                         */
-                        cpu_find_by_instance(0, &dp, NULL);
+                        for_each_node_by_type(dp, "cpu") {
-                        if (!strcmp(dp->name, "SUNW,UltraSPARC-IIe"))
+                                if (!strcmp(dp->name, "SUNW,UltraSPARC-IIe"))
-                                hummingbird_p = 1;
+                                        hummingbird_p = 1;
+                        }
                }
        }
diff --git a/arch/sparc64/kernel/pci_sun4v.c b/arch/sparc64/kernel/pci_sun4v.c
index 044e8ec4c0f5..6b3fe2c1d65e 100644
--- a/arch/sparc64/kernel/pci_sun4v.c
+++ b/arch/sparc64/kernel/pci_sun4v.c
@@ -12,6 +12,7 @@
 #include <linux/percpu.h>
 #include <linux/irq.h>
 #include <linux/msi.h>
+#include <linux/log2.h>
 #include <asm/iommu.h>
 #include <asm/irq.h>
@@ -26,6 +27,9 @@
 #include "pci_sun4v.h"
+static unsigned long vpci_major = 1;
+static unsigned long vpci_minor = 1;
 #define PGLIST_NENTS    (PAGE_SIZE / sizeof(u64))
 struct iommu_batch {
@@ -638,9 +642,8 @@ static void pci_sun4v_iommu_init(struct pci_pbm_info *pbm)
 {
        struct iommu *iommu = pbm->iommu;
        struct property *prop;
-        unsigned long num_tsb_entries, sz;
+        unsigned long num_tsb_entries, sz, tsbsize;
        u32 vdma[2], dma_mask, dma_offset;
-        int tsbsize;
        prop = of_find_property(pbm->prom_node, "virtual-dma", NULL);
        if (prop) {
@@ -654,31 +657,15 @@ static void pci_sun4v_iommu_init(struct pci_pbm_info *pbm)
                vdma[1] = 0x80000000;
        }
-        dma_mask = vdma[0];
+        if ((vdma[0] | vdma[1]) & ~IO_PAGE_MASK) {
-        switch (vdma[1]) {
+                prom_printf("PCI-SUN4V: strange virtual-dma[%08x:%08x].\n",
-                case 0x20000000:
+                            vdma[0], vdma[1]);
-                        dma_mask |= 0x1fffffff;
+                prom_halt();
-                        tsbsize = 64;
-                        break;
-                case 0x40000000:
-                        dma_mask |= 0x3fffffff;
-                        tsbsize = 128;
-                        break;
-                case 0x80000000:
-                        dma_mask |= 0x7fffffff;
-                        tsbsize = 256;
-                        break;
-                default:
-                        prom_printf("PCI-SUN4V: strange virtual-dma size.\n");
-                        prom_halt();
        };
-        tsbsize *= (8 * 1024);
+        dma_mask = (roundup_pow_of_two(vdma[1]) - 1UL);
+        num_tsb_entries = vdma[1] / IO_PAGE_SIZE;
-        num_tsb_entries = tsbsize / sizeof(iopte_t);
+        tsbsize = num_tsb_entries * sizeof(iopte_t);
        dma_offset = vdma[0];
@@ -689,7 +676,7 @@ static void pci_sun4v_iommu_init(struct pci_pbm_info *pbm)
        iommu->dma_addr_mask = dma_mask;
        /* Allocate and initialize the free area map.  */
-        sz = num_tsb_entries / 8;
+        sz = (num_tsb_entries + 7) / 8;
        sz = (sz + 7UL) & ~7UL;
        iommu->arena.map = kzalloc(sz, GFP_KERNEL);
        if (!iommu->arena.map) {
@@ -1178,6 +1165,7 @@ static void pci_sun4v_pbm_init(struct pci_controller_info *p, struct device_node
 void sun4v_pci_init(struct device_node *dp, char *model_name)
 {
+        static int hvapi_negotiated = 0;
        struct pci_controller_info *p;
        struct pci_pbm_info *pbm;
        struct iommu *iommu;
@@ -1186,6 +1174,20 @@ void sun4v_pci_init(struct device_node *dp, char *model_name)
        u32 devhandle;
        int i;
+        if (!hvapi_negotiated++) {
+                int err = sun4v_hvapi_register(HV_GRP_PCI,
+                                               vpci_major,
+                                               &vpci_minor);
+                if (err) {
+                        prom_printf("SUN4V_PCI: Could not register hvapi, "
+                                    "err=%d\n", err);
+                        prom_halt();
+                }
+                printk("SUN4V_PCI: Registered hvapi major[%lu] minor[%lu]\n",
+                       vpci_major, vpci_minor);
+        }
        prop = of_find_property(dp, "reg", NULL);
        regs = prop->value;
diff --git a/arch/sparc64/kernel/power.c b/arch/sparc64/kernel/power.c
index 699b24b890df..5d6adea3967f 100644
--- a/arch/sparc64/kernel/power.c
+++ b/arch/sparc64/kernel/power.c
@@ -19,6 +19,7 @@
 #include <asm/prom.h>
 #include <asm/of_device.h>
 #include <asm/io.h>
+#include <asm/sstate.h>
 #include <linux/unistd.h>
@@ -53,6 +54,7 @@ static void (*poweroff_method)(void) = machine_alt_power_off;
 void machine_power_off(void)
 {
+        sstate_poweroff();
        if (!serial_console || scons_pwroff) {
 #ifdef CONFIG_PCI
                if (power_reg) {
diff --git a/arch/sparc64/kernel/process.c b/arch/sparc64/kernel/process.c
index 952762bfb4c0..f5f97e2c669c 100644
--- a/arch/sparc64/kernel/process.c
+++ b/arch/sparc64/kernel/process.c
@@ -45,6 +45,7 @@
 #include <asm/mmu_context.h>
 #include <asm/unistd.h>
 #include <asm/hypervisor.h>
+#include <asm/sstate.h>
 /* #define VERBOSE_SHOWREGS */
@@ -106,6 +107,7 @@ extern void (*prom_keyboard)(void);
 void machine_halt(void)
 {
+        sstate_halt();
        if (!serial_console && prom_palette)
                prom_palette (1);
        if (prom_keyboard)
@@ -116,6 +118,7 @@ void machine_halt(void)
 void machine_alt_power_off(void)
 {
+        sstate_poweroff();
        if (!serial_console && prom_palette)
                prom_palette(1);
        if (prom_keyboard)
@@ -128,6 +131,7 @@ void machine_restart(char * cmd)
 {
        char *p;
        
+        sstate_reboot();
        p = strchr (reboot_command, '\n');
        if (p) *p = 0;
        if (!serial_console && prom_palette)
diff --git a/arch/sparc64/kernel/prom.c b/arch/sparc64/kernel/prom.c
index 02830e4671f5..dad4b3ba705f 100644
--- a/arch/sparc64/kernel/prom.c
+++ b/arch/sparc64/kernel/prom.c
@@ -28,6 +28,7 @@
 #include <asm/irq.h>
 #include <asm/asi.h>
 #include <asm/upa.h>
+#include <asm/smp.h>
 static struct device_node *allnodes;
@@ -1665,6 +1666,150 @@ static struct device_node * __init build_tree(struct device_node *parent, phandl
        return ret;
 }
+static const char *get_mid_prop(void)
+{
+        return (tlb_type == spitfire ? "upa-portid" : "portid");
+}
+struct device_node *of_find_node_by_cpuid(int cpuid)
+{
+        struct device_node *dp;
+        const char *mid_prop = get_mid_prop();
+        for_each_node_by_type(dp, "cpu") {
+                int id = of_getintprop_default(dp, mid_prop, -1);
+                const char *this_mid_prop = mid_prop;
+                if (id < 0) {
