257 files changed, 4615 insertions, 2208 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,
-          DEC 2000 model 500
+          Jensen              DECpc 150, DEC 2000 models 300, 500
           LX164               AlphaPC164-LX
           Lynx                AS 2100A
-          Miata               Personal Workstation 433a, 433au, 500a,
-          500au, 600a, or 600au
+          Miata               Personal Workstation 433/500/600 a/au
           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
-          Takara              Takara
-          Titan               AlphaServer ES45 / DS25
+          Shark               DS 20L
+          Takara              Takara (OEM)
+          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)"
-        range 2 64
+        int "Maximum number of CPUs (2-32)"
+        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
 
-/*
- * Externally-visible vga hose bases
- */
-unsigned long __vga_hose_io_base = 0;   /* base for default hose */
-unsigned long __vga_hose_mem_base = 0;  /* base for default hose */
+struct pci_controller *pci_vga_hose;
+static struct resource alpha_vga = {
+        .name   = "alpha-vga+",
+        .start  = 0x3C0,
+        .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;
-                else hose = sel_func(hose, dev->sysdata);
+                if (!hose)
+                        hose = dev->sysdata;
+                else
+                        hose = sel_func(hose, dev->sysdata);
         }
 
-        /* Did we already inititialize the correct one? */
-        if (conswitchp == &vga_con &&
-            __vga_hose_io_base == hose->io_space->start &&
-            __vga_hose_mem_base == hose->mem_space->start)
+        /* Did we already initialize the correct one? Is there one? */
+        if (!hose || (conswitchp == &vga_con && pci_vga_hose == hose))
                 return;
 
-        /* Set the VGA hose and init the new console */
-        set_vga_hose(hose);
+        /* Create a new VGA ioport resource WRT the hose it is on. */
+        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
-        if (pci_vga_hose && __marvel_is_mem_vga(addr)) {
-                addr += pci_vga_hose->mem_space->start;
-        }
-#endif
+        FIXUP_MEMADDR_VGA(addr);
 
         /*
          * 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))
-                ;
-#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;
+        FIXUP_IOADDR_VGA(addr);
         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 (pchip1_present) {
+        if (titan_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 && __titan_is_mem_vga(addr)) {
+        if (pci_vga_hose && __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, $18
-        mov     $31, $16
+        mov     $sp, $17
+        mov     $5, $18
         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 *,
-                     unsigned long, unsigned long);
+static void do_signal(struct pt_regs *, struct switch_stack *,
+                      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;
-                schedule();
-                if (do_signal(&oldset, regs, sw, 0, 0))
-                        return -EINTR;
-        }
+        current->state = TASK_INTERRUPTIBLE;
+        schedule();
+        set_thread_flag(TIF_RESTORE_SIGMASK);
+        return -ERESTARTNOHAND;
 }
 
 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;
-                schedule();
-                if (do_signal(&oldset, regs, sw, 0, 0))
-                        return -EINTR;
-        }
+        current->state = TASK_INTERRUPTIBLE;
+        schedule();
+        set_thread_flag(TIF_RESTORE_SIGMASK);
+        return -ERESTARTNOHAND;
 }
 
 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)
-                ka->sa.sa_handler = SIG_DFL;
+        if (ret == 0) {
+                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);
-        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);
+        return ret;
 }
 
 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
-do_signal(sigset_t *oldset, struct pt_regs * regs, struct switch_stack * sw,
+static void
+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);
-                handle_signal(signr, &ka, &info, oldset, regs, sw);
+                if (r0)
+                        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,
-                 struct switch_stack *sw, unsigned long r0,
-                 unsigned long r19, unsigned long thread_info_flags)
+do_notify_resume(struct pt_regs *regs, struct switch_stack *sw,
+                 unsigned long thread_info_flags,
+                 unsigned long r0, unsigned long r19)
 {
-        if (thread_info_flags & _TIF_SIGPENDING)
-                do_signal(oldset, regs, sw, r0, r19);
+        if (thread_info_flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK))
+                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/boot/compressed/head.S b/arch/arm/boot/compressed/head.S
index 2568d311be21..23348e9561b9 100644
--- a/arch/arm/boot/compressed/head.S
+++ b/arch/arm/boot/compressed/head.S
@@ -247,7 +247,7 @@ not_relocated:	mov	r0, #0
                 mov     r3, r7
                 bl      decompress_kernel
 
-                add     r0, r0, #127
+                add     r0, r0, #127 + 128      @ alignment + stack
                 bic     r0, r0, #127            @ align the kernel length
 /*
  * r0     = decompressed kernel length
@@ -269,6 +269,7 @@ not_relocated:	mov	r0, #0
                 stmia   r1!, {r9 - r14}
                 cmp     r2, r3
                 blo     1b
+                add     sp, r1, #128            @ relocate the stack
 
                 bl      cache_clean_flush
                 add     pc, r5, r0              @ call relocation code
@@ -476,6 +477,7 @@ __common_mmu_cache_on:
  */
                 .align  5
 reloc_start:    add     r9, r5, r0
+                sub     r9, r9, #128            @ do not copy the stack
                 debug_reloc_start
                 mov     r1, r4
 1:
@@ -486,6 +488,7 @@ reloc_start:	add	r9, r5, r0
 
                 cmp     r5, r9
                 blo     1b
+                add     sp, r1, #128            @ relocate the stack
                 debug_reloc_end
 
 call_kernel:    bl      cache_clean_flush
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-at91/clock.c b/arch/arm/mach-at91/clock.c
index 06c9a0507d0d..848efb2a4ebf 100644
--- a/arch/arm/mach-at91/clock.c
+++ b/arch/arm/mach-at91/clock.c
@@ -364,19 +364,14 @@ static int at91_clk_show(struct seq_file *s, void *unused)
 {
         u32             scsr, pcsr, sr;
         struct clk      *clk;
-        unsigned        i;
 
         seq_printf(s, "SCSR = %8x\n", scsr = at91_sys_read(AT91_PMC_SCSR));
         seq_printf(s, "PCSR = %8x\n", pcsr = at91_sys_read(AT91_PMC_PCSR));
-
         seq_printf(s, "MOR  = %8x\n", at91_sys_read(AT91_CKGR_MOR));
         seq_printf(s, "MCFR = %8x\n", at91_sys_read(AT91_CKGR_MCFR));
         seq_printf(s, "PLLA = %8x\n", at91_sys_read(AT91_CKGR_PLLAR));
         seq_printf(s, "PLLB = %8x\n", at91_sys_read(AT91_CKGR_PLLBR));
-
         seq_printf(s, "MCKR = %8x\n", at91_sys_read(AT91_PMC_MCKR));
-        for (i = 0; i < 4; i++)
-                seq_printf(s, "PCK%d = %8x\n", i, at91_sys_read(AT91_PMC_PCKR(i)));
         seq_printf(s, "SR   = %8x\n", sr = at91_sys_read(AT91_PMC_SR));
 
         seq_printf(s, "\n");
diff --git a/arch/arm/mach-at91/pm.c b/arch/arm/mach-at91/pm.c
index ff8db29e989e..47ff676aca5f 100644
--- a/arch/arm/mach-at91/pm.c
+++ b/arch/arm/mach-at91/pm.c
@@ -76,12 +76,11 @@ static int at91_pm_verify_clocks(void)
                         pr_debug("AT91: PM - Suspend-to-RAM with USB still active\n");
                         return 0;
                 }
-        } else if (cpu_is_at91sam9260()) {
-#warning "Check SAM9260 USB clocks"
-        } else if (cpu_is_at91sam9261()) {
-#warning "Check SAM9261 USB clocks"
-        } else if (cpu_is_at91sam9263()) {
-#warning "Check SAM9263 USB clocks"
+        } else if (cpu_is_at91sam9260() || cpu_is_at91sam9261() || cpu_is_at91sam9263()) {
+                if ((scsr & (AT91SAM926x_PMC_UHP | AT91SAM926x_PMC_UDP)) != 0) {
+                        pr_debug("AT91: PM - Suspend-to-RAM with USB still active\n");
+                        return 0;
+                }
         }
 
 #ifdef CONFIG_AT91_PROGRAMMABLE_CLOCKS
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,
-                .end            = NSLU2_LED_GRN,
+                .start          = NSLU2_LED_GRN_GPIO,
+                .end            = NSLU2_LED_GRN_GPIO,
                 .flags          = IXP4XX_GPIO_HIGH,
         },
         {
                 .name           = "status", /* red led */
-                .start          = NSLU2_LED_RED,
-                .end            = NSLU2_LED_RED,
+                .start          = NSLU2_LED_RED_GPIO,
+                .end            = NSLU2_LED_RED_GPIO,
                 .flags          = IXP4XX_GPIO_HIGH,
         },
         {
                 .name           = "disk-1",
-                .start          = NSLU2_LED_DISK1,
-                .end            = NSLU2_LED_DISK1,
+                .start          = NSLU2_LED_DISK1_GPIO,
+                .end            = NSLU2_LED_DISK1_GPIO,
                 .flags          = IXP4XX_GPIO_LOW,
         },
         {
                 .name           = "disk-2",
-                .start          = NSLU2_LED_DISK2,
-                .end            = NSLU2_LED_DISK2,
+                .start          = NSLU2_LED_DISK2_GPIO,
+                .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-s3c2440/mach-anubis.c b/arch/arm/mach-s3c2440/mach-anubis.c
index b5d387ef37e1..bff7ddd06a52 100644
--- a/arch/arm/mach-s3c2440/mach-anubis.c
+++ b/arch/arm/mach-s3c2440/mach-anubis.c
@@ -76,8 +76,8 @@ static struct map_desc anubis_iodesc[] __initdata = {
         .length         = SZ_4K,
         .type           = MT_DEVICE,
   }, {
-        .virtual        = (u32)ANUBIS_VA_CTRL2,
-        .pfn            = __phys_to_pfn(ANUBIS_PA_CTRL2),
+        .virtual        = (u32)ANUBIS_VA_IDREG,
+        .pfn            = __phys_to_pfn(ANUBIS_PA_IDREG),
         .length         = SZ_4K,
         .type           = MT_DEVICE,
   },
diff --git a/arch/arm/mach-s3c2440/mach-osiris.c b/arch/arm/mach-s3c2440/mach-osiris.c
index 4d6c7a574c1a..15811601f03d 100644
--- a/arch/arm/mach-s3c2440/mach-osiris.c
+++ b/arch/arm/mach-s3c2440/mach-osiris.c
@@ -16,6 +16,7 @@
 #include <linux/timer.h>
 #include <linux/init.h>
 #include <linux/device.h>
+#include <linux/sysdev.h>
 #include <linux/serial_core.h>
 
 #include <asm/mach/arch.h>
@@ -65,6 +66,11 @@ static struct map_desc osiris_iodesc[] __initdata = {
   /* CPLD control registers */
 
   {
+          .virtual      = (u32)OSIRIS_VA_CTRL0,
+          .pfn          = __phys_to_pfn(OSIRIS_PA_CTRL0),
+          .length       = SZ_16K,
+          .type         = MT_DEVICE,
+  }, {
           .virtual      = (u32)OSIRIS_VA_CTRL1,
           .pfn          = __phys_to_pfn(OSIRIS_PA_CTRL1),
           .length       = SZ_16K,
@@ -74,6 +80,11 @@ static struct map_desc osiris_iodesc[] __initdata = {
           .pfn          = __phys_to_pfn(OSIRIS_PA_CTRL2),
           .length       = SZ_16K,
           .type         = MT_DEVICE,
+  }, {
+          .virtual      = (u32)OSIRIS_VA_IDREG,
+          .pfn          = __phys_to_pfn(OSIRIS_PA_IDREG),
+          .length       = SZ_16K,
+          .type         = MT_DEVICE,
   },
 };
 
@@ -195,13 +206,13 @@ static void osiris_nand_select(struct s3c2410_nand_set *set, int slot)
         pr_debug("osiris_nand: selecting slot %d (set %p,%p)\n",
                  slot, set, set->nr_map);
 
-        tmp = __raw_readb(OSIRIS_VA_CTRL1);
-        tmp &= ~OSIRIS_CTRL1_NANDSEL;
+        tmp = __raw_readb(OSIRIS_VA_CTRL0);
+        tmp &= ~OSIRIS_CTRL0_NANDSEL;
         tmp |= slot;
 
-        pr_debug("osiris_nand: ctrl1 now %02x\n", tmp);
+        pr_debug("osiris_nand: ctrl0 now %02x\n", tmp);
 
-        __raw_writeb(tmp, OSIRIS_VA_CTRL1);
+        __raw_writeb(tmp, OSIRIS_VA_CTRL0);
 }
 
 static struct s3c2410_platform_nand osiris_nand_info = {
@@ -235,10 +246,45 @@ static struct platform_device osiris_pcmcia = {
         .resource       = osiris_pcmcia_resource,
 };
 
+/* Osiris power management device */
+
+#ifdef CONFIG_PM
+static unsigned char pm_osiris_ctrl0;
+
+static int osiris_pm_suspend(struct sys_device *sd, pm_message_t state)
+{
+        pm_osiris_ctrl0 = __raw_readb(OSIRIS_VA_CTRL0);
+        return 0;
+}
+
+static int osiris_pm_resume(struct sys_device *sd)
+{
+        if (pm_osiris_ctrl0 & OSIRIS_CTRL0_FIX8)
+                __raw_writeb(OSIRIS_CTRL1_FIX8, OSIRIS_VA_CTRL1);
+
+        return 0;
+}
+
+#else
+#define osiris_pm_suspend NULL
+#define osiris_pm_resume NULL
+#endif
+
+static struct sysdev_class osiris_pm_sysclass = {
+        set_kset_name("mach-osiris"),
+        .suspend        = osiris_pm_suspend,
+        .resume         = osiris_pm_resume,
+};
+
+static struct sys_device osiris_pm_sysdev = {
+        .cls            = &osiris_pm_sysclass,
+};
+
 /* Standard Osiris devices */
 
 static struct platform_device *osiris_devices[] __initdata = {
         &s3c_device_i2c,
+        &s3c_device_wdt,
         &s3c_device_nand,
         &osiris_pcmcia,
 };
@@ -288,6 +334,9 @@ static void __init osiris_map_io(void)
 
 static void __init osiris_init(void)
 {
+        sysdev_class_register(&osiris_pm_sysclass);
+        sysdev_register(&osiris_pm_sysdev);
+
         platform_add_devices(osiris_devices, ARRAY_SIZE(osiris_devices));
 };
 