+                        this_mid_prop = "cpuid";
+                        id = of_getintprop_default(dp, this_mid_prop, -1);
+                }
+                if (id < 0) {
+                        prom_printf("OF: Serious problem, cpu lacks "
+                                    "%s property", this_mid_prop);
+                        prom_halt();
+                }
+                if (cpuid == id)
+                        return dp;
+        }
+        return NULL;
+}
+static void __init of_fill_in_cpu_data(void)
+{
+        struct device_node *dp;
+        const char *mid_prop = get_mid_prop();
+        ncpus_probed = 0;
+        for_each_node_by_type(dp, "cpu") {
+                int cpuid = of_getintprop_default(dp, mid_prop, -1);
+                const char *this_mid_prop = mid_prop;
+                struct device_node *portid_parent;
+                int portid = -1;
+                portid_parent = NULL;
+                if (cpuid < 0) {
+                        this_mid_prop = "cpuid";
+                        cpuid = of_getintprop_default(dp, this_mid_prop, -1);
+                        if (cpuid >= 0) {
+                                int limit = 2;
+                                portid_parent = dp;
+                                while (limit--) {
+                                        portid_parent = portid_parent->parent;
+                                        if (!portid_parent)
+                                                break;
+                                        portid = of_getintprop_default(portid_parent,
+                                                                       "portid", -1);
+                                        if (portid >= 0)
+                                                break;
+                                }
+                        }
+                }
+                if (cpuid < 0) {
+                        prom_printf("OF: Serious problem, cpu lacks "
+                                    "%s property", this_mid_prop);
+                        prom_halt();
+                }
+                ncpus_probed++;
+#ifdef CONFIG_SMP
+                if (cpuid >= NR_CPUS)
+                        continue;
+#else
+                /* On uniprocessor we only want the values for the
+                 * real physical cpu the kernel booted onto, however
+                 * cpu_data() only has one entry at index 0.
+                 */
+                if (cpuid != real_hard_smp_processor_id())
+                        continue;
+                cpuid = 0;
+#endif
+                cpu_data(cpuid).clock_tick =
+                        of_getintprop_default(dp, "clock-frequency", 0);
+                if (portid_parent) {
+                        cpu_data(cpuid).dcache_size =
+                                of_getintprop_default(dp, "l1-dcache-size",
+                                                      16 * 1024);
+                        cpu_data(cpuid).dcache_line_size =
+                                of_getintprop_default(dp, "l1-dcache-line-size",
+                                                      32);
+                        cpu_data(cpuid).icache_size =
+                                of_getintprop_default(dp, "l1-icache-size",
+                                                      8 * 1024);
+                        cpu_data(cpuid).icache_line_size =
+                                of_getintprop_default(dp, "l1-icache-line-size",
+                                                      32);
+                        cpu_data(cpuid).ecache_size =
+                                of_getintprop_default(dp, "l2-cache-size", 0);
+                        cpu_data(cpuid).ecache_line_size =
+                                of_getintprop_default(dp, "l2-cache-line-size", 0);
+                        if (!cpu_data(cpuid).ecache_size ||
+                            !cpu_data(cpuid).ecache_line_size) {
+                                cpu_data(cpuid).ecache_size =
+                                        of_getintprop_default(portid_parent,
+                                                              "l2-cache-size",
+                                                              (4 * 1024 * 1024));
+                                cpu_data(cpuid).ecache_line_size =
+                                        of_getintprop_default(portid_parent,
+                                                              "l2-cache-line-size", 64);
+                        }
+                        cpu_data(cpuid).core_id = portid + 1;
+                } else {
+                        cpu_data(cpuid).dcache_size =
+                                of_getintprop_default(dp, "dcache-size", 16 * 1024);
+                        cpu_data(cpuid).dcache_line_size =
+                                of_getintprop_default(dp, "dcache-line-size", 32);
+                        cpu_data(cpuid).icache_size =
+                                of_getintprop_default(dp, "icache-size", 16 * 1024);
+                        cpu_data(cpuid).icache_line_size =
+                                of_getintprop_default(dp, "icache-line-size", 32);
+                        cpu_data(cpuid).ecache_size =
+                                of_getintprop_default(dp, "ecache-size",
+                                                      (4 * 1024 * 1024));
+                        cpu_data(cpuid).ecache_line_size =
+                                of_getintprop_default(dp, "ecache-line-size", 64);
+                        cpu_data(cpuid).core_id = 0;
+                }
+#ifdef CONFIG_SMP
+                cpu_set(cpuid, cpu_present_map);
+                cpu_set(cpuid, phys_cpu_present_map);
+#endif
+        }
+        smp_fill_in_sib_core_maps();
+}
 void __init prom_build_devicetree(void)
 {
        struct device_node **nextp;
@@ -1679,4 +1824,7 @@ void __init prom_build_devicetree(void)
                                     &nextp);
        printk("PROM: Built device tree with %u bytes of memory.\n",
               prom_early_allocated);
+        if (tlb_type != hypervisor)
+                of_fill_in_cpu_data();
 }
diff --git a/arch/sparc64/kernel/setup.c b/arch/sparc64/kernel/setup.c
index dea9c3c9ec5f..de9b4c13f1c7 100644
--- a/arch/sparc64/kernel/setup.c
+++ b/arch/sparc64/kernel/setup.c
@@ -46,11 +46,17 @@
 #include <asm/sections.h>
 #include <asm/setup.h>
 #include <asm/mmu.h>
+#include <asm/ns87303.h>
 #ifdef CONFIG_IP_PNP
 #include <net/ipconfig.h>
 #endif
+/* Used to synchronize accesses to NatSemi SUPER I/O chip configure
+ * operations in asm/ns87303.h
+ */
+DEFINE_SPINLOCK(ns87303_lock);
 struct screen_info screen_info = {
        0, 0,                   /* orig-x, orig-y */
        0,                      /* unused */
@@ -370,8 +376,6 @@ void __init setup_arch(char **cmdline_p)
        init_cur_cpu_trap(current_thread_info());
        paging_init();
-        smp_setup_cpu_possible_map();
 }
 static int __init set_preferred_console(void)
@@ -424,7 +428,7 @@ extern void mmu_info(struct seq_file *);
 unsigned int dcache_parity_tl1_occurred;
 unsigned int icache_parity_tl1_occurred;
-static int ncpus_probed;
+int ncpus_probed;
 static int show_cpuinfo(struct seq_file *m, void *__unused)
 {
@@ -516,14 +520,6 @@ static int __init topology_init(void)
        err = -ENOMEM;
-        /* Count the number of physically present processors in
-         * the machine, even on uniprocessor, so that /proc/cpuinfo
-         * output is consistent with 2.4.x
-         */
-        ncpus_probed = 0;
-        while (!cpu_find_by_instance(ncpus_probed, NULL, NULL))
-                ncpus_probed++;
        for_each_possible_cpu(i) {
                struct cpu *p = kzalloc(sizeof(*p), GFP_KERNEL);
                if (p) {
diff --git a/arch/sparc64/kernel/smp.c b/arch/sparc64/kernel/smp.c
index 24fdf1d0adc5..c550bba3490a 100644
--- a/arch/sparc64/kernel/smp.c
+++ b/arch/sparc64/kernel/smp.c
@@ -40,6 +40,7 @@
 #include <asm/tlb.h>
 #include <asm/sections.h>
 #include <asm/prom.h>
+#include <asm/mdesc.h>
 extern void calibrate_delay(void);
@@ -75,53 +76,6 @@ void smp_bogo(struct seq_file *m)
                           i, cpu_data(i).clock_tick);
 }
-void __init smp_store_cpu_info(int id)
-{
-        struct device_node *dp;
-        int def;
-        cpu_data(id).udelay_val                 = loops_per_jiffy;
-        cpu_find_by_mid(id, &dp);
-        cpu_data(id).clock_tick =
-                of_getintprop_default(dp, "clock-frequency", 0);
-        def = ((tlb_type == hypervisor) ? (8 * 1024) : (16 * 1024));
-        cpu_data(id).dcache_size =
-                of_getintprop_default(dp, "dcache-size", def);
-        def = 32;
-        cpu_data(id).dcache_line_size =
-                of_getintprop_default(dp, "dcache-line-size", def);
-        def = 16 * 1024;
-        cpu_data(id).icache_size =
-                of_getintprop_default(dp, "icache-size", def);
-        def = 32;
-        cpu_data(id).icache_line_size =
-                of_getintprop_default(dp, "icache-line-size", def);
-        def = ((tlb_type == hypervisor) ?
-               (3 * 1024 * 1024) :
-               (4 * 1024 * 1024));
-        cpu_data(id).ecache_size =
-                of_getintprop_default(dp, "ecache-size", def);
-        def = 64;
-        cpu_data(id).ecache_line_size =
-                of_getintprop_default(dp, "ecache-line-size", def);
-        printk("CPU[%d]: Caches "
-               "D[sz(%d):line_sz(%d)] "
-               "I[sz(%d):line_sz(%d)] "
-               "E[sz(%d):line_sz(%d)]\n",
-               id,
-               cpu_data(id).dcache_size, cpu_data(id).dcache_line_size,
-               cpu_data(id).icache_size, cpu_data(id).icache_line_size,
-               cpu_data(id).ecache_size, cpu_data(id).ecache_line_size);
-}
 extern void setup_sparc64_timer(void);
 static volatile unsigned long callin_flag = 0;
@@ -145,7 +99,7 @@ void __init smp_callin(void)
        local_irq_enable();
        calibrate_delay();
-        smp_store_cpu_info(cpuid);
+        cpu_data(cpuid).udelay_val = loops_per_jiffy;
        callin_flag = 1;
        __asm__ __volatile__("membar #Sync\n\t"
                             "flush  %%g6" : : : "memory");
@@ -340,9 +294,8 @@ static int __devinit smp_boot_one_cpu(unsigned int cpu)
                prom_startcpu_cpuid(cpu, entry, cookie);
        } else {
-                struct device_node *dp;
+                struct device_node *dp = of_find_node_by_cpuid(cpu);
-                cpu_find_by_mid(cpu, &dp);
                prom_startcpu(dp->node, entry, cookie);
        }
@@ -447,7 +400,7 @@ static __inline__ void spitfire_xcall_deliver(u64 data0, u64 data1, u64 data2, c
 static void cheetah_xcall_deliver(u64 data0, u64 data1, u64 data2, cpumask_t mask)
 {
        u64 pstate, ver;
-        int nack_busy_id, is_jbus;
+        int nack_busy_id, is_jbus, need_more;
        if (cpus_empty(mask))
                return;
@@ -463,6 +416,7 @@ static void cheetah_xcall_deliver(u64 data0, u64 data1, u64 data2, cpumask_t mas
        __asm__ __volatile__("rdpr %%pstate, %0" : "=r" (pstate));
 retry:
+        need_more = 0;
        __asm__ __volatile__("wrpr %0, %1, %%pstate\n\t"
                             : : "r" (pstate), "i" (PSTATE_IE));
@@ -491,6 +445,10 @@ retry:
                                : /* no outputs */
                                : "r" (target), "i" (ASI_INTR_W));
                        nack_busy_id++;
+                        if (nack_busy_id == 32) {
+                                need_more = 1;
+                                break;
+                        }
                }
        }
@@ -507,6 +465,16 @@ retry:
                        if (dispatch_stat == 0UL) {
                                __asm__ __volatile__("wrpr %0, 0x0, %%pstate"
                                                     : : "r" (pstate));
+                                if (unlikely(need_more)) {
+                                        int i, cnt = 0;
+                                        for_each_cpu_mask(i, mask) {
+                                                cpu_clear(i, mask);
+                                                cnt++;
+                                                if (cnt == 32)
+                                                        break;
+                                        }
+                                        goto retry;
+                                }
                                return;
                        }
                        if (!--stuck)
@@ -544,6 +512,8 @@ retry:
                                if ((dispatch_stat & check_mask) == 0)
                                        cpu_clear(i, mask);
                                this_busy_nack += 2;
+                                if (this_busy_nack == 64)
+                                        break;
                        }
                        goto retry;
@@ -1191,23 +1161,14 @@ int setup_profiling_timer(unsigned int multiplier)
 static void __init smp_tune_scheduling(void)
 {
-        struct device_node *dp;
+        unsigned int smallest = ~0U;
-        int instance;
+        int i;
-        unsigned int def, smallest = ~0U;
-        def = ((tlb_type == hypervisor) ?