@@ -299,5 +348,6 @@ MACHINE_START(OSIRIS, "Simtec-OSIRIS")
         .map_io         = osiris_map_io,
         .init_machine   = osiris_init,
         .init_irq       = s3c24xx_init_irq,
+        .init_machine   = osiris_init,
         .timer          = &s3c24xx_timer,
 MACHINE_END
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..75bae067922d 100644
--- a/arch/arm/oprofile/op_model_mpcore.c
+++ b/arch/arm/oprofile/op_model_mpcore.c
@@ -200,8 +200,10 @@ static int em_call_function(int (*fn)(void))
         data.fn = fn;
         data.ret = 0;
 
+        preempt_disable();
         smp_call_function(em_func, &data, 1, 1);
         em_func(&data);
+        preempt_enable();
 
         return data.ret;
 }
@@ -257,8 +259,13 @@ static void em_stop(void)
  */
 static void em_route_irq(int irq, unsigned int cpu)
 {
-        irq_desc[irq].affinity = cpumask_of_cpu(cpu);
-        irq_desc[irq].chip->set_affinity(irq, cpumask_of_cpu(cpu));
+        struct irq_desc *desc = irq_desc + irq;
+        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/arm/vfp/entry.S b/arch/arm/vfp/entry.S
index ca2a5ad19ea6..806ce26d5243 100644
--- a/arch/arm/vfp/entry.S
+++ b/arch/arm/vfp/entry.S
@@ -29,6 +29,10 @@ do_vfp:
         add     r10, r10, #TI_VFPSTATE  @ r10 = workspace
         ldr     pc, [r4]                @ call VFP entry point
 
+ENTRY(vfp_null_entry)
+        mov     pc, lr
+ENDPROC(vfp_null_entry)
+
 .LCvfp:
         .word   vfp_vector
 
diff --git a/arch/arm/vfp/vfpmodule.c b/arch/arm/vfp/vfpmodule.c
index f1e5951dc721..1106b5f9cf19 100644
--- a/arch/arm/vfp/vfpmodule.c
+++ b/arch/arm/vfp/vfpmodule.c
@@ -26,8 +26,9 @@
  */
 void vfp_testing_entry(void);
 void vfp_support_entry(void);
+void vfp_null_entry(void);
 
-void (*vfp_vector)(void) = vfp_testing_entry;
+void (*vfp_vector)(void) = vfp_null_entry;
 union vfp_state *last_VFP_context[NR_CPUS];
 
 /*
@@ -321,8 +322,10 @@ static int __init vfp_init(void)
          * The handler is already setup to just log calls, so
          * we just need to read the VFPSID register.
          */
+        vfp_vector = vfp_testing_entry;
         vfpsid = fmrx(FPSID);
         barrier();
+        vfp_vector = vfp_null_entry;
 
         printk(KERN_INFO "VFP support v0.3: ");
         if (VFP_arch) {
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/Kconfig b/arch/i386/Kconfig
index c2d54b802232..8770a5d0b143 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -891,7 +891,7 @@ config PHYSICAL_ALIGN
           Don't change this unless you know what you are doing.
 
 config HOTPLUG_CPU
-        bool "Support for hot-pluggable CPUs (EXPERIMENTAL)"
+        bool "Support for suspend on SMP and hot-pluggable CPUs (EXPERIMENTAL)"
         depends on SMP && HOTPLUG && EXPERIMENTAL && !X86_VOYAGER
         ---help---
           Say Y here to experiment with turning CPUs off and on, and to
diff --git a/arch/i386/kernel/cpu/mtrr/main.c b/arch/i386/kernel/cpu/mtrr/main.c
index 1cf466df330a..7202b98aac4f 100644
--- a/arch/i386/kernel/cpu/mtrr/main.c
+++ b/arch/i386/kernel/cpu/mtrr/main.c
@@ -734,10 +734,13 @@ void mtrr_ap_init(void)
  */
 void mtrr_save_state(void)
 {
-        if (smp_processor_id() == 0)
+        int cpu = get_cpu();
+
+        if (cpu == 0)
                 mtrr_save_fixed_ranges(NULL);
         else
                 smp_call_function_single(0, mtrr_save_fixed_ranges, NULL, 1, 1);
+        put_cpu();
 }
 
 static int __init mtrr_init_finialize(void)
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/mach-generic/bigsmp.c b/arch/i386/mach-generic/bigsmp.c
index e932d3485ae2..58a477baec30 100644
--- a/arch/i386/mach-generic/bigsmp.c
+++ b/arch/i386/mach-generic/bigsmp.c
@@ -21,7 +21,7 @@
 
 static int dmi_bigsmp; /* can be set by dmi scanners */
 
-static __init int hp_ht_bigsmp(struct dmi_system_id *d)
+static int hp_ht_bigsmp(struct dmi_system_id *d)
 {
 #ifdef CONFIG_X86_GENERICARCH
         printk(KERN_NOTICE "%s detected: force use of apic=bigsmp\n", d->ident);
@@ -31,7 +31,7 @@ static __init int hp_ht_bigsmp(struct dmi_system_id *d)
 }
 
 
-static struct dmi_system_id __initdata bigsmp_dmi_table[] = {
+static struct dmi_system_id bigsmp_dmi_table[] = {
         { hp_ht_bigsmp, "HP ProLiant DL760 G2", {
                 DMI_MATCH(DMI_BIOS_VENDOR, "HP"),
                 DMI_MATCH(DMI_BIOS_VERSION, "P44-"),
@@ -45,7 +45,7 @@ static struct dmi_system_id __initdata bigsmp_dmi_table[] = {
 };
 
 
-static int __init probe_bigsmp(void)
+static int probe_bigsmp(void)
 { 
         if (def_to_bigsmp)
                 dmi_bigsmp = 1;
diff --git a/arch/i386/math-emu/fpu_entry.c b/arch/i386/math-emu/fpu_entry.c
index ddf8fa3bbd01..1853524c8b57 100644
--- a/arch/i386/math-emu/fpu_entry.c
+++ b/arch/i386/math-emu/fpu_entry.c
@@ -754,7 +754,7 @@ int save_i387_soft(void *s387, struct _fpstate __user * buf)
     return -1;
   if ( offset )
     if (__copy_to_user(d+other, (u_char *)&S387->st_space, offset))
-      return -1
+      return -1;
   RE_ENTRANT_CHECK_ON;
 
   return 1;
diff --git a/arch/i386/mm/fault.c b/arch/i386/mm/fault.c
index 29d7d61543a1..1ecb3e43b523 100644
--- a/arch/i386/mm/fault.c
+++ b/arch/i386/mm/fault.c
@@ -458,6 +458,11 @@ bad_area:
 bad_area_nosemaphore:
         /* User mode accesses just cause a SIGSEGV */
         if (error_code & 4) {
+                /*
+                 * It's possible to have interrupts off here.
+                 */
+                local_irq_enable();
+
                 /* 
                  * Valid to do another page fault here because this one came 
                  * from user space.
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)
-                        cpu_msrs[cpu] = cpu_msrs[0];
+                if (cpu != 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/ia64/kernel/acpi-processor.c b/arch/ia64/kernel/acpi-processor.c
index 4d4993a47e55..5a216c019924 100644
--- a/arch/ia64/kernel/acpi-processor.c
+++ b/arch/ia64/kernel/acpi-processor.c
@@ -44,7 +44,7 @@ static void init_intel_pdc(struct acpi_processor *pr)
 
         buf[0] = ACPI_PDC_REVISION_ID;
         buf[1] = 1;
-        buf[2] |= ACPI_PDC_EST_CAPABILITY_SMP;
+        buf[2] = ACPI_PDC_EST_CAPABILITY_SMP;
 
         obj->type = ACPI_TYPE_BUFFER;
         obj->buffer.length = 12;
diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c
index c4784494970e..103dd8edda71 100644
--- a/arch/ia64/kernel/acpi.c
+++ b/arch/ia64/kernel/acpi.c
@@ -67,7 +67,8 @@ EXPORT_SYMBOL(pm_power_off);
 unsigned int acpi_cpei_override;
 unsigned int acpi_cpei_phys_cpuid;
 
-const char *acpi_get_sysname(void)
+const char __init *
+acpi_get_sysname(void)
 {
 #ifdef CONFIG_IA64_GENERIC
         unsigned long rsdp_phys;
diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c
index d1c3ed9943e5..af73b8dfde28 100644
--- a/arch/ia64/kernel/process.c
+++ b/arch/ia64/kernel/process.c
@@ -763,6 +763,9 @@ get_wchan (struct task_struct *p)
         unsigned long ip;
         int count = 0;
 
+        if (!p || p == current || p->state == TASK_RUNNING)
+                return 0;
+
         /*
          * Note: p may not be a blocked task (it could be current or
          * another process running on some other CPU.  Rather than
@@ -773,6 +776,8 @@ get_wchan (struct task_struct *p)
          */
         unw_init_from_blocked_task(&info, p);
         do {
+                if (p->state == TASK_RUNNING)
+                        return 0;
                 if (unw_unwind(&info) < 0)
                         return 0;
                 unw_get_ip(&info, &ip);
diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c
index 542958079f1b..3c9d8e6089cf 100644
--- a/arch/ia64/kernel/smpboot.c
+++ b/arch/ia64/kernel/smpboot.c
@@ -370,7 +370,7 @@ smp_setup_percpu_timer (void)
 {
 }
 
-static void __devinit
+static void __cpuinit
 smp_callin (void)
 {
         int cpuid, phys_id, itc_master;
@@ -456,7 +456,7 @@ smp_callin (void)
 /*
  * Activate a secondary processor.  head.S calls this.
  */
-int __devinit
+int __cpuinit
 start_secondary (void *unused)
 {
         /* Early console may use I/O ports */
diff --git a/arch/ia64/kernel/unwind.c b/arch/ia64/kernel/unwind.c
index 7d3dd6cdafa4..b0b08b5f3eca 100644
--- a/arch/ia64/kernel/unwind.c
+++ b/arch/ia64/kernel/unwind.c
@@ -1860,7 +1860,7 @@ int
 unw_unwind (struct unw_frame_info *info)
 {
         unsigned long prev_ip, prev_sp, prev_bsp;
-        unsigned long ip, pr, num_regs;
+        unsigned long ip, pr, num_regs, rp_loc, pfs_loc;
         STAT(unsigned long start, flags;)
         int retval;
 
@@ -1870,14 +1870,16 @@ unw_unwind (struct unw_frame_info *info)
         prev_sp = info->sp;
         prev_bsp = info->bsp;
 
-        /* restore the ip */
-        if (!info->rp_loc) {
+        /* validate the return IP pointer */
+        rp_loc = (unsigned long) info->rp_loc;
+        if ((rp_loc < info->regstk.limit) || (rp_loc > info->regstk.top)) {
                 /* FIXME: should really be level 0 but it occurs too often. KAO */
                 UNW_DPRINT(1, "unwind.%s: failed to locate return link (ip=0x%lx)!\n",
                            __FUNCTION__, info->ip);
                 STAT(unw.stat.api.unwind_time += ia64_get_itc() - start; local_irq_restore(flags));
                 return -1;
         }
+        /* restore the ip */
         ip = info->ip = *info->rp_loc;
         if (ip < GATE_ADDR) {
                 UNW_DPRINT(2, "unwind.%s: reached user-space (ip=0x%lx)\n", __FUNCTION__, ip);
@@ -1885,12 +1887,14 @@ unw_unwind (struct unw_frame_info *info)
                 return -1;
         }
 
-        /* restore the cfm: */
-        if (!info->pfs_loc) {
+        /* validate the previous stack frame pointer */
+        pfs_loc = (unsigned long) info->pfs_loc;
+        if ((pfs_loc < info->regstk.limit) || (pfs_loc > info->regstk.top)) {
                 UNW_DPRINT(0, "unwind.%s: failed to locate ar.pfs!\n", __FUNCTION__);
                 STAT(unw.stat.api.unwind_time += ia64_get_itc() - start; local_irq_restore(flags));
                 return -1;
         }
+        /* restore the cfm: */
         info->cfm_loc = info->pfs_loc;
 
         /* restore the bsp: */
@@ -1992,13 +1996,16 @@ init_frame_info (struct unw_frame_info *info, struct task_struct *t,
         memset(info, 0, sizeof(*info));
 
         rbslimit = (unsigned long) t + IA64_RBS_OFFSET;
+        stklimit = (unsigned long) t + IA64_STK_OFFSET;
+
         rbstop   = sw->ar_bspstore;
-        if (rbstop - (unsigned long) t >= IA64_STK_OFFSET)
+        if (rbstop > stklimit || rbstop < rbslimit)
                 rbstop = rbslimit;
 