-               (3 * 1024 * 1024) :
-               (4 * 1024 * 1024));
-        instance = 0;
+        for (i = 0; i < NR_CPUS; i++) {
-        while (!cpu_find_by_instance(instance, &dp, NULL)) {
+                unsigned int val = cpu_data(i).ecache_size;
-                unsigned int val;
-                val = of_getintprop_default(dp, "ecache-size", def);
+                if (val && val < smallest)
-                if (val < smallest)
                        smallest = val;
-                instance++;
        }
        /* Any value less than 256K is nonsense.  */
@@ -1230,58 +1191,42 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
        int i;
        if (num_possible_cpus() > max_cpus) {
-                int instance, mid;
+                for_each_possible_cpu(i) {
+                        if (i != boot_cpu_id) {
-                instance = 0;
+                                cpu_clear(i, phys_cpu_present_map);
-                while (!cpu_find_by_instance(instance, NULL, &mid)) {
+                                cpu_clear(i, cpu_present_map);
-                        if (mid != boot_cpu_id) {
-                                cpu_clear(mid, phys_cpu_present_map);
-                                cpu_clear(mid, cpu_present_map);
                                if (num_possible_cpus() <= max_cpus)
                                        break;
                        }
-                        instance++;
                }
        }
-        for_each_possible_cpu(i) {
+        cpu_data(boot_cpu_id).udelay_val = loops_per_jiffy;
-                if (tlb_type == hypervisor) {
-                        int j;
-                        /* XXX get this mapping from machine description */
-                        for_each_possible_cpu(j) {
-                                if ((j >> 2) == (i >> 2))
-                                        cpu_set(j, cpu_sibling_map[i]);
-                        }
-                } else {
-                        cpu_set(i, cpu_sibling_map[i]);
-                }
-        }
-        smp_store_cpu_info(boot_cpu_id);
        smp_tune_scheduling();
 }
-/* Set this up early so that things like the scheduler can init
+void __devinit smp_prepare_boot_cpu(void)
- * properly.  We use the same cpu mask for both the present and
- * possible cpu map.
- */
-void __init smp_setup_cpu_possible_map(void)
 {
-        int instance, mid;
-        instance = 0;
-        while (!cpu_find_by_instance(instance, NULL, &mid)) {
-                if (mid < NR_CPUS) {
-                        cpu_set(mid, phys_cpu_present_map);
-                        cpu_set(mid, cpu_present_map);
-                }
-                instance++;
-        }
 }
-void __devinit smp_prepare_boot_cpu(void)
+void __devinit smp_fill_in_sib_core_maps(void)
 {
+        unsigned int i;
+        for_each_possible_cpu(i) {
+                unsigned int j;
+                if (cpu_data(i).core_id == 0) {
+                        cpu_set(i, cpu_sibling_map[i]);
+                        continue;
+                }
+                for_each_possible_cpu(j) {
+                        if (cpu_data(i).core_id ==
+                            cpu_data(j).core_id)
+                                cpu_set(j, cpu_sibling_map[i]);
+                }
+        }
 }
 int __cpuinit __cpu_up(unsigned int cpu)
@@ -1337,7 +1282,7 @@ unsigned long __per_cpu_shift __read_mostly;
 EXPORT_SYMBOL(__per_cpu_base);
 EXPORT_SYMBOL(__per_cpu_shift);
-void __init setup_per_cpu_areas(void)
+void __init real_setup_per_cpu_areas(void)
 {
        unsigned long goal, size, i;
        char *ptr;
diff --git a/arch/sparc64/kernel/sstate.c b/arch/sparc64/kernel/sstate.c
new file mode 100644
index 000000000000..5b6e75b7f052
--- /dev/null
+++ b/arch/sparc64/kernel/sstate.c
@@ -0,0 +1,104 @@
+/* sstate.c: System soft state support.
+ *
+ * Copyright (C) 2007 David S. Miller <davem@davemloft.net>
+ */
+#include <linux/kernel.h>
+#include <linux/notifier.h>
+#include <linux/init.h>
+#include <asm/hypervisor.h>
+#include <asm/sstate.h>
+#include <asm/oplib.h>
+#include <asm/head.h>
+#include <asm/io.h>
+static int hv_supports_soft_state;
+static unsigned long kimage_addr_to_ra(const char *p)
+{
+        unsigned long val = (unsigned long) p;
+        return kern_base + (val - KERNBASE);
+}
+static void do_set_sstate(unsigned long state, const char *msg)
+{
+        unsigned long err;
+        if (!hv_supports_soft_state)
+                return;
+        err = sun4v_mach_set_soft_state(state, kimage_addr_to_ra(msg));
+        if (err) {
+                printk(KERN_WARNING "SSTATE: Failed to set soft-state to "
+                       "state[%lx] msg[%s], err=%lu\n",
+                       state, msg, err);
+        }
+}
+static const char booting_msg[32] __attribute__((aligned(32))) =
+        "Linux booting";
+static const char running_msg[32] __attribute__((aligned(32))) =
+        "Linux running";
+static const char halting_msg[32] __attribute__((aligned(32))) =
+        "Linux halting";
+static const char poweroff_msg[32] __attribute__((aligned(32))) =
+        "Linux powering off";
+static const char rebooting_msg[32] __attribute__((aligned(32))) =
+        "Linux rebooting";
+static const char panicing_msg[32] __attribute__((aligned(32))) =
+        "Linux panicing";
+void sstate_booting(void)
+{
+        do_set_sstate(HV_SOFT_STATE_TRANSITION, booting_msg);
+}
+void sstate_running(void)
+{
+        do_set_sstate(HV_SOFT_STATE_NORMAL, running_msg);
+}
+void sstate_halt(void)
+{
+        do_set_sstate(HV_SOFT_STATE_TRANSITION, halting_msg);
+}
+void sstate_poweroff(void)
+{
+        do_set_sstate(HV_SOFT_STATE_TRANSITION, poweroff_msg);
+}
+void sstate_reboot(void)
+{
+        do_set_sstate(HV_SOFT_STATE_TRANSITION, rebooting_msg);
+}
+static int sstate_panic_event(struct notifier_block *n, unsigned long event, void *ptr)
+{
+        do_set_sstate(HV_SOFT_STATE_TRANSITION, panicing_msg);
+        return NOTIFY_DONE;
+}
+static struct notifier_block sstate_panic_block = {
+        .notifier_call  =       sstate_panic_event,
+        .priority       =       INT_MAX,
+};
+void __init sun4v_sstate_init(void)
+{
+        unsigned long major, minor;
+        major = 1;
+        minor = 0;
+        if (sun4v_hvapi_register(HV_GRP_SOFT_STATE, major, &minor))
+                return;
+        hv_supports_soft_state = 1;
+        prom_sun4v_guest_soft_state();
+        atomic_notifier_chain_register(&panic_notifier_list,
+                                       &sstate_panic_block);
+}
diff --git a/arch/sparc64/kernel/sun4v_ivec.S b/arch/sparc64/kernel/sun4v_ivec.S
index 405855dd886b..574bc248bca6 100644
--- a/arch/sparc64/kernel/sun4v_ivec.S
+++ b/arch/sparc64/kernel/sun4v_ivec.S
@@ -22,12 +22,12 @@ sun4v_cpu_mondo:
        be,pn   %xcc, sun4v_cpu_mondo_queue_empty
         nop
-        /* Get &trap_block[smp_processor_id()] into %g3.  */
+        /* Get &trap_block[smp_processor_id()] into %g4.  */
-        ldxa    [%g0] ASI_SCRATCHPAD, %g3
+        ldxa    [%g0] ASI_SCRATCHPAD, %g4
-        sub     %g3, TRAP_PER_CPU_FAULT_INFO, %g3
+        sub     %g4, TRAP_PER_CPU_FAULT_INFO, %g4
        /* Get CPU mondo queue base phys address into %g7.  */
-        ldx     [%g3 + TRAP_PER_CPU_CPU_MONDO_PA], %g7
+        ldx     [%g4 + TRAP_PER_CPU_CPU_MONDO_PA], %g7
        /* Now get the cross-call arguments and handler PC, same
         * layout as sun4u:
@@ -47,8 +47,7 @@ sun4v_cpu_mondo:
        add     %g2, 0x40 - 0x8 - 0x8, %g2
        /* Update queue head pointer.  */
-        sethi   %hi(8192 - 1), %g4
+        lduw    [%g4 + TRAP_PER_CPU_CPU_MONDO_QMASK], %g4
-        or      %g4, %lo(8192 - 1), %g4
        and     %g2, %g4, %g2
        mov     INTRQ_CPU_MONDO_HEAD, %g4
@@ -71,12 +70,12 @@ sun4v_dev_mondo:
        be,pn   %xcc, sun4v_dev_mondo_queue_empty
         nop
-        /* Get &trap_block[smp_processor_id()] into %g3.  */
+        /* Get &trap_block[smp_processor_id()] into %g4.  */
-        ldxa    [%g0] ASI_SCRATCHPAD, %g3
+        ldxa    [%g0] ASI_SCRATCHPAD, %g4
-        sub     %g3, TRAP_PER_CPU_FAULT_INFO, %g3
+        sub     %g4, TRAP_PER_CPU_FAULT_INFO, %g4
        /* Get DEV mondo queue base phys address into %g5.  */
-        ldx     [%g3 + TRAP_PER_CPU_DEV_MONDO_PA], %g5
+        ldx     [%g4 + TRAP_PER_CPU_DEV_MONDO_PA], %g5
        /* Load IVEC into %g3.  */
        ldxa    [%g5 + %g2] ASI_PHYS_USE_EC, %g3
@@ -90,8 +89,7 @@ sun4v_dev_mondo:
         */
        /* Update queue head pointer, this frees up some registers.  */
-        sethi   %hi(8192 - 1), %g4
+        lduw    [%g4 + TRAP_PER_CPU_DEV_MONDO_QMASK], %g4
-        or      %g4, %lo(8192 - 1), %g4
        and     %g2, %g4, %g2
        mov     INTRQ_DEVICE_MONDO_HEAD, %g4