-        stklimit = (unsigned long) t + IA64_STK_OFFSET;
         if (stktop <= rbstop)
                 stktop = rbstop;
+        if (stktop > stklimit)
+                stktop = stklimit;
 
         info->regstk.limit = rbslimit;
         info->regstk.top   = rbstop;
diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c
index 3549f3b42592..73696b4a2eed 100644
--- a/arch/ia64/pci/pci.c
+++ b/arch/ia64/pci/pci.c
@@ -354,10 +354,13 @@ pci_acpi_scan_root(struct acpi_device *device, int domain, int bus)
 
         acpi_walk_resources(device->handle, METHOD_NAME__CRS, count_window,
                         &windows);
-        controller->window = kmalloc_node(sizeof(*controller->window) * windows,
-                        GFP_KERNEL, controller->node);
-        if (!controller->window)
-                goto out2;
+        if (windows) {
+                controller->window =
+                        kmalloc_node(sizeof(*controller->window) * windows,
+                                     GFP_KERNEL, controller->node);
+                if (!controller->window)
+                        goto out2;
+        }
 
         name = kmalloc(16, GFP_KERNEL);
         if (!name)
diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c
index a574fcd163dd..684b1c984a44 100644
--- a/arch/ia64/sn/kernel/setup.c
+++ b/arch/ia64/sn/kernel/setup.c
@@ -194,7 +194,7 @@ void __init early_sn_setup(void)
 }
 
 extern int platform_intr_list[];
-static int __initdata shub_1_1_found;
+static int __cpuinitdata shub_1_1_found;
 
 /*
  * sn_check_for_wars
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"
-        depends on ADVANCED && !SUN3
+        bool "Use one physical chunk of memory only" if 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
-        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);
+        paging_init();
 
+#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;
-    int free = 0, total = 0, reserved = 0, shared = 0;
-    int cached = 0;
-
-    printk("\nMem-info:\n");
-    show_free_areas();
-    printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
-    i = max_mapnr;
-    while (i-- > 0) {
-        total++;
-        if (PageReserved(mem_map+i))
-            reserved++;
-        else if (PageSwapCache(mem_map+i))
-            cached++;
-        else if (!page_count(mem_map+i))
-            free++;
-        else
-            shared += page_count(mem_map+i) - 1;
-    }
-    printk("%d pages of RAM\n",total);
-    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);
+        pg_data_t *pgdat;
+        int free = 0, total = 0, reserved = 0, shared = 0;
+        int cached = 0;
+        int i;
+
+        printk("\nMem-info:\n");
+        show_free_areas();
+        printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
+        for_each_online_pgdat(pgdat) {
+                for (i = 0; i < pgdat->node_spanned_pages; i++) {
+                        struct page *page = pgdat->node_mem_map + i;
+                        total++;
+                        if (PageReserved(page))
+                                reserved++;
+                        else if (PageSwapCache(page))
+                                cached++;
+                        else if (!page_count(page))
+                                free++;
+                        else
+                                shared += page_count(page) - 1;
+                }
+        }
+        printk("%d pages of RAM\n",total);
+        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 __init_begin, __init_end;
+extern char _text[], _etext[];
+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();
-
-        for (tmp = PAGE_OFFSET ; tmp < (unsigned long)high_memory; tmp += PAGE_SIZE) {
-                if (PageReserved(virt_to_page(tmp))) {
-                        if (tmp >= (unsigned long)&_text
-                            && tmp < (unsigned long)&_etext)
+        totalram_pages = num_physpages = 0;
+        for_each_online_pgdat(pgdat) {
+                num_physpages += pgdat->node_present_pages;
+
+                totalram_pages += free_all_bootmem_node(pgdat);
+                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
-                                 && tmp < (unsigned long) &__init_end)
+                        else if (addr >= __init_begin &&
+                                 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
-map_chunk (unsigned long addr, long size)
+static void __init map_node(int node)
 {
 #define PTRTREESIZE (256*1024)
 #define ROOTTREESIZE (32*1024*1024)
-        static unsigned long virtaddr = PAGE_OFFSET;
-        unsigned long physaddr;
+        unsigned long physaddr, virtaddr, size;
         pgd_t *pgd_dir;
         pmd_t *pmd_dir;
         pte_t *pte_dir;
 
-        physaddr = (addr | m68k_supervisor_cachemode |
-                    _PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_DIRTY);
+        size = m68k_memory[node].size;
+        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",
-                        kernel_pg_dir, availmem, start_mem, end_mem);
+                printk ("start of paging_init (%p, %lx)\n",
+                        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
-         * tables and thus modify availmem.
+         * address space. First initialize the bootmem allocator with
+         * 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++) {
-                mem_avail = map_chunk (m68k_memory[chunk].addr,
-                                       m68k_memory[chunk].size);
-
-        }
+        for (i = 1; i < m68k_num_memory; i++)
+                map_node(i);
 
         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 ?
-                                (mach_max_dma_address+1) : (unsigned long)high_memory);
-        zones_size[ZONE_NORMAL] = (unsigned long)high_memory - zones_size[0];
-
-        zones_size[ZONE_DMA] = (zones_size[ZONE_DMA] - PAGE_OFFSET) >> PAGE_SHIFT;
-        zones_size[ZONE_NORMAL] >>= PAGE_SHIFT;
-
-        free_area_init(zones_size);
+        for (i = 0; i < m68k_num_memory; i++) {
+                zones_size[ZONE_DMA] = m68k_memory[i].size >> PAGE_SHIFT;
+                free_area_init_node(i, pg_data_map + i, zones_size,
+                                    m68k_memory[i].addr >> PAGE_SHIFT, NULL);
+        }
 }
 
 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/m68knommu/platform/5307/timers.c b/arch/m68knommu/platform/5307/timers.c
index 92e58070b016..fb66eadd5896 100644
--- a/arch/m68knommu/platform/5307/timers.c
+++ b/arch/m68knommu/platform/5307/timers.c
@@ -62,10 +62,13 @@ void coldfire_tick(void)
 
 /***************************************************************************/
 
+static int ticks_per_intr;
+
 void coldfire_timer_init(irq_handler_t handler)
 {
         __raw_writew(MCFTIMER_TMR_DISABLE, TA(MCFTIMER_TMR));
-        __raw_writetrr(((MCF_BUSCLK / 16) / HZ), TA(MCFTIMER_TRR));
+        ticks_per_intr = (MCF_BUSCLK / 16) / HZ;
+        __raw_writetrr(ticks_per_intr - 1, TA(MCFTIMER_TRR));
         __raw_writew(MCFTIMER_TMR_ENORI | MCFTIMER_TMR_CLK16 |
                 MCFTIMER_TMR_RESTART | MCFTIMER_TMR_ENABLE, TA(MCFTIMER_TMR));
 
@@ -81,11 +84,10 @@ void coldfire_timer_init(irq_handler_t handler)
 
 unsigned long coldfire_timer_offset(void)
 {
-        unsigned long trr, tcn, offset;
+        unsigned long tcn, offset;
 
         tcn = __raw_readw(TA(MCFTIMER_TCN));
-        trr = __raw_readtrr(TA(MCFTIMER_TRR));
-        offset = (tcn * (1000000 / HZ)) / trr;
+        offset = ((tcn + 1) * (1000000 / HZ)) / ticks_per_intr;
 
         /* Check if we just wrapped the counters and maybe missed a tick */
         if ((offset < (1000000 / HZ / 2)) && mcf_timerirqpending(1))
diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 0f09412e1b7f..9528ee90640a 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -747,9 +747,9 @@ config EARLY_PRINTK
           to print messages very early in the bootup process.
 
           This is useful for kernel debugging when your machine crashes very
-          early before the console code is initialized. For normal operation
-          it is not recommended because it looks on some machines ugly and
-          oesn't cooperate with an X server. You should normally N here,
+          early before the console code is initialized. For normal operation,
+          it is not recommended because it looks ugly on some machines and
+          doesn't cooperate with an X server. You should normally say N here,
           unless you want to debug such a crash.
 
 config SYS_HAS_EARLY_PRINTK
diff --git a/arch/mips/emma2rh/markeins/setup.c b/arch/mips/emma2rh/markeins/setup.c
index b29a44739230..2f060e1ed36c 100644
--- a/arch/mips/emma2rh/markeins/setup.c
+++ b/arch/mips/emma2rh/markeins/setup.c
@@ -115,30 +115,6 @@ extern void markeins_irq_setup(void);
 
 static void inline __init markeins_sio_setup(void)
 {
-#ifdef CONFIG_KGDB_8250
-        struct uart_port emma_port;
-
-        memset(&emma_port, 0, sizeof(emma_port));
-
-        emma_port.flags =
-            UPF_BOOT_AUTOCONF | UPF_SKIP_TEST;
-        emma_port.iotype = UPIO_MEM;
-        emma_port.regshift = 4; /* I/O addresses are every 8 bytes */
-        emma_port.uartclk = 18544000;   /* Clock rate of the chip */
-
-        emma_port.line = 0;
-        emma_port.mapbase = KSEG1ADDR(EMMA2RH_PFUR0_BASE + 3);
-        emma_port.membase = (u8*)emma_port.mapbase;
-        early_serial_setup(&emma_port);
-
-        emma_port.line = 1;
-        emma_port.mapbase = KSEG1ADDR(EMMA2RH_PFUR1_BASE + 3);
-        emma_port.membase = (u8*)emma_port.mapbase;
-        early_serial_setup(&emma_port);
-
-        emma_port.irq = EMMA2RH_IRQ_PFUR1;
-        kgdb8250_add_port(1, &emma_port);
-#endif
 }
 
 void __init plat_mem_setup(void)
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/kernel/linux32.c b/arch/mips/kernel/linux32.c
index 37849edd0645..06e04da211d5 100644
--- a/arch/mips/kernel/linux32.c
+++ b/arch/mips/kernel/linux32.c
@@ -556,6 +556,16 @@ asmlinkage long sys32_sync_file_range(int fd, int __pad,
                         flags);
 }
 
+asmlinkage long sys32_fadvise64_64(int fd, int __pad,
+        unsigned long a2, unsigned long a3,
+        unsigned long a4, unsigned long a5,
+        int flags)
+{
+        return sys_fadvise64_64(fd,
+                        merge_64(a2, a3), merge_64(a4, a5),
+                        flags);
+}
+
 save_static_function(sys32_clone);
 __attribute_used__ noinline static int
 _sys32_clone(nabi_no_regargs struct pt_regs regs)
diff --git a/arch/mips/kernel/r4k_switch.S b/arch/mips/kernel/r4k_switch.S
index cc566cf12246..06729596812f 100644
--- a/arch/mips/kernel/r4k_switch.S
+++ b/arch/mips/kernel/r4k_switch.S
@@ -174,7 +174,7 @@ LEAF(_init_fpu)
         or      t0, t1
         mtc0    t0, CP0_STATUS
 #endif /* CONFIG_MIPS_MT_SMTC */
-        fpu_enable_hazard
+        enable_fpu_hazard
 
         li      t1, FPU_DEFAULT
         ctc1    t1, fcr31
diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S
index 0c9a9ff8cd25..ae985d1fcca1 100644
--- a/arch/mips/kernel/scall32-o32.S
+++ b/arch/mips/kernel/scall32-o32.S
@@ -657,7 +657,11 @@ einval:	li	v0, -EINVAL
         sys     sys_getcpu              3
         sys     sys_epoll_pwait         6
         sys     sys_ioprio_set          3
-        sys     sys_ioprio_get          2
+        sys     sys_ioprio_get          2       /* 4315 */
+        sys     sys_utimensat           4
+        sys     sys_signalfd            3
+        sys     sys_timerfd             4
+        sys     sys_eventfd             1
         .endm
 
         /* We pre-compute the number of _instruction_ bytes needed to
diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S
index 23f3b118f718..7bcd5a1a85f5 100644
--- a/arch/mips/kernel/scall64-64.S
+++ b/arch/mips/kernel/scall64-64.S
@@ -473,4 +473,8 @@ sys_call_table:
         PTR     sys_epoll_pwait
         PTR     sys_ioprio_set
         PTR     sys_ioprio_get
+        PTR     sys_utimensat                   /* 5275 */
+        PTR     sys_signalfd
+        PTR     sys_timerfd
+        PTR     sys_eventfd
         .size   sys_call_table,.-sys_call_table
diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S
index 6eac28337423..532a2f3b42fc 100644
--- a/arch/mips/kernel/scall64-n32.S
+++ b/arch/mips/kernel/scall64-n32.S
@@ -299,7 +299,7 @@ EXPORT(sysn32_call_table)
         PTR     sys_ni_syscall                  /* res. for afs_syscall */
         PTR     sys_ni_syscall                  /* res. for security */
         PTR     sys_gettid
-        PTR     sys32_readahead
+        PTR     sys_readahead
         PTR     sys_setxattr                    /* 6180 */
         PTR     sys_lsetxattr
         PTR     sys_fsetxattr
@@ -399,4 +399,8 @@ EXPORT(sysn32_call_table)
         PTR     compat_sys_epoll_pwait
         PTR     sys_ioprio_set
         PTR     sys_ioprio_get
+        PTR     compat_sys_utimensat
+        PTR     compat_sys_signalfd             /* 5280 */
+        PTR     compat_sys_timerfd
+        PTR     sys_eventfd
         .size   sysn32_call_table,.-sysn32_call_table
diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S
index 7e74b412a782..6bbe0f4ed8ba 100644
--- a/arch/mips/kernel/scall64-o32.S
+++ b/arch/mips/kernel/scall64-o32.S
@@ -459,7 +459,7 @@ sys_call_table:
         PTR     sys_remap_file_pages
         PTR     sys_set_tid_address
         PTR     sys_restart_syscall
-        PTR     sys_fadvise64_64
+        PTR     sys32_fadvise64_64
         PTR     compat_sys_statfs64             /* 4255 */
         PTR     compat_sys_fstatfs64
         PTR     compat_sys_timer_create
@@ -521,4 +521,8 @@ sys_call_table:
         PTR     compat_sys_epoll_pwait
         PTR     sys_ioprio_set
         PTR     sys_ioprio_get                  /* 4315 */
+        PTR     compat_sys_utimensat
+        PTR     compat_sys_signalfd
+        PTR     compat_sys_timerfd
+        PTR     sys_eventfd
         .size   sys_call_table,.-sys_call_table
diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
index c46e479c992b..67edfa7ed93a 100644
--- a/arch/mips/kernel/smp.c
+++ b/arch/mips/kernel/smp.c
@@ -68,7 +68,7 @@ extern ATTRIB_NORET void cpu_idle(void);
  * First C code run on the secondary CPUs after being started up by
  * the master.
  */
-asmlinkage void start_secondary(void)
+asmlinkage __cpuinit void start_secondary(void)
 {
         unsigned int cpu;
 
diff --git a/arch/mips/kernel/smtc-asm.S b/arch/mips/kernel/smtc-asm.S
index 921207c4a83c..20938a4cb52d 100644
--- a/arch/mips/kernel/smtc-asm.S
+++ b/arch/mips/kernel/smtc-asm.S
@@ -121,10 +121,7 @@ LEAF(self_ipi)
         subu    t1,sp,PT_SIZE
         sw      ra,PT_EPC(t1)
         sw      a0,PT_PADSLOT4(t1)
-        LONG_L  s0, TI_REGS($28)
-        LONG_S  sp, TI_REGS($28)
         la      t2,ipi_decode
-        LONG_S  s0, TI_REGS($28)
         sw      t2,PT_PADSLOT5(t1)
         /* Save pre-disable value of TCStatus */
         sw      t0,PT_TCSTATUS(t1)
diff --git a/arch/mips/kernel/smtc.c b/arch/mips/kernel/smtc.c
index b361edb83dc6..21eb5993a19f 100644
--- a/arch/mips/kernel/smtc.c
+++ b/arch/mips/kernel/smtc.c
@@ -611,12 +611,12 @@ void smtc_cpus_done(void)
 int setup_irq_smtc(unsigned int irq, struct irqaction * new,
                         unsigned long hwmask)
 {
+#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG
         unsigned int vpe = current_cpu_data.vpe_id;
 