@@ -143,6 +141,8 @@ sun4v_res_mondo:
        brnz,pn %g1, sun4v_res_mondo_queue_full
         nop
+        lduw    [%g3 + TRAP_PER_CPU_RESUM_QMASK], %g4
        /* Remember this entry's offset in %g1.  */
        mov     %g2, %g1
@@ -173,8 +173,6 @@ sun4v_res_mondo:
        add     %g2, 0x08, %g2
        /* Update queue head pointer.  */
-        sethi   %hi(8192 - 1), %g4
-        or      %g4, %lo(8192 - 1), %g4
        and     %g2, %g4, %g2
        mov     INTRQ_RESUM_MONDO_HEAD, %g4
@@ -254,6 +252,8 @@ sun4v_nonres_mondo:
        brnz,pn %g1, sun4v_nonres_mondo_queue_full
         nop
+        lduw    [%g3 + TRAP_PER_CPU_NONRESUM_QMASK], %g4
        /* Remember this entry's offset in %g1.  */
        mov     %g2, %g1
@@ -284,8 +284,6 @@ sun4v_nonres_mondo:
        add     %g2, 0x08, %g2
        /* Update queue head pointer.  */
-        sethi   %hi(8192 - 1), %g4
-        or      %g4, %lo(8192 - 1), %g4
        and     %g2, %g4, %g2
        mov     INTRQ_NONRESUM_MONDO_HEAD, %g4
diff --git a/arch/sparc64/kernel/time.c b/arch/sparc64/kernel/time.c
index 2d63d7689962..a31a0439244f 100644
--- a/arch/sparc64/kernel/time.c
+++ b/arch/sparc64/kernel/time.c
@@ -680,22 +680,14 @@ static int starfire_set_time(u32 val)
 static u32 hypervisor_get_time(void)
 {
-        register unsigned long func asm("%o5");
+        unsigned long ret, time;
-        register unsigned long arg0 asm("%o0");
-        register unsigned long arg1 asm("%o1");
        int retries = 10000;
 retry:
-        func = HV_FAST_TOD_GET;
+        ret = sun4v_tod_get(&time);
-        arg0 = 0;
+        if (ret == HV_EOK)
-        arg1 = 0;
+                return time;
-        __asm__ __volatile__("ta        %6"
+        if (ret == HV_EWOULDBLOCK) {
-                             : "=&r" (func), "=&r" (arg0), "=&r" (arg1)
-                             : "0" (func), "1" (arg0), "2" (arg1),
-                               "i" (HV_FAST_TRAP));
-        if (arg0 == HV_EOK)
-                return arg1;
-        if (arg0 == HV_EWOULDBLOCK) {
                if (--retries > 0) {
                        udelay(100);
                        goto retry;
@@ -709,20 +701,14 @@ retry:
 static int hypervisor_set_time(u32 secs)
 {
-        register unsigned long func asm("%o5");
+        unsigned long ret;
-        register unsigned long arg0 asm("%o0");
        int retries = 10000;
 retry:
-        func = HV_FAST_TOD_SET;
+        ret = sun4v_tod_set(secs);
-        arg0 = secs;
+        if (ret == HV_EOK)
-        __asm__ __volatile__("ta        %4"
-                             : "=&r" (func), "=&r" (arg0)
-                             : "0" (func), "1" (arg0),
-                               "i" (HV_FAST_TRAP));
-        if (arg0 == HV_EOK)
                return 0;
-        if (arg0 == HV_EWOULDBLOCK) {
+        if (ret == HV_EWOULDBLOCK) {
                if (--retries > 0) {
                        udelay(100);
                        goto retry;
@@ -862,7 +848,6 @@ fs_initcall(clock_init);
 static unsigned long sparc64_init_timers(void)
 {
        struct device_node *dp;
-        struct property *prop;
        unsigned long clock;
 #ifdef CONFIG_SMP
        extern void smp_tick_init(void);
@@ -879,17 +864,15 @@ static unsigned long sparc64_init_timers(void)
                if (manuf == 0x17 && impl == 0x13) {
                        /* Hummingbird, aka Ultra-IIe */
                        tick_ops = &hbtick_operations;
-                        prop = of_find_property(dp, "stick-frequency", NULL);
+                        clock = of_getintprop_default(dp, "stick-frequency", 0);
                } else {
                        tick_ops = &tick_operations;
-                        cpu_find_by_instance(0, &dp, NULL);
+                        clock = local_cpu_data().clock_tick;
-                        prop = of_find_property(dp, "clock-frequency", NULL);
                }
        } else {
                tick_ops = &stick_operations;
-                prop = of_find_property(dp, "stick-frequency", NULL);
+                clock = of_getintprop_default(dp, "stick-frequency", 0);
        }
-        clock = *(unsigned int *) prop->value;
 #ifdef CONFIG_SMP
        smp_tick_init();
@@ -1365,6 +1348,7 @@ static int hypervisor_set_rtc_time(struct rtc_time *time)
        return hypervisor_set_time(seconds);
 }
+#ifdef CONFIG_PCI
 static void bq4802_get_rtc_time(struct rtc_time *time)
 {
        unsigned char val = readb(bq4802_regs + 0x0e);
@@ -1436,6 +1420,7 @@ static int bq4802_set_rtc_time(struct rtc_time *time)
        return 0;
 }
+#endif /* CONFIG_PCI */
 struct mini_rtc_ops {
        void (*get_rtc_time)(struct rtc_time *);
@@ -1452,10 +1437,12 @@ static struct mini_rtc_ops hypervisor_rtc_ops = {
        .set_rtc_time = hypervisor_set_rtc_time,
 };
+#ifdef CONFIG_PCI
 static struct mini_rtc_ops bq4802_rtc_ops = {
        .get_rtc_time = bq4802_get_rtc_time,
        .set_rtc_time = bq4802_set_rtc_time,
 };
+#endif /* CONFIG_PCI */
 static struct mini_rtc_ops *mini_rtc_ops;
@@ -1579,8 +1566,10 @@ static int __init rtc_mini_init(void)
                mini_rtc_ops = &hypervisor_rtc_ops;
        else if (this_is_starfire)
                mini_rtc_ops = &starfire_rtc_ops;
+#ifdef CONFIG_PCI
        else if (bq4802_regs)
                mini_rtc_ops = &bq4802_rtc_ops;
+#endif /* CONFIG_PCI */
        else
                return -ENODEV;
diff --git a/arch/sparc64/kernel/traps.c b/arch/sparc64/kernel/traps.c
index d0fde36395b4..00a9e3286c83 100644
--- a/arch/sparc64/kernel/traps.c
+++ b/arch/sparc64/kernel/traps.c
@@ -795,8 +795,7 @@ extern unsigned int cheetah_deferred_trap_vector[], cheetah_deferred_trap_vector
 void __init cheetah_ecache_flush_init(void)
 {
        unsigned long largest_size, smallest_linesize, order, ver;
-        struct device_node *dp;
+        int i, sz;
-        int i, instance, sz;
        /* Scan all cpu device tree nodes, note two values:
         * 1) largest E-cache size
@@ -805,18 +804,20 @@ void __init cheetah_ecache_flush_init(void)
        largest_size = 0UL;
        smallest_linesize = ~0UL;
-        instance = 0;
+        for (i = 0; i < NR_CPUS; i++) {
-        while (!cpu_find_by_instance(instance, &dp, NULL)) {
                unsigned long val;
-                val = of_getintprop_default(dp, "ecache-size",
+                val = cpu_data(i).ecache_size;
-                                            (2 * 1024 * 1024));
+                if (!val)
+                        continue;
                if (val > largest_size)
                        largest_size = val;
-                val = of_getintprop_default(dp, "ecache-line-size", 64);
+                val = cpu_data(i).ecache_line_size;
                if (val < smallest_linesize)
                        smallest_linesize = val;
-                instance++;
        }
        if (largest_size == 0UL || smallest_linesize == ~0UL) {
@@ -2564,7 +2565,15 @@ void __init trap_init(void)
            (TRAP_PER_CPU_TSB_HUGE_TEMP !=
             offsetof(struct trap_per_cpu, tsb_huge_temp)) ||
            (TRAP_PER_CPU_IRQ_WORKLIST !=
-             offsetof(struct trap_per_cpu, irq_worklist)))
+             offsetof(struct trap_per_cpu, irq_worklist)) ||
+            (TRAP_PER_CPU_CPU_MONDO_QMASK !=
+             offsetof(struct trap_per_cpu, cpu_mondo_qmask)) ||
+            (TRAP_PER_CPU_DEV_MONDO_QMASK !=
+             offsetof(struct trap_per_cpu, dev_mondo_qmask)) ||
+            (TRAP_PER_CPU_RESUM_QMASK !=
+             offsetof(struct trap_per_cpu, resum_qmask)) ||
+            (TRAP_PER_CPU_NONRESUM_QMASK !=
+             offsetof(struct trap_per_cpu, nonresum_qmask)))
                trap_per_cpu_offsets_are_bolixed_dave();
        if ((TSB_CONFIG_TSB !=
diff --git a/arch/sparc64/kernel/vmlinux.lds.S b/arch/sparc64/kernel/vmlinux.lds.S
index fb648de18a8d..3ad10f3027e4 100644
--- a/arch/sparc64/kernel/vmlinux.lds.S
+++ b/arch/sparc64/kernel/vmlinux.lds.S
@@ -1,5 +1,6 @@
 /* ld script to make UltraLinux kernel */
+#include <asm/page.h>
 #include <asm-generic/vmlinux.lds.h>
 OUTPUT_FORMAT("elf64-sparc", "elf64-sparc", "elf64-sparc")
@@ -23,7 +24,7 @@ SECTIONS
  _etext = .;
  PROVIDE (etext = .);
-  RODATA
+  RO_DATA(PAGE_SIZE)
  .data    :
  {
@@ -44,7 +45,7 @@ SECTIONS
  __ex_table : { *(__ex_table) }
  __stop___ex_table = .;
-  . = ALIGN(8192);
+  . = ALIGN(PAGE_SIZE);
  __init_begin = .;
  .init.text : { 
        _sinittext = .;
@@ -83,17 +84,17 @@ SECTIONS
  __sun4v_2insn_patch_end = .;
 #ifdef CONFIG_BLK_DEV_INITRD
-  . = ALIGN(8192); 
+  . = ALIGN(PAGE_SIZE);
  __initramfs_start = .;
  .init.ramfs : { *(.init.ramfs) }
  __initramfs_end = .;
 #endif
-  . = ALIGN(8192);
+  . = ALIGN(PAGE_SIZE);
  __per_cpu_start = .;
  .data.percpu  : { *(.data.percpu) }
  __per_cpu_end = .;
-  . = ALIGN(8192);
+  . = ALIGN(PAGE_SIZE);
  __init_end = .;
  __bss_start = .;
  .sbss      : { *(.sbss) *(.scommon) }
diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c
index 6e5b01d779d2..3010227fe243 100644
--- a/arch/sparc64/mm/init.c
+++ b/arch/sparc64/mm/init.c
@@ -23,6 +23,7 @@
 #include <linux/kprobes.h>
 #include <linux/cache.h>
 #include <linux/sort.h>
+#include <linux/percpu.h>
 #include <asm/head.h>
 #include <asm/system.h>
@@ -43,8 +44,8 @@
 #include <asm/tsb.h>
 #include <asm/hypervisor.h>
 #include <asm/prom.h>
+#include <asm/sstate.h>
-extern void device_scan(void);
+#include <asm/mdesc.h>
 #define MAX_PHYS_ADDRESS        (1UL << 42UL)
 #define KPTE_BITMAP_CHUNK_SZ    (256UL * 1024UL * 1024UL)
@@ -60,8 +61,11 @@ unsigned long kern_linear_pte_xor[2] __read_mostly;
 unsigned long kpte_linear_bitmap[KPTE_BITMAP_BYTES / sizeof(unsigned long)];
 #ifndef CONFIG_DEBUG_PAGEALLOC
-/* A special kernel TSB for 4MB and 256MB linear mappings.  */
+/* A special kernel TSB for 4MB and 256MB linear mappings.