-        irq_hwmask[irq] = hwmask;
-#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG
         vpemask[vpe][irq - MIPSCPU_INT_BASE] = 1;
 #endif
+        irq_hwmask[irq] = hwmask;
 
         return setup_irq(irq, new);
 }
diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c
index 200de027f354..a7a17eb9bfcd 100644
--- a/arch/mips/kernel/traps.c
+++ b/arch/mips/kernel/traps.c
@@ -11,6 +11,7 @@
  * Copyright (C) 2000, 01 MIPS Technologies, Inc.
  * Copyright (C) 2002, 2003, 2004, 2005  Maciej W. Rozycki
  */
+#include <linux/bug.h>
 #include <linux/init.h>
 #include <linux/mm.h>
 #include <linux/module.h>
@@ -927,12 +928,6 @@ asmlinkage void do_reserved(struct pt_regs *regs)
               (regs->cp0_cause & 0x7f) >> 2);
 }
 
-static asmlinkage void do_default_vi(void)
-{
-        show_regs(get_irq_regs());
-        panic("Caught unexpected vectored interrupt.");
-}
-
 /*
  * Some MIPS CPUs can enable/disable for cache parity detection, but do
  * it different ways.
@@ -1128,6 +1123,12 @@ void mips_srs_free(int set)
         clear_bit(set, &sr->sr_allocated);
 }
 
+static asmlinkage void do_default_vi(void)
+{
+        show_regs(get_irq_regs());
+        panic("Caught unexpected vectored interrupt.");
+}
+
 static void *set_vi_srs_handler(int n, vi_handler_t addr, int srs)
 {
         unsigned long handler;
@@ -1190,8 +1191,8 @@ static void *set_vi_srs_handler(int n, vi_handler_t addr, int srs)
 
                 memcpy (b, &except_vec_vi, handler_len);
 #ifdef CONFIG_MIPS_MT_SMTC
-                if (n > 7)
-                        printk("Vector index %d exceeds SMTC maximum\n", n);
+                BUG_ON(n > 7);  /* Vector index %d exceeds SMTC maximum. */
+
                 w = (u32 *)(b + mori_offset);
                 *w = (*w & 0xffff0000) | (0x100 << n);
 #endif /* CONFIG_MIPS_MT_SMTC */
@@ -1383,6 +1384,13 @@ void __init per_cpu_trap_init(void)
                 cpu_cache_init();
                 tlb_init();
 #ifdef CONFIG_MIPS_MT_SMTC
+        } else if (!secondaryTC) {
+                /*
+                 * First TC in non-boot VPE must do subset of tlb_init()
+                 * for MMU countrol registers.
+                 */
+                write_c0_pagemask(PM_DEFAULT_MASK);
+                write_c0_wired(0);
         }
 #endif /* CONFIG_MIPS_MT_SMTC */
 }
@@ -1531,8 +1539,7 @@ void __init trap_init(void)
         if (cpu_has_mipsmt)
                 set_except_vector(25, handle_mt);
 
-        if (cpu_has_dsp)
-                set_except_vector(26, handle_dsp);
+        set_except_vector(26, handle_dsp);
 
         if (cpu_has_vce)
                 /* Special exception: R4[04]00 uses also the divec space. */
diff --git a/arch/mips/mips-boards/atlas/atlas_int.c b/arch/mips/mips-boards/atlas/atlas_int.c
index dfa0acbd7fc2..9f49da95aacf 100644
--- a/arch/mips/mips-boards/atlas/atlas_int.c
+++ b/arch/mips/mips-boards/atlas/atlas_int.c
@@ -248,14 +248,13 @@ void __init arch_init_irq(void)
         case MIPS_REVISION_CORID_CORE_24K:
         case MIPS_REVISION_CORID_CORE_EMUL_MSC:
                 if (cpu_has_veic)
-                        init_msc_irqs (MSC01E_INT_BASE,
+                        init_msc_irqs (MSC01E_INT_BASE, MSC01E_INT_BASE,
                                        msc_eicirqmap, msc_nr_eicirqs);
                 else
-                        init_msc_irqs (MSC01C_INT_BASE,
+                        init_msc_irqs (MSC01E_INT_BASE, MSC01C_INT_BASE,
                                        msc_irqmap, msc_nr_irqs);
         }
 
-
         if (cpu_has_veic) {
                 set_vi_handler (MSC01E_INT_ATLAS, atlas_hw0_irqdispatch);
                 setup_irq (MSC01E_INT_BASE + MSC01E_INT_ATLAS, &atlasirq);
diff --git a/arch/mips/mips-boards/atlas/atlas_setup.c b/arch/mips/mips-boards/atlas/atlas_setup.c
index 0c6b0ce15028..1cc6ebbedfdd 100644
--- a/arch/mips/mips-boards/atlas/atlas_setup.c
+++ b/arch/mips/mips-boards/atlas/atlas_setup.c
@@ -48,6 +48,8 @@ const char *get_system_type(void)
         return "MIPS Atlas";
 }
 
+const char display_string[] = "        LINUX ON ATLAS       ";
+
 void __init plat_mem_setup(void)
 {
         mips_pcibios_init();
diff --git a/arch/mips/mips-boards/generic/display.c b/arch/mips/mips-boards/generic/display.c
index 548dbe5ce7c8..5d600054090a 100644
--- a/arch/mips/mips-boards/generic/display.c
+++ b/arch/mips/mips-boards/generic/display.c
@@ -19,9 +19,14 @@
  */
 
 #include <linux/compiler.h>
+#include <linux/timer.h>
 #include <asm/io.h>
 #include <asm/mips-boards/generic.h>
 
+extern const char display_string[];
+static unsigned int display_count;
+static unsigned int max_display_count;
+
 void mips_display_message(const char *str)
 {
         static unsigned int __iomem *display = NULL;
@@ -37,3 +42,22 @@ void mips_display_message(const char *str)
                          writel(' ', display + i);
         }
 }
+
+static void scroll_display_message(unsigned long data);
+static DEFINE_TIMER(mips_scroll_timer, scroll_display_message, HZ, 0);
+
+static void scroll_display_message(unsigned long data)
+{
+        mips_display_message(&display_string[display_count++]);
+        if (display_count == max_display_count)
+                display_count = 0;
+
+        mod_timer(&mips_scroll_timer, jiffies + HZ);
+}
+
+void mips_scroll_message(void)
+{
+        del_timer_sync(&mips_scroll_timer);
+        max_display_count = strlen(display_string) + 1 - 8;
+        mod_timer(&mips_scroll_timer, jiffies + 1);
+}
diff --git a/arch/mips/mips-boards/generic/time.c b/arch/mips/mips-boards/generic/time.c
index df2a2bd3aa5d..b41db9e7ab1f 100644
--- a/arch/mips/mips-boards/generic/time.c
+++ b/arch/mips/mips-boards/generic/time.c
@@ -53,37 +53,11 @@
 
 unsigned long cpu_khz;
 
-#if defined(CONFIG_MIPS_ATLAS)
-static char display_string[] = "        LINUX ON ATLAS       ";
-#endif
-#if defined(CONFIG_MIPS_MALTA)
-#if defined(CONFIG_MIPS_MT_SMTC)
-static char display_string[] = "       SMTC LINUX ON MALTA       ";
-#else
-static char display_string[] = "        LINUX ON MALTA       ";
-#endif /* CONFIG_MIPS_MT_SMTC */
-#endif
-#if defined(CONFIG_MIPS_SEAD)
-static char display_string[] = "        LINUX ON SEAD       ";
-#endif
-static unsigned int display_count;
-#define MAX_DISPLAY_COUNT (sizeof(display_string) - 8)
-
 #define CPUCTR_IMASKBIT (0x100 << MIPSCPU_INT_CPUCTR)
 
-static unsigned int timer_tick_count;
 static int mips_cpu_timer_irq;
 extern void smtc_timer_broadcast(int);
 
-static inline void scroll_display_message(void)
-{
-        if ((timer_tick_count++ % HZ) == 0) {
-                mips_display_message(&display_string[display_count++]);
-                if (display_count == MAX_DISPLAY_COUNT)
-                        display_count = 0;
-        }
-}
-
 static void mips_timer_dispatch(void)
 {
         do_IRQ(mips_cpu_timer_irq);
@@ -114,8 +88,6 @@ irqreturn_t mips_timer_interrupt(int irq, void *dev_id)
          *  the general MIPS timer_interrupt routine.
          */
 
-        int vpflags;
-
         /*
          * We could be here due to timer interrupt,
          * perf counter overflow, or both.
@@ -124,15 +96,6 @@ irqreturn_t mips_timer_interrupt(int irq, void *dev_id)
                 perf_irq();
 
         if (read_c0_cause() & (1 << 30)) {
-                /* If timer interrupt, make it de-assert */
-                write_c0_compare (read_c0_count() - 1);
-                /*
-                 * DVPE is necessary so long as cross-VPE interrupts
-                 * are done via read-modify-write of Cause register.
-                 */
-                vpflags = dvpe();
-                clear_c0_cause(CPUCTR_IMASKBIT);
-                evpe(vpflags);
                 /*
                  * There are things we only want to do once per tick
                  * in an "MP" system.   One TC of each VPE will take
@@ -141,15 +104,13 @@ irqreturn_t mips_timer_interrupt(int irq, void *dev_id)
                  * the tick on VPE 0 to run the full timer_interrupt().
                  */
                 if (cpu_data[cpu].vpe_id == 0) {
-                                timer_interrupt(irq, NULL);
-                                smtc_timer_broadcast(cpu_data[cpu].vpe_id);
-                                scroll_display_message();
+                        timer_interrupt(irq, NULL);
                 } else {
                         write_c0_compare(read_c0_count() +
                                          (mips_hpt_frequency/HZ));
                         local_timer_interrupt(irq, dev_id);
-                        smtc_timer_broadcast(cpu_data[cpu].vpe_id);
                 }
+                smtc_timer_broadcast(cpu_data[cpu].vpe_id);
         }
 #else /* CONFIG_MIPS_MT_SMTC */
         int r2 = cpu_has_mips_r2;
@@ -167,8 +128,6 @@ irqreturn_t mips_timer_interrupt(int irq, void *dev_id)
                 /* we keep interrupt disabled all the time */
                 if (!r2 || (read_c0_cause() & (1 << 30)))
                         timer_interrupt(irq, NULL);
-
-                scroll_display_message();
         } else {
                 /* Everyone else needs to reset the timer int here as
                    ll_local_timer_interrupt doesn't */
@@ -262,6 +221,8 @@ void __init mips_time_init(void)
                (est_freq%1000000)*100/1000000);
 
         cpu_khz = est_freq / 1000;
+
+        mips_scroll_message();
 }
 
 void __init plat_timer_setup(struct irqaction *irq)
diff --git a/arch/mips/mips-boards/malta/malta_setup.c b/arch/mips/mips-boards/malta/malta_setup.c
index 7873932532a1..c14b7bf89950 100644
--- a/arch/mips/mips-boards/malta/malta_setup.c
+++ b/arch/mips/mips-boards/malta/malta_setup.c
@@ -56,6 +56,12 @@ const char *get_system_type(void)
         return "MIPS Malta";
 }
 
+#if defined(CONFIG_MIPS_MT_SMTC)
+const char display_string[] = "       SMTC LINUX ON MALTA       ";
+#else
+const char display_string[] = "        LINUX ON MALTA       ";
+#endif /* CONFIG_MIPS_MT_SMTC */
+
 #ifdef CONFIG_BLK_DEV_FD
 void __init fd_activate(void)
 {
diff --git a/arch/mips/mips-boards/sead/sead_setup.c b/arch/mips/mips-boards/sead/sead_setup.c
index a189dec7c7bc..811aba100605 100644
--- a/arch/mips/mips-boards/sead/sead_setup.c
+++ b/arch/mips/mips-boards/sead/sead_setup.c
@@ -43,6 +43,8 @@ const char *get_system_type(void)
         return "MIPS SEAD";
 }
 
+const char display_string[] = "        LINUX ON SEAD       ";
+
 void __init plat_mem_setup(void)
 {
         ioport_resource.end = 0x7fffffff;
diff --git a/arch/mips/mm/dma-default.c b/arch/mips/mm/dma-default.c
index f0eb29917d9a..76903c727647 100644
--- a/arch/mips/mm/dma-default.c
+++ b/arch/mips/mm/dma-default.c
@@ -168,8 +168,9 @@ int dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
                 addr = (unsigned long) page_address(sg->page);
                 if (!plat_device_is_coherent(dev) && addr)
                         __dma_sync(addr + sg->offset, sg->length, direction);
-                sg->dma_address = plat_map_dma_mem_page(dev, sg->page) +
-                                  sg->offset;
+                sg->dma_address = plat_map_dma_mem(dev,
+                                                   (void *)(addr + sg->offset),
+                                                   sg->length);
         }
 
         return nents;
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;
-        end     = GT_PCI_MEM_BASE + GT_PCI_MEM_BASE - 1;
+        .start  = GT_PCI_MEM_BASE,
+        .end    = GT_PCI_MEM_BASE + GT_PCI_MEM_BASE - 1,
 };
 
 static struct resource ocelot_io_resource = {
-        start   = GT_PCI_IO_BASE;
-        end     = GT_PCI_IO_BASE + GT_PCI_IO_SIZE - 1;
+        .start  = GT_PCI_IO_BASE,
+        .end    = GT_PCI_IO_BASE + GT_PCI_IO_SIZE - 1,
 };
 
 static struct pci_controller ocelot_pci_controller = {
-        .pci_ops        = gt64xxx_pci0_ops;
-        .mem_resource   = &ocelot_mem_resource;
-        .io_resource    = &ocelot_io_resource;
+        .pci_ops        = gt64xxx_pci0_ops,
+        .mem_resource   = &ocelot_mem_resource,
+        .io_resource    = &ocelot_io_resource,
 };
 
 static int __init ocelot_pcibios_init(void)
diff --git a/arch/mips/qemu/q-irq.c b/arch/mips/qemu/q-irq.c
index f5ea2fe10f14..89891e984b3b 100644
--- a/arch/mips/qemu/q-irq.c
+++ b/arch/mips/qemu/q-irq.c
@@ -7,8 +7,6 @@
 #include <asm/system.h>
 #include <asm/time.h>
 
-extern asmlinkage void qemu_handle_int(void);
-
 asmlinkage void plat_irq_dispatch(void)
 {
         unsigned int pending = read_c0_status() & read_c0_cause();
diff --git a/arch/mips/sgi-ip27/ip27-memory.c b/arch/mips/sgi-ip27/ip27-memory.c
index fe8a1066aec1..e5e023f50a07 100644
--- a/arch/mips/sgi-ip27/ip27-memory.c
+++ b/arch/mips/sgi-ip27/ip27-memory.c
@@ -517,7 +517,7 @@ void __init paging_init(void)
                 pfn_t start_pfn = slot_getbasepfn(node, 0);
                 pfn_t end_pfn = node_getmaxclick(node) + 1;
 
-                zones_size[ZONE_DMA] = end_pfn - start_pfn;
+                zones_size[ZONE_NORMAL] = end_pfn - start_pfn;
                 free_area_init_node(node, NODE_DATA(node),
                                 zones_size, start_pfn, NULL);
 
diff --git a/arch/mips/sgi-ip32/Makefile b/arch/mips/sgi-ip32/Makefile
index 7e1416768a60..60f0227425e7 100644
--- a/arch/mips/sgi-ip32/Makefile
+++ b/arch/mips/sgi-ip32/Makefile
@@ -3,5 +3,5 @@
 # under Linux.
 #
 
-obj-y   += ip32-berr.o ip32-irq.o ip32-setup.o ip32-reset.o \
+obj-y   += ip32-berr.o ip32-irq.o ip32-platform.o ip32-setup.o ip32-reset.o \
            crime.o ip32-memory.o
diff --git a/arch/mips/sgi-ip32/ip32-platform.c b/arch/mips/sgi-ip32/ip32-platform.c
new file mode 100644
index 000000000000..120b15932caf
--- /dev/null
+++ b/arch/mips/sgi-ip32/ip32-platform.c
@@ -0,0 +1,20 @@
+#include <linux/init.h>
+#include <linux/platform_device.h>
+
+static __init int meth_devinit(void)
+{
+        struct platform_device *pd;
+        int ret;
+
+        pd = platform_device_alloc("meth", -1);
+        if (!pd)
+                return -ENOMEM;
+
+        ret = platform_device_add(pd);
+        if (ret)
+                platform_device_put(pd);
+
+        return ret;
+}
+
+device_initcall(meth_devinit);
diff --git a/arch/mips/sni/pcimt.c b/arch/mips/sni/pcimt.c
index 9ee208daa8b1..97b234361b4d 100644
--- a/arch/mips/sni/pcimt.c
+++ b/arch/mips/sni/pcimt.c
@@ -6,7 +6,7 @@
  * for more details.
  *
  * Copyright (C) 1996, 97, 98, 2000, 03, 04, 06 Ralf Baechle (ralf@linux-mips.org)
- * Copyright (C) 2006 Thomas Bogendoerfer (tsbogend@alpha.franken.de)
+ * Copyright (C) 2006,2007 Thomas Bogendoerfer (tsbogend@alpha.franken.de)
  */
 
 #include <linux/init.h>
@@ -131,6 +131,19 @@ static struct resource pcimt_io_resources[] = {
         }
 };
 
+static struct resource pcimt_mem_resources[] = {
+        {
+                /*
+                 * this region should only be 4 bytes long,
+                 * but it's 16MB on all RM300C I've checked
+                 */
+                .start  = 0x1a000000,
+                .end    = 0x1affffff,
+                .name   = "PCI INT ACK",
+                .flags  = IORESOURCE_BUSY
+        }
+};
+
 static struct resource sni_mem_resource = {
         .start  = 0x18000000UL,
         .end    = 0x1fbfffffUL,
@@ -145,6 +158,9 @@ static void __init sni_pcimt_resource_init(void)
         /* request I/O space for devices used on all i[345]86 PCs */
         for (i = 0; i < ARRAY_SIZE(pcimt_io_resources); i++)
                 request_resource(&sni_io_resource, pcimt_io_resources + i);
+        /* request MEM space for devices used on all i[345]86 PCs */
+        for (i = 0; i < ARRAY_SIZE(pcimt_mem_resources); i++)
+                request_resource(&sni_mem_resource, pcimt_mem_resources + i);
 }
 
 extern struct pci_ops sni_pcimt_ops;
diff --git a/arch/mips/sni/setup.c b/arch/mips/sni/setup.c
index 68d7cf609b4f..4fedfbda0c79 100644
--- a/arch/mips/sni/setup.c
+++ b/arch/mips/sni/setup.c
@@ -6,7 +6,7 @@
  * for more details.
  *
  * Copyright (C) 1996, 97, 98, 2000, 03, 04, 06 Ralf Baechle (ralf@linux-mips.org)
- * Copyright (C) 2006 Thomas Bogendoerfer (tsbogend@alpha.franken.de)
+ * Copyright (C) 2006,2007 Thomas Bogendoerfer (tsbogend@alpha.franken.de)
  */
 #include <linux/eisa.h>
 #include <linux/init.h>
@@ -92,3 +92,34 @@ void __init plat_mem_setup(void)
 
         sni_display_setup();
 }
+
+#if CONFIG_PCI
+
+#include <linux/pci.h>
+#include <video/vga.h>
+#include <video/cirrus.h>
+
+static void __devinit quirk_cirrus_ram_size(struct pci_dev *dev)
+{
+        u16 cmd;
+
+        /*
+         * firmware doesn't set the ram size correct, so we
+         * need to do it here, otherwise we get screen corruption
+         * on older Cirrus chips
+         */
+        pci_read_config_word (dev, PCI_COMMAND, &cmd);
+        if ((cmd & (PCI_COMMAND_IO|PCI_COMMAND_MEMORY))
+                == (PCI_COMMAND_IO|PCI_COMMAND_MEMORY)) {
+                vga_wseq (NULL, CL_SEQR6, 0x12);        /* unlock all extension registers */
+                vga_wseq (NULL, CL_SEQRF, 0x18);
+        }
+}
+
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_CIRRUS, PCI_DEVICE_ID_CIRRUS_5434_8,
+                        quirk_cirrus_ram_size);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_CIRRUS, PCI_DEVICE_ID_CIRRUS_5436,
+                        quirk_cirrus_ram_size);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_CIRRUS, PCI_DEVICE_ID_CIRRUS_5446,
+                        quirk_cirrus_ram_size);
+#endif
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/Makefile b/arch/powerpc/Makefile
index 6238b5875fd1..fbafd965dcd2 100644
--- a/arch/powerpc/Makefile
+++ b/arch/powerpc/Makefile
@@ -142,7 +142,6 @@ drivers-$(CONFIG_OPROFILE)	+= arch/powerpc/oprofile/
 
 # Default to zImage, override when needed
 defaultimage-y                  := zImage
-defaultimage-$(CONFIG_PPC_ISERIES) := vmlinux
 defaultimage-$(CONFIG_DEFAULT_UIMAGE) := uImage
 KBUILD_IMAGE := $(defaultimage-y)
 all: $(KBUILD_IMAGE)
diff --git a/arch/powerpc/boot/Makefile b/arch/powerpc/boot/Makefile
index 83788986b93b..ff2701949ee1 100644
--- a/arch/powerpc/boot/Makefile
+++ b/arch/powerpc/boot/Makefile
@@ -11,20 +11,18 @@
 #       bootloader and increase compatibility with OpenFirmware.
 #
 #       To this end we need to define BOOTCC, etc, as the tools
-#       needed to build the 32 bit image.  These are normally HOSTCC,
-#       but may be a third compiler if, for example, you are cross
-#       compiling from an intel box.  Once the 64bit ppc gcc is
-#       stable it will probably simply be a compiler switch to
-#       compile for 32bit mode.
+#       needed to build the 32 bit image.  That's normally the same
+#       compiler for the rest of the kernel, with the -m32 flag added.
 #       To make it easier to setup a cross compiler,
 #       CROSS32_COMPILE is setup as a prefix just like CROSS_COMPILE
 #       in the toplevel makefile.
 
 all: $(obj)/zImage
 
-HOSTCC          := gcc
-BOOTCFLAGS      := $(HOSTCFLAGS) -fno-builtin -nostdinc -isystem \
-                   $(shell $(CROSS32CC) -print-file-name=include) -fPIC
+BOOTCFLAGS    := -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs \
+                 -fno-strict-aliasing -Os -msoft-float -pipe \
+                 -fomit-frame-pointer -fno-builtin -fPIC -nostdinc \
+                 -isystem $(shell $(CROSS32CC) -print-file-name=include)
 BOOTAFLAGS      := -D__ASSEMBLY__ $(BOOTCFLAGS) -traditional -nostdinc
 
 ifeq ($(call cc-option-yn, -fstack-protector),y)
@@ -33,8 +31,8 @@ endif
 
 BOOTCFLAGS      += -I$(obj) -I$(srctree)/$(obj)
 
-$(obj)/44x.o: BOOTCFLAGS += -Wa,-mbooke
-$(obj)/ebony.o: BOOTCFLAGS += -Wa,-mbooke
+$(obj)/44x.o: BOOTCFLAGS += -mcpu=440
+$(obj)/ebony.o: BOOTCFLAGS += -mcpu=440
 
 zlib       := inffast.c inflate.c inftrees.c
 zlibheader := inffast.h inffixed.h inflate.h inftrees.h infutil.h
@@ -136,6 +134,7 @@ image-$(CONFIG_PPC_EFIKA)		+= zImage.chrp
 image-$(CONFIG_PPC_PMAC)                += zImage.pmac
 image-$(CONFIG_PPC_HOLLY)               += zImage.holly-elf
 image-$(CONFIG_PPC_PRPMC2800)           += zImage.prpmc2800
+image-$(CONFIG_PPC_ISERIES)             += zImage.iseries
 image-$(CONFIG_DEFAULT_UIMAGE)          += uImage
 
 ifneq ($(CONFIG_DEVICE_TREE),"")
@@ -185,6 +184,9 @@ $(obj)/zImage.initrd.%: vmlinux $(wrapperbits)
 $(obj)/zImage.%: vmlinux $(wrapperbits)
         $(call if_changed,wrap,$*)
 
+$(obj)/zImage.iseries: vmlinux
+        $(STRIP) -s -R .comment $< -o $@
+
 $(obj)/zImage.ps3: vmlinux
         $(STRIP) -s -R .comment $< -o $@
 
diff --git a/arch/powerpc/boot/crt0.S b/arch/powerpc/boot/crt0.S
index 5a4215c4b014..f1c4dfc635be 100644
--- a/arch/powerpc/boot/crt0.S
+++ b/arch/powerpc/boot/crt0.S
@@ -13,6 +13,7 @@
 
         .text
         /* a procedure descriptor used when booting this as a COFF file */
+        .globl  _zimage_start_opd
 _zimage_start_opd:
         .long   _zimage_start, 0, 0, 0
 