-struct tsb swapper_4m_tsb[KERNEL_TSB4M_NENTRIES];
+ * Space is allocated for this right after the trap table
+ * in arch/sparc64/kernel/head.S
+ */
+extern struct tsb swapper_4m_tsb[KERNEL_TSB4M_NENTRIES];
 #endif
 #define MAX_BANKS       32
@@ -190,12 +194,9 @@ inline void flush_dcache_page_impl(struct page *page)
 }
 #define PG_dcache_dirty         PG_arch_1
-#define PG_dcache_cpu_shift     24UL
+#define PG_dcache_cpu_shift     32UL
-#define PG_dcache_cpu_mask      (256UL - 1UL)
+#define PG_dcache_cpu_mask      \
+        ((1UL<<ilog2(roundup_pow_of_two(NR_CPUS)))-1UL)
-#if NR_CPUS > 256
-#error D-cache dirty tracking and thread_info->cpu need fixing for > 256 cpus
-#endif
 #define dcache_dirty_cpu(page) \
        (((page)->flags >> PG_dcache_cpu_shift) & PG_dcache_cpu_mask)
@@ -557,26 +558,11 @@ static void __init hypervisor_tlb_lock(unsigned long vaddr,
                                       unsigned long pte,
                                       unsigned long mmu)
 {
-        register unsigned long func asm("%o5");
+        unsigned long ret = sun4v_mmu_map_perm_addr(vaddr, 0, pte, mmu);
-        register unsigned long arg0 asm("%o0");
-        register unsigned long arg1 asm("%o1");
+        if (ret != 0) {
-        register unsigned long arg2 asm("%o2");
-        register unsigned long arg3 asm("%o3");
-        func = HV_FAST_MMU_MAP_PERM_ADDR;
-        arg0 = vaddr;
-        arg1 = 0;
-        arg2 = pte;
-        arg3 = mmu;
-        __asm__ __volatile__("ta        0x80"
-                             : "=&r" (func), "=&r" (arg0),
-                               "=&r" (arg1), "=&r" (arg2),
-                               "=&r" (arg3)
-                             : "0" (func), "1" (arg0), "2" (arg1),
-                               "3" (arg2), "4" (arg3));
-        if (arg0 != 0) {
                prom_printf("hypervisor_tlb_lock[%lx:%lx:%lx:%lx]: "
-                            "errors with %lx\n", vaddr, 0, pte, mmu, arg0);
+                            "errors with %lx\n", vaddr, 0, pte, mmu, ret);
                prom_halt();
        }
 }
@@ -1313,20 +1299,16 @@ static void __init sun4v_ktsb_init(void)
 void __cpuinit sun4v_ktsb_register(void)
 {
-        register unsigned long func asm("%o5");
+        unsigned long pa, ret;
-        register unsigned long arg0 asm("%o0");
-        register unsigned long arg1 asm("%o1");
-        unsigned long pa;
        pa = kern_base + ((unsigned long)&ktsb_descr[0] - KERNBASE);
-        func = HV_FAST_MMU_TSB_CTX0;
+        ret = sun4v_mmu_tsb_ctx0(NUM_KTSB_DESCR, pa);
-        arg0 = NUM_KTSB_DESCR;
+        if (ret != 0) {
-        arg1 = pa;
+                prom_printf("hypervisor_mmu_tsb_ctx0[%lx]: "
-        __asm__ __volatile__("ta        %6"
+                            "errors with %lx\n", pa, ret);
-                             : "=&r" (func), "=&r" (arg0), "=&r" (arg1)
+                prom_halt();
-                             : "0" (func), "1" (arg0), "2" (arg1),
+        }
-                               "i" (HV_FAST_TRAP));
 }
 /* paging_init() sets up the page tables */
@@ -1334,6 +1316,9 @@ void __cpuinit sun4v_ktsb_register(void)
 extern void cheetah_ecache_flush_init(void);
 extern void sun4v_patch_tlb_handlers(void);
+extern void cpu_probe(void);
+extern void central_probe(void);
 static unsigned long last_valid_pfn;
 pgd_t swapper_pg_dir[2048];
@@ -1345,9 +1330,24 @@ void __init paging_init(void)
        unsigned long end_pfn, pages_avail, shift, phys_base;
        unsigned long real_end, i;
+        /* These build time checkes make sure that the dcache_dirty_cpu()
+         * page->flags usage will work.
+         *
+         * When a page gets marked as dcache-dirty, we store the
+         * cpu number starting at bit 32 in the page->flags.  Also,
+         * functions like clear_dcache_dirty_cpu use the cpu mask
+         * in 13-bit signed-immediate instruction fields.
+         */
+        BUILD_BUG_ON(FLAGS_RESERVED != 32);
+        BUILD_BUG_ON(SECTIONS_WIDTH + NODES_WIDTH + ZONES_WIDTH +
+                     ilog2(roundup_pow_of_two(NR_CPUS)) > FLAGS_RESERVED);
+        BUILD_BUG_ON(NR_CPUS > 4096);
        kern_base = (prom_boot_mapping_phys_low >> 22UL) << 22UL;
        kern_size = (unsigned long)&_end - (unsigned long)KERNBASE;
+        sstate_booting();
        /* Invalidate both kernel TSBs.  */
        memset(swapper_tsb, 0x40, sizeof(swapper_tsb));
 #ifndef CONFIG_DEBUG_PAGEALLOC
@@ -1416,8 +1416,13 @@ void __init paging_init(void)
        kernel_physical_mapping_init();
+        real_setup_per_cpu_areas();
        prom_build_devicetree();
+        if (tlb_type == hypervisor)
+                sun4v_mdesc_init();
        {
                unsigned long zones_size[MAX_NR_ZONES];
                unsigned long zholes_size[MAX_NR_ZONES];
@@ -1434,7 +1439,10 @@ void __init paging_init(void)
                                    zholes_size);
        }
-        device_scan();
+        prom_printf("Booting Linux...\n");
+        central_probe();
+        cpu_probe();
 }
 static void __init taint_real_pages(void)
diff --git a/arch/sparc64/prom/misc.c b/arch/sparc64/prom/misc.c
index 0b4213720d43..f3e0c14e9eef 100644
--- a/arch/sparc64/prom/misc.c
+++ b/arch/sparc64/prom/misc.c
@@ -15,6 +15,25 @@
 #include <asm/oplib.h>
 #include <asm/system.h>
+int prom_service_exists(const char *service_name)
+{
+        int err = p1275_cmd("test", P1275_ARG(0, P1275_ARG_IN_STRING) |
+                            P1275_INOUT(1, 1), service_name);
+        if (err)
+                return 0;
+        return 1;
+}
+void prom_sun4v_guest_soft_state(void)
+{
+        const char *svc = "SUNW,soft-state-supported";
+        if (!prom_service_exists(svc))
+                return;
+        p1275_cmd(svc, P1275_INOUT(0, 0));
+}
 /* Reset and reboot the machine with the command 'bcommand'. */
 void prom_reboot(const char *bcommand)
 {
diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index 1336da8bdee1..1ad5111aec38 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -761,3 +761,9 @@ int in_gate_area_no_task(unsigned long addr)
 {
        return (addr >= VSYSCALL_START) && (addr < VSYSCALL_END);
 }
+void *alloc_bootmem_high_node(pg_data_t *pgdat, unsigned long size)
+{
+        return __alloc_bootmem_core(pgdat->bdata, size,
+                        SMP_CACHE_BYTES, (4UL*1024*1024*1024), 0);
+}