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/boot/wrapper b/arch/powerpc/boot/wrapper
index 2ed8b8b3f0ec..da77adc73078 100755
--- a/arch/powerpc/boot/wrapper
+++ b/arch/powerpc/boot/wrapper
@@ -129,7 +129,7 @@ case "$platform" in
 pmac|pseries|chrp)
     platformo=$object/of.o
     ;;
-pmaccoff)
+coff)
     platformo=$object/of.o
     lds=$object/zImage.coff.lds
     ;;
@@ -220,7 +220,7 @@ case "$platform" in
 pseries|chrp)
     $object/addnote "$ofile"
     ;;
-pmaccoff)
+coff)
     ${CROSS}objcopy -O aixcoff-rs6000 --set-start "$entry" "$ofile"
     $object/hack-coff "$ofile"
     ;;
diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c
index 6ef87fb90b8e..b2b5d664d328 100644
--- a/arch/powerpc/kernel/cputable.c
+++ b/arch/powerpc/kernel/cputable.c
@@ -235,6 +235,7 @@ static struct cpu_spec cpu_specs[] = {
                 .icache_bsize           = 128,
                 .dcache_bsize           = 128,
                 .num_pmcs               = 8,
+                .pmc_type               = PPC_PMC_IBM,
                 .cpu_setup              = __setup_cpu_ppc970,
                 .cpu_restore            = __restore_cpu_ppc970,
                 .oprofile_cpu_type      = "ppc64/970MP",
@@ -251,6 +252,7 @@ static struct cpu_spec cpu_specs[] = {
                 .icache_bsize           = 128,
                 .dcache_bsize           = 128,
                 .num_pmcs               = 8,
+                .pmc_type               = PPC_PMC_IBM,
                 .cpu_setup              = __setup_cpu_ppc970MP,
                 .cpu_restore            = __restore_cpu_ppc970,
                 .oprofile_cpu_type      = "ppc64/970MP",
@@ -317,6 +319,7 @@ static struct cpu_spec cpu_specs[] = {
                 .icache_bsize           = 128,
                 .dcache_bsize           = 128,
                 .num_pmcs               = 6,
+                .pmc_type               = PPC_PMC_IBM,
                 .oprofile_cpu_type      = "ppc64/power6",
                 .oprofile_type          = PPC_OPROFILE_POWER4,
                 .oprofile_mmcra_sihv    = POWER6_MMCRA_SIHV,
@@ -335,6 +338,7 @@ static struct cpu_spec cpu_specs[] = {
                 .icache_bsize           = 128,
                 .dcache_bsize           = 128,
                 .num_pmcs               = 6,
+                .pmc_type               = PPC_PMC_IBM,
                 .oprofile_cpu_type      = "ppc64/power6",
                 .oprofile_type          = PPC_OPROFILE_POWER4,
                 .oprofile_mmcra_sihv    = POWER6_MMCRA_SIHV,
diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c
index 068377a2a8dc..42c8ed6ed528 100644
--- a/arch/powerpc/kernel/irq.c
+++ b/arch/powerpc/kernel/irq.c
@@ -489,7 +489,7 @@ struct irq_host *irq_alloc_host(unsigned int revmap_type,
         case IRQ_HOST_MAP_LINEAR:
                 rmap = (unsigned int *)(host + 1);
                 for (i = 0; i < revmap_arg; i++)
-                        rmap[i] = IRQ_NONE;
+                        rmap[i] = NO_IRQ;
                 host->revmap_data.linear.size = revmap_arg;
                 smp_wmb();
                 host->revmap_data.linear.revmap = rmap;
@@ -614,7 +614,7 @@ unsigned int irq_create_mapping(struct irq_host *host,
          * host->ops->map() to update the flags
          */
         virq = irq_find_mapping(host, hwirq);
-        if (virq != IRQ_NONE) {
+        if (virq != NO_IRQ) {
                 if (host->ops->remap)
                         host->ops->remap(host, virq, hwirq);
                 pr_debug("irq: -> existing mapping on virq %d\n", virq);
@@ -741,7 +741,7 @@ void irq_dispose_mapping(unsigned int virq)
         switch(host->revmap_type) {
         case IRQ_HOST_MAP_LINEAR:
                 if (hwirq < host->revmap_data.linear.size)
-                        host->revmap_data.linear.revmap[hwirq] = IRQ_NONE;
+                        host->revmap_data.linear.revmap[hwirq] = NO_IRQ;
                 break;
         case IRQ_HOST_MAP_TREE:
                 /* Check if radix tree allocated yet */
diff --git a/arch/powerpc/kernel/of_platform.c b/arch/powerpc/kernel/of_platform.c
index d501c23e5159..d454f61c9c7c 100644
--- a/arch/powerpc/kernel/of_platform.c
+++ b/arch/powerpc/kernel/of_platform.c
@@ -433,7 +433,7 @@ static int __devinit of_pci_phb_probe(struct of_device *dev,
          * Note also that we don't do ISA, this will also be fixed with a
          * more massive rework.
          */
-        pci_setup_phb_io(phb, 0);
+        pci_setup_phb_io(phb, pci_io_base == 0);
 
         /* Init pci_dn data structures */
         pci_devs_phb_init_dynamic(phb);
diff --git a/arch/powerpc/kernel/pmc.c b/arch/powerpc/kernel/pmc.c
index 24d7b7c99bb9..ea04e0ab3f2f 100644
--- a/arch/powerpc/kernel/pmc.c
+++ b/arch/powerpc/kernel/pmc.c
@@ -20,8 +20,8 @@
 #include <asm/cputable.h>
 #include <asm/pmc.h>
 
-#ifndef MMCR0_PMA0
-#define MMCR0_PMA0      0
+#ifndef MMCR0_PMAO
+#define MMCR0_PMAO      0
 #endif
 
 static void dummy_perf(struct pt_regs *regs)
@@ -30,7 +30,7 @@ static void dummy_perf(struct pt_regs *regs)
         mtpmr(PMRN_PMGC0, mfpmr(PMRN_PMGC0) & ~PMGC0_PMIE);
 #elif defined(CONFIG_PPC64) || defined(CONFIG_6xx)
         if (cur_cpu_spec->pmc_type == PPC_PMC_IBM)
-                mtspr(SPRN_MMCR0, mfspr(SPRN_MMCR0) & ~(MMCR0_PMXE|MMCR0_PMA0));
+                mtspr(SPRN_MMCR0, mfspr(SPRN_MMCR0) & ~(MMCR0_PMXE|MMCR0_PMAO));
 #else
         mtspr(SPRN_MMCR0, mfspr(SPRN_MMCR0) & ~MMCR0_PMXE);
 #endif
diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
index 066a6a7a25b8..af42ddab3ab4 100644
--- a/arch/powerpc/kernel/prom.c
+++ b/arch/powerpc/kernel/prom.c
@@ -1171,11 +1171,12 @@ EXPORT_SYMBOL(of_find_node_by_name);
 
 /**
  *      of_find_node_by_type - Find a node by its "device_type" property
- *      @from:  The node to start searching from or NULL, the node
- *              you pass will not be searched, only the next one
- *              will; typically, you pass what the previous call
- *              returned. of_node_put() will be called on it
- *      @name:  The type string to match against
+ *      @from:  The node to start searching from, or NULL to start searching
+ *              the entire device tree. The node you pass will not be
+ *              searched, only the next one will; typically, you pass
+ *              what the previous call returned. of_node_put() will be
+ *              called on from for you.
+ *      @type:  The type string to match against
  *
  *      Returns a node pointer with refcount incremented, use
  *      of_node_put() on it when done.
diff --git a/arch/powerpc/kernel/ptrace.c b/arch/powerpc/kernel/ptrace.c
index f4f391cdd8f5..bf76562167c3 100644
--- a/arch/powerpc/kernel/ptrace.c
+++ b/arch/powerpc/kernel/ptrace.c
@@ -218,6 +218,7 @@ set_single_step(struct task_struct *task)
                 regs->msr |= MSR_SE;
 #endif
         }
+        set_tsk_thread_flag(task, TIF_SINGLESTEP);
 }
 
 static inline void
@@ -233,6 +234,7 @@ clear_single_step(struct task_struct *task)
                 regs->msr &= ~MSR_SE;
 #endif
         }
+        clear_tsk_thread_flag(task, TIF_SINGLESTEP);
 }
 #endif /* CONFIG_PPC32 */
 
diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index 22f1ef1b3100..d577b71db375 100644
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -201,13 +201,6 @@ int smp_call_function_map(void (*func) (void *info), void *info, int nonatomic,
         /* Can deadlock when called with interrupts disabled */
         WARN_ON(irqs_disabled());
 
-        /* remove 'self' from the map */
-        if (cpu_isset(smp_processor_id(), map))
-                cpu_clear(smp_processor_id(), map);
-
-        /* sanity check the map, remove any non-online processors. */
-        cpus_and(map, map, cpu_online_map);
-
         if (unlikely(smp_ops == NULL))
                 return ret;
 
@@ -222,10 +215,17 @@ int smp_call_function_map(void (*func) (void *info), void *info, int nonatomic,
         /* Must grab online cpu count with preempt disabled, otherwise
          * it can change. */
         num_cpus = num_online_cpus() - 1;
-        if (!num_cpus || cpus_empty(map)) {
-                ret = 0;
-                goto out;
-        }
+        if (!num_cpus)
+                goto done;
+
+        /* remove 'self' from the map */
+        if (cpu_isset(smp_processor_id(), map))
+                cpu_clear(smp_processor_id(), map);
+
+        /* sanity check the map, remove any non-online processors. */
+        cpus_and(map, map, cpu_online_map);
+        if (cpus_empty(map))
+                goto done;
 
         call_data = &data;
         smp_wmb();
@@ -263,6 +263,7 @@ int smp_call_function_map(void (*func) (void *info), void *info, int nonatomic,
                 }
         }
 
+ done:
         ret = 0;
 
  out:
@@ -282,16 +283,17 @@ EXPORT_SYMBOL(smp_call_function);
 int smp_call_function_single(int cpu, void (*func) (void *info), void *info, int nonatomic,
                         int wait)
 {
-        cpumask_t map=CPU_MASK_NONE;
+        cpumask_t map = CPU_MASK_NONE;
+        int ret = -EBUSY;
 
         if (!cpu_online(cpu))
                 return -EINVAL;
 
-        if (cpu == smp_processor_id())
-                return -EBUSY;
-
         cpu_set(cpu, map);
-        return smp_call_function_map(func,info,nonatomic,wait,map);
+        if (cpu != get_cpu())
+                ret = smp_call_function_map(func,info,nonatomic,wait,map);
+        put_cpu();
+        return ret;
 }
 EXPORT_SYMBOL(smp_call_function_single);
 
diff --git a/arch/powerpc/kernel/vmlinux.lds.S b/arch/powerpc/kernel/vmlinux.lds.S
index f7d7bf19e4fb..21c39ff2dc39 100644
--- a/arch/powerpc/kernel/vmlinux.lds.S
+++ b/arch/powerpc/kernel/vmlinux.lds.S
@@ -34,7 +34,6 @@ SECTIONS
         /* Text and gots */
         .text : {
                 _text = .;
-                *(.text.*)
                 TEXT_TEXT
                 SCHED_TEXT
                 LOCK_TEXT
diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index 246eeea40ece..0266a94d83b6 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -310,11 +310,12 @@ void __init paging_init(void)
 
 #ifdef CONFIG_HIGHMEM
         map_page(PKMAP_BASE, 0, 0);     /* XXX gross */
-        pkmap_page_table = pte_offset_kernel(pmd_offset(pgd_offset_k
-                        (PKMAP_BASE), PKMAP_BASE), PKMAP_BASE);
+        pkmap_page_table = pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k
+                        (PKMAP_BASE), PKMAP_BASE), PKMAP_BASE), PKMAP_BASE);
         map_page(KMAP_FIX_BEGIN, 0, 0); /* XXX gross */
-        kmap_pte = pte_offset_kernel(pmd_offset(pgd_offset_k
-                        (KMAP_FIX_BEGIN), KMAP_FIX_BEGIN), KMAP_FIX_BEGIN);
+        kmap_pte = pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k
+                        (KMAP_FIX_BEGIN), KMAP_FIX_BEGIN), KMAP_FIX_BEGIN),
+                         KMAP_FIX_BEGIN);
         kmap_prot = PAGE_KERNEL;
 #endif /* CONFIG_HIGHMEM */
 
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/cell/cbe_cpufreq.c b/arch/powerpc/platforms/cell/cbe_cpufreq.c
index f9ac3fe3be97..ac445998d831 100644
--- a/arch/powerpc/platforms/cell/cbe_cpufreq.c
+++ b/arch/powerpc/platforms/cell/cbe_cpufreq.c
@@ -67,6 +67,7 @@ static u64 MIC_Slow_Next_Timer_table[] = {
         0x00003FC000000000ull,
 };
 
+static unsigned int pmi_frequency_limit = 0;
 /*
  * hardware specific functions
  */
@@ -164,7 +165,6 @@ static int set_pmode(int cpu, unsigned int slow_mode) {
 
 static void cbe_cpufreq_handle_pmi(struct of_device *dev, pmi_message_t pmi_msg)
 {
-        struct cpufreq_policy policy;
         u8 cpu;
         u8 cbe_pmode_new;
 
@@ -173,15 +173,27 @@ static void cbe_cpufreq_handle_pmi(struct of_device *dev, pmi_message_t pmi_msg)
         cpu = cbe_node_to_cpu(pmi_msg.data1);
         cbe_pmode_new = pmi_msg.data2;
 
-        cpufreq_get_policy(&policy, cpu);
+        pmi_frequency_limit = cbe_freqs[cbe_pmode_new].frequency;
 
-        policy.max = min(policy.max, cbe_freqs[cbe_pmode_new].frequency);
-        policy.min = min(policy.min, policy.max);
+        pr_debug("cbe_handle_pmi: max freq=%d\n", pmi_frequency_limit);
+}
+
+static int pmi_notifier(struct notifier_block *nb,
+                                       unsigned long event, void *data)
+{
+        struct cpufreq_policy *policy = data;
 
-        pr_debug("cbe_handle_pmi: new policy.min=%d policy.max=%d\n", policy.min, policy.max);
-        cpufreq_set_policy(&policy);
+        if (event != CPUFREQ_INCOMPATIBLE)
+                return 0;
+
+        cpufreq_verify_within_limits(policy, 0, pmi_frequency_limit);
+        return 0;
 }
 
+static struct notifier_block pmi_notifier_block = {
+        .notifier_call = pmi_notifier,
+};
+
 static struct pmi_handler cbe_pmi_handler = {
         .type                   = PMI_TYPE_FREQ_CHANGE,
         .handle_pmi_message     = cbe_cpufreq_handle_pmi,
@@ -238,12 +250,21 @@ static int cbe_cpufreq_cpu_init(struct cpufreq_policy *policy)
 
         cpufreq_frequency_table_get_attr(cbe_freqs, policy->cpu);
 
+        if (pmi_dev) {
+                /* frequency might get limited later, initialize limit with max_freq */
+                pmi_frequency_limit = max_freq;
+                cpufreq_register_notifier(&pmi_notifier_block, CPUFREQ_POLICY_NOTIFIER);
+        }
+
         /* this ensures that policy->cpuinfo_min and policy->cpuinfo_max are set correctly */
         return cpufreq_frequency_table_cpuinfo(policy, cbe_freqs);
 }
 
 static int cbe_cpufreq_cpu_exit(struct cpufreq_policy *policy)
 {
+        if (pmi_dev)
+                cpufreq_unregister_notifier(&pmi_notifier_block, CPUFREQ_POLICY_NOTIFIER);
+
         cpufreq_frequency_table_put_attr(policy->cpu);
         return 0;
 }
diff --git a/arch/powerpc/platforms/cell/spufs/context.c b/arch/powerpc/platforms/cell/spufs/context.c
index 8654749e317b..7c51cb54bca1 100644
--- a/arch/powerpc/platforms/cell/spufs/context.c
+++ b/arch/powerpc/platforms/cell/spufs/context.c
@@ -39,7 +39,7 @@ struct spu_context *alloc_spu_context(struct spu_gang *gang)
         if (spu_init_csa(&ctx->csa))
                 goto out_free;
         spin_lock_init(&ctx->mmio_lock);
-        spin_lock_init(&ctx->mapping_lock);
+        mutex_init(&ctx->mapping_lock);
         kref_init(&ctx->kref);
         mutex_init(&ctx->state_mutex);
         mutex_init(&ctx->run_mutex);
@@ -103,6 +103,7 @@ void spu_forget(struct spu_context *ctx)
 
 void spu_unmap_mappings(struct spu_context *ctx)
 {
+        mutex_lock(&ctx->mapping_lock);
         if (ctx->local_store)
                 unmap_mapping_range(ctx->local_store, 0, LS_SIZE, 1);
         if (ctx->mfc)
@@ -117,6 +118,7 @@ void spu_unmap_mappings(struct spu_context *ctx)
                 unmap_mapping_range(ctx->mss, 0, 0x1000, 1);
         if (ctx->psmap)
                 unmap_mapping_range(ctx->psmap, 0, 0x20000, 1);
+        mutex_unlock(&ctx->mapping_lock);
 }
 
 /**
diff --git a/arch/powerpc/platforms/cell/spufs/file.c b/arch/powerpc/platforms/cell/spufs/file.c
index 45614c73c784..b1e7e2f8a2e9 100644
--- a/arch/powerpc/platforms/cell/spufs/file.c
+++ b/arch/powerpc/platforms/cell/spufs/file.c
@@ -45,11 +45,11 @@ spufs_mem_open(struct inode *inode, struct file *file)
         struct spufs_inode_info *i = SPUFS_I(inode);
         struct spu_context *ctx = i->i_ctx;
 
-        spin_lock(&ctx->mapping_lock);
+        mutex_lock(&ctx->mapping_lock);
         file->private_data = ctx;
         if (!i->i_openers++)
                 ctx->local_store = inode->i_mapping;
-        spin_unlock(&ctx->mapping_lock);
+        mutex_unlock(&ctx->mapping_lock);
         return 0;
 }
 
@@ -59,10 +59,10 @@ spufs_mem_release(struct inode *inode, struct file *file)
         struct spufs_inode_info *i = SPUFS_I(inode);
         struct spu_context *ctx = i->i_ctx;
 
-        spin_lock(&ctx->mapping_lock);
+        mutex_lock(&ctx->mapping_lock);
         if (!--i->i_openers)
                 ctx->local_store = NULL;
-        spin_unlock(&ctx->mapping_lock);
+        mutex_unlock(&ctx->mapping_lock);
         return 0;
 }
 
@@ -217,6 +217,7 @@ unsigned long spufs_get_unmapped_area(struct file *file, unsigned long addr,
 
 static const struct file_operations spufs_mem_fops = {
         .open                   = spufs_mem_open,
+        .release                = spufs_mem_release,
         .read                   = spufs_mem_read,
         .write                  = spufs_mem_write,
         .llseek                 = generic_file_llseek,
@@ -309,11 +310,11 @@ static int spufs_cntl_open(struct inode *inode, struct file *file)
         struct spufs_inode_info *i = SPUFS_I(inode);
         struct spu_context *ctx = i->i_ctx;
 
-        spin_lock(&ctx->mapping_lock);
+        mutex_lock(&ctx->mapping_lock);
         file->private_data = ctx;
         if (!i->i_openers++)
                 ctx->cntl = inode->i_mapping;
-        spin_unlock(&ctx->mapping_lock);
+        mutex_unlock(&ctx->mapping_lock);
         return simple_attr_open(inode, file, spufs_cntl_get,
                                         spufs_cntl_set, "0x%08lx");
 }
@@ -326,10 +327,10 @@ spufs_cntl_release(struct inode *inode, struct file *file)
 
         simple_attr_close(inode, file);
 
-        spin_lock(&ctx->mapping_lock);
+        mutex_lock(&ctx->mapping_lock);
         if (!--i->i_openers)
                 ctx->cntl = NULL;
-        spin_unlock(&ctx->mapping_lock);
+        mutex_unlock(&ctx->mapping_lock);
         return 0;
 }
 
@@ -812,11 +813,11 @@ static int spufs_signal1_open(struct inode *inode, struct file *file)
         struct spufs_inode_info *i = SPUFS_I(inode);
         struct spu_context *ctx = i->i_ctx;
 
-        spin_lock(&ctx->mapping_lock);
+        mutex_lock(&ctx->mapping_lock);
         file->private_data = ctx;
         if (!i->i_openers++)
                 ctx->signal1 = inode->i_mapping;
-        spin_unlock(&ctx->mapping_lock);
+        mutex_unlock(&ctx->mapping_lock);
         return nonseekable_open(inode, file);
 }
 
@@ -826,10 +827,10 @@ spufs_signal1_release(struct inode *inode, struct file *file)
         struct spufs_inode_info *i = SPUFS_I(inode);
         struct spu_context *ctx = i->i_ctx;
 
-        spin_lock(&ctx->mapping_lock);
+        mutex_lock(&ctx->mapping_lock);
         if (!--i->i_openers)
                 ctx->signal1 = NULL;
-        spin_unlock(&ctx->mapping_lock);
+        mutex_unlock(&ctx->mapping_lock);
         return 0;
 }
 
@@ -936,11 +937,11 @@ static int spufs_signal2_open(struct inode *inode, struct file *file)
         struct spufs_inode_info *i = SPUFS_I(inode);
         struct spu_context *ctx = i->i_ctx;
 
-        spin_lock(&ctx->mapping_lock);
+        mutex_lock(&ctx->mapping_lock);
         file->private_data = ctx;
         if (!i->i_openers++)
                 ctx->signal2 = inode->i_mapping;
-        spin_unlock(&ctx->mapping_lock);
+        mutex_unlock(&ctx->mapping_lock);
         return nonseekable_open(inode, file);
 }
 
@@ -950,10 +951,10 @@ spufs_signal2_release(struct inode *inode, struct file *file)
         struct spufs_inode_info *i = SPUFS_I(inode);
         struct spu_context *ctx = i->i_ctx;
 
-        spin_lock(&ctx->mapping_lock);
+        mutex_lock(&ctx->mapping_lock);
         if (!--i->i_openers)
                 ctx->signal2 = NULL;
-        spin_unlock(&ctx->mapping_lock);
+        mutex_unlock(&ctx->mapping_lock);
         return 0;
 }
 
@@ -1154,10 +1155,10 @@ static int spufs_mss_open(struct inode *inode, struct file *file)
 
         file->private_data = i->i_ctx;
 
-        spin_lock(&ctx->mapping_lock);
+        mutex_lock(&ctx->mapping_lock);
         if (!i->i_openers++)
                 ctx->mss = inode->i_mapping;
-        spin_unlock(&ctx->mapping_lock);
+        mutex_unlock(&ctx->mapping_lock);
         return nonseekable_open(inode, file);
 }
 
@@ -1167,10 +1168,10 @@ spufs_mss_release(struct inode *inode, struct file *file)
         struct spufs_inode_info *i = SPUFS_I(inode);
         struct spu_context *ctx = i->i_ctx;
 
-        spin_lock(&ctx->mapping_lock);
+        mutex_lock(&ctx->mapping_lock);
         if (!--i->i_openers)
                 ctx->mss = NULL;
-        spin_unlock(&ctx->mapping_lock);
+        mutex_unlock(&ctx->mapping_lock);
         return 0;
 }
 
@@ -1211,11 +1212,11 @@ static int spufs_psmap_open(struct inode *inode, struct file *file)
         struct spufs_inode_info *i = SPUFS_I(inode);
         struct spu_context *ctx = i->i_ctx;
 
-        spin_lock(&ctx->mapping_lock);
+        mutex_lock(&ctx->mapping_lock);
         file->private_data = i->i_ctx;
         if (!i->i_openers++)
                 ctx->psmap = inode->i_mapping;
-        spin_unlock(&ctx->mapping_lock);
+        mutex_unlock(&ctx->mapping_lock);
         return nonseekable_open(inode, file);
 }
 
@@ -1225,10 +1226,10 @@ spufs_psmap_release(struct inode *inode, struct file *file)
         struct spufs_inode_info *i = SPUFS_I(inode);
         struct spu_context *ctx = i->i_ctx;
 
-        spin_lock(&ctx->mapping_lock);
+        mutex_lock(&ctx->mapping_lock);
         if (!--i->i_openers)
                 ctx->psmap = NULL;
-        spin_unlock(&ctx->mapping_lock);
+        mutex_unlock(&ctx->mapping_lock);
         return 0;
 }
 
@@ -1281,11 +1282,11 @@ static int spufs_mfc_open(struct inode *inode, struct file *file)
         if (atomic_read(&inode->i_count) != 1)
                 return -EBUSY;
 
-        spin_lock(&ctx->mapping_lock);
+        mutex_lock(&ctx->mapping_lock);
         file->private_data = ctx;
         if (!i->i_openers++)
                 ctx->mfc = inode->i_mapping;
-        spin_unlock(&ctx->mapping_lock);
+        mutex_unlock(&ctx->mapping_lock);
         return nonseekable_open(inode, file);
 }
 
@@ -1295,10 +1296,10 @@ spufs_mfc_release(struct inode *inode, struct file *file)
         struct spufs_inode_info *i = SPUFS_I(inode);
         struct spu_context *ctx = i->i_ctx;
 
-        spin_lock(&ctx->mapping_lock);
+        mutex_lock(&ctx->mapping_lock);
         if (!--i->i_openers)
                 ctx->mfc = NULL;
-        spin_unlock(&ctx->mapping_lock);
+        mutex_unlock(&ctx->mapping_lock);
         return 0;
 }
 
diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c
index 7150730e2ff1..9807206e0219 100644
--- a/arch/powerpc/platforms/cell/spufs/inode.c
+++ b/arch/powerpc/platforms/cell/spufs/inode.c
@@ -177,7 +177,7 @@ static int spufs_rmdir(struct inode *parent, struct dentry *dir)
 static int spufs_fill_dir(struct dentry *dir, struct tree_descr *files,
                           int mode, struct spu_context *ctx)
 {
-        struct dentry *dentry;
+        struct dentry *dentry, *tmp;
         int ret;
 
         while (files->name && files->name[0]) {
@@ -193,7 +193,20 @@ static int spufs_fill_dir(struct dentry *dir, struct tree_descr *files,
         }
         return 0;
 out:
-        spufs_prune_dir(dir);
+        /*
+         * remove all children from dir. dir->inode is not set so don't
+         * just simply use spufs_prune_dir() and panic afterwards :)
+         * dput() looks like it will do the right thing:
+         * - dec parent's ref counter
+         * - remove child from parent's child list
+         * - free child's inode if possible
+         * - free child
+         */
+        list_for_each_entry_safe(dentry, tmp, &dir->d_subdirs, d_u.d_child) {
+                dput(dentry);
+        }
+
+        shrink_dcache_parent(dir);
         return ret;
 }
 
@@ -274,6 +287,7 @@ spufs_mkdir(struct inode *dir, struct dentry *dentry, unsigned int flags,
         goto out;
 
 out_free_ctx:
+        spu_forget(ctx);
         put_spu_context(ctx);
 out_iput:
         iput(inode);
@@ -349,37 +363,6 @@ out:
         return ret;
 }
 
-static int spufs_rmgang(struct inode *root, struct dentry *dir)
-{
-        /* FIXME: this fails if the dir is not empty,
-                  which causes a leak of gangs. */
-        return simple_rmdir(root, dir);
-}
-
-static int spufs_gang_close(struct inode *inode, struct file *file)
-{
-        struct inode *parent;
-        struct dentry *dir;
-        int ret;
-
-        dir = file->f_path.dentry;
-        parent = dir->d_parent->d_inode;
-
-        ret = spufs_rmgang(parent, dir);
-        WARN_ON(ret);
-
-        return dcache_dir_close(inode, file);
-}
-
-const struct file_operations spufs_gang_fops = {
-        .open           = dcache_dir_open,
-        .release        = spufs_gang_close,
-        .llseek         = dcache_dir_lseek,
-        .read           = generic_read_dir,
-        .readdir        = dcache_readdir,
-        .fsync          = simple_sync_file,
-};
-
 static int
 spufs_mkgang(struct inode *dir, struct dentry *dentry, int mode)
 {
@@ -407,7 +390,6 @@ spufs_mkgang(struct inode *dir, struct dentry *dentry, int mode)
         inode->i_fop = &simple_dir_operations;
 
         d_instantiate(dentry, inode);
-        dget(dentry);
         dir->i_nlink++;
         dentry->d_inode->i_nlink++;
         return ret;
@@ -437,7 +419,7 @@ static int spufs_gang_open(struct dentry *dentry, struct vfsmount *mnt)
                 goto out;
         }
 
-        filp->f_op = &spufs_gang_fops;
+        filp->f_op = &simple_dir_operations;
         fd_install(ret, filp);
 out:
         return ret;
@@ -458,8 +440,10 @@ static int spufs_create_gang(struct inode *inode,
          * in error path of *_open().
          */
         ret = spufs_gang_open(dget(dentry), mntget(mnt));
-        if (ret < 0)
-                WARN_ON(spufs_rmgang(inode, dentry));
+        if (ret < 0) {
+                int err = simple_rmdir(inode, dentry);
+                WARN_ON(err);
+        }
 
 out:
         mutex_unlock(&inode->i_mutex);
@@ -600,6 +584,10 @@ spufs_create_root(struct super_block *sb, void *data)
         struct inode *inode;
         int ret;
 
+        ret = -ENODEV;
+        if (!spu_management_ops)
+                goto out;
+
         ret = -ENOMEM;
         inode = spufs_new_inode(sb, S_IFDIR | 0775);
         if (!inode)
diff --git a/arch/powerpc/platforms/cell/spufs/sched.c b/arch/powerpc/platforms/cell/spufs/sched.c
index b6ecb30e7d58..3b831e07f1ed 100644
--- a/arch/powerpc/platforms/cell/spufs/sched.c
+++ b/arch/powerpc/platforms/cell/spufs/sched.c
@@ -93,43 +93,6 @@ void spu_stop_tick(struct spu_context *ctx)
         }
 }
 
-void spu_sched_tick(struct work_struct *work)
-{
-        struct spu_context *ctx =
-                container_of(work, struct spu_context, sched_work.work);
-        struct spu *spu;
-        int preempted = 0;
-
-        /*
-         * If this context is being stopped avoid rescheduling from the
-         * scheduler tick because we would block on the state_mutex.
-         * The caller will yield the spu later on anyway.
-         */
-        if (test_bit(SPU_SCHED_EXITING, &ctx->sched_flags))
-                return;
-
-        mutex_lock(&ctx->state_mutex);
-        spu = ctx->spu;
-        if (spu) {
-                int best = sched_find_first_bit(spu_prio->bitmap);
-                if (best <= ctx->prio) {
-                        spu_deactivate(ctx);
-                        preempted = 1;
-                }
-        }
-        mutex_unlock(&ctx->state_mutex);
-
-        if (preempted) {
-                /*
-                 * We need to break out of the wait loop in spu_run manually
-                 * to ensure this context gets put on the runqueue again
-                 * ASAP.
-                 */
-                wake_up(&ctx->stop_wq);
-        } else
-                spu_start_tick(ctx);
-}
-
 /**
  * spu_add_to_active_list - add spu to active list
  * @spu:        spu to add to the active list
@@ -273,34 +236,6 @@ static void spu_prio_wait(struct spu_context *ctx)
         remove_wait_queue(&ctx->stop_wq, &wait);
 }
 
-/**
- * spu_reschedule - try to find a runnable context for a spu
- * @spu:       spu available
- *
- * This function is called whenever a spu becomes idle.  It looks for the
- * most suitable runnable spu context and schedules it for execution.
- */
-static void spu_reschedule(struct spu *spu)
-{
-        int best;
-
-        spu_free(spu);
-
-        spin_lock(&spu_prio->runq_lock);
-        best = sched_find_first_bit(spu_prio->bitmap);
-        if (best < MAX_PRIO) {
-                struct list_head *rq = &spu_prio->runq[best];
-                struct spu_context *ctx;
-
-                BUG_ON(list_empty(rq));
-
-                ctx = list_entry(rq->next, struct spu_context, rq);
-                __spu_del_from_rq(ctx);
-                wake_up(&ctx->stop_wq);
-        }
-        spin_unlock(&spu_prio->runq_lock);
-}
-
 static struct spu *spu_get_idle(struct spu_context *ctx)
 {
         struct spu *spu = NULL;
@@ -429,6 +364,51 @@ int spu_activate(struct spu_context *ctx, unsigned long flags)
 }
 
 /**
+ * grab_runnable_context - try to find a runnable context
+ *
+ * Remove the highest priority context on the runqueue and return it
+ * to the caller.  Returns %NULL if no runnable context was found.
+ */
+static struct spu_context *grab_runnable_context(int prio)
+{
+        struct spu_context *ctx = NULL;
+        int best;
+
+        spin_lock(&spu_prio->runq_lock);
+        best = sched_find_first_bit(spu_prio->bitmap);
+        if (best < prio) {
+                struct list_head *rq = &spu_prio->runq[best];
+
+                BUG_ON(list_empty(rq));
+
+                ctx = list_entry(rq->next, struct spu_context, rq);
+                __spu_del_from_rq(ctx);
+        }
+        spin_unlock(&spu_prio->runq_lock);
+
+        return ctx;
+}
+
+static int __spu_deactivate(struct spu_context *ctx, int force, int max_prio)
+{
+        struct spu *spu = ctx->spu;
+        struct spu_context *new = NULL;
+
+        if (spu) {
+                new = grab_runnable_context(max_prio);
+                if (new || force) {
+                        spu_unbind_context(spu, ctx);
+                        spu_free(spu);
+                        if (new)
+                                wake_up(&new->stop_wq);
+                }
+
+        }
+
+        return new != NULL;
+}
+
+/**
  * spu_deactivate - unbind a context from it's physical spu
  * @ctx:        spu context to unbind
  *
@@ -437,12 +417,7 @@ int spu_activate(struct spu_context *ctx, unsigned long flags)
  */
 void spu_deactivate(struct spu_context *ctx)
 {
-        struct spu *spu = ctx->spu;
-
-        if (spu) {
-                spu_unbind_context(spu, ctx);
-                spu_reschedule(spu);
-        }
+        __spu_deactivate(ctx, 1, MAX_PRIO);
 }
 
 /**
@@ -455,21 +430,43 @@ void spu_deactivate(struct spu_context *ctx)
  */
 void spu_yield(struct spu_context *ctx)
 {
-        struct spu *spu;
-
-        if (mutex_trylock(&ctx->state_mutex)) {
-                if ((spu = ctx->spu) != NULL) {
-                        int best = sched_find_first_bit(spu_prio->bitmap);
-                        if (best < MAX_PRIO) {
-                                pr_debug("%s: yielding SPU %d NODE %d\n",
-                                         __FUNCTION__, spu->number, spu->node);
-                                spu_deactivate(ctx);
-                        }
-                }
+        if (!(ctx->flags & SPU_CREATE_NOSCHED)) {
+                mutex_lock(&ctx->state_mutex);
+                __spu_deactivate(ctx, 0, MAX_PRIO);
                 mutex_unlock(&ctx->state_mutex);
         }
 }
 
+void spu_sched_tick(struct work_struct *work)
+{
+        struct spu_context *ctx =
+                container_of(work, struct spu_context, sched_work.work);
+        int preempted;
+
+        /*
+         * If this context is being stopped avoid rescheduling from the
+         * scheduler tick because we would block on the state_mutex.
+         * The caller will yield the spu later on anyway.
+         */
+        if (test_bit(SPU_SCHED_EXITING, &ctx->sched_flags))
+                return;
+
+        mutex_lock(&ctx->state_mutex);
+        preempted = __spu_deactivate(ctx, 0, ctx->prio + 1);
+        mutex_unlock(&ctx->state_mutex);
+
+        if (preempted) {
+                /*
+                 * We need to break out of the wait loop in spu_run manually
+                 * to ensure this context gets put on the runqueue again
+                 * ASAP.
+                 */
+                wake_up(&ctx->stop_wq);
+        } else {
+                spu_start_tick(ctx);
+        }
+}
+
 int __init spu_sched_init(void)
 {
         int i;
diff --git a/arch/powerpc/platforms/cell/spufs/spufs.h b/arch/powerpc/platforms/cell/spufs/spufs.h
index 0a947fd7de57..47617e8014a5 100644
--- a/arch/powerpc/platforms/cell/spufs/spufs.h
+++ b/arch/powerpc/platforms/cell/spufs/spufs.h
@@ -55,7 +55,7 @@ struct spu_context {
         struct address_space *signal2;     /* 'signal2' area mappings. */
         struct address_space *mss;         /* 'mss' area mappings. */
         struct address_space *psmap;       /* 'psmap' area mappings. */
-        spinlock_t mapping_lock;
+        struct mutex mapping_lock;
         u64 object_id;             /* user space pointer for oprofile */
 
         enum { SPU_STATE_RUNNABLE, SPU_STATE_SAVED } state;
diff --git a/arch/powerpc/platforms/celleb/Makefile b/arch/powerpc/platforms/celleb/Makefile
index f4f82520dc4f..5240046d8671 100644
--- a/arch/powerpc/platforms/celleb/Makefile
+++ b/arch/powerpc/platforms/celleb/Makefile
@@ -4,5 +4,5 @@ obj-y				+= interrupt.o iommu.o setup.o \
 
 obj-$(CONFIG_SMP)               += smp.o
 obj-$(CONFIG_PPC_UDBG_BEAT)     += udbg_beat.o
-obj-$(CONFIG_HAS_TXX9_SERIAL)   += scc_sio.o
+obj-$(CONFIG_SERIAL_TXX9)       += scc_sio.o
 obj-$(CONFIG_SPU_BASE)          += spu_priv1.o
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/pasemi/idle.c b/arch/powerpc/platforms/pasemi/idle.c
index 03cd45d8fefa..3c962d5757be 100644
--- a/arch/powerpc/platforms/pasemi/idle.c
+++ b/arch/powerpc/platforms/pasemi/idle.c
@@ -26,6 +26,7 @@
 
 #include <asm/machdep.h>
 #include <asm/reg.h>
+#include <asm/smp.h>
 
 #include "pasemi.h"
 
diff --git a/arch/powerpc/platforms/pasemi/iommu.c b/arch/powerpc/platforms/pasemi/iommu.c
index 95fa6a7d15ee..f33b21b9f5d4 100644
--- a/arch/powerpc/platforms/pasemi/iommu.c
+++ b/arch/powerpc/platforms/pasemi/iommu.c
@@ -31,8 +31,6 @@
 #define IOBMAP_PAGE_SIZE        (1 << IOBMAP_PAGE_SHIFT)
 #define IOBMAP_PAGE_MASK        (IOBMAP_PAGE_SIZE - 1)
 
-#define IOBMAP_PAGE_FACTOR      (PAGE_SHIFT - IOBMAP_PAGE_SHIFT)
-
 #define IOB_BASE                0xe0000000
 #define IOB_SIZE                0x3000
 /* Configuration registers */
@@ -97,9 +95,6 @@ static void iobmap_build(struct iommu_table *tbl, long index,
 
         bus_addr = (tbl->it_offset + index) << PAGE_SHIFT;
 
-        npages <<= IOBMAP_PAGE_FACTOR;
-        index <<= IOBMAP_PAGE_FACTOR;
-
         ip = ((u32 *)tbl->it_base) + index;
 
         while (npages--) {
@@ -125,9 +120,6 @@ static void iobmap_free(struct iommu_table *tbl, long index,
 
         bus_addr = (tbl->it_offset + index) << PAGE_SHIFT;
 
-        npages <<= IOBMAP_PAGE_FACTOR;
-        index <<= IOBMAP_PAGE_FACTOR;
-
         ip = ((u32 *)tbl->it_base) + index;
 
         while (npages--) {
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/platforms/ps3/interrupt.c b/arch/powerpc/platforms/ps3/interrupt.c
index 9da82c266ba9..ec9030dbb5f1 100644
--- a/arch/powerpc/platforms/ps3/interrupt.c
+++ b/arch/powerpc/platforms/ps3/interrupt.c
@@ -25,6 +25,7 @@
 #include <asm/machdep.h>
 #include <asm/udbg.h>
 #include <asm/lv1call.h>
+#include <asm/smp.h>
 
 #include "platform.h"
 
diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c
index b854e7f1001c..f1df942072bb 100644
--- a/arch/powerpc/platforms/pseries/xics.c
+++ b/arch/powerpc/platforms/pseries/xics.c
@@ -752,6 +752,7 @@ skip_gserver_check:
 void xics_request_IPIs(void)
 {
         unsigned int ipi;
+        int rc;
 
         ipi = irq_create_mapping(xics_host, XICS_IPI);
         BUG_ON(ipi == NO_IRQ);
@@ -762,11 +763,12 @@ void xics_request_IPIs(void)
          */
         set_irq_handler(ipi, handle_percpu_irq);
         if (firmware_has_feature(FW_FEATURE_LPAR))
-                request_irq(ipi, xics_ipi_action_lpar, IRQF_DISABLED,
-                            "IPI", NULL);
+                rc = request_irq(ipi, xics_ipi_action_lpar, IRQF_DISABLED,
+                                "IPI", NULL);
         else
-                request_irq(ipi, xics_ipi_action_direct, IRQF_DISABLED,
-                            "IPI", NULL);
+                rc = request_irq(ipi, xics_ipi_action_direct, IRQF_DISABLED,
+                                "IPI", NULL);
+        BUG_ON(rc);
 }
 #endif /* CONFIG_SMP */
 
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
-        select UCC
+        default y if UCC_GETH
         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
-        .comm   primary_pteg_full,4
-        .comm   htab_hash_searches,4
+        .section .bss
+        .align  2
+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/ppc/syslib/ibm_ocp.c b/arch/ppc/syslib/ibm_ocp.c
index 3f6e55c79181..2ee176610e7c 100644
--- a/arch/ppc/syslib/ibm_ocp.c
+++ b/arch/ppc/syslib/ibm_ocp.c
@@ -1,4 +1,5 @@
 #include <linux/module.h>
+#include <asm/ibm4xx.h>
 #include <asm/ocp.h>
 
 struct ocp_sys_info_data ocp_sys_info = {
diff --git a/arch/ppc/syslib/qspan_pci.c b/arch/ppc/syslib/qspan_pci.c
index 85053b2816a9..7a97c7440b30 100644
--- a/arch/ppc/syslib/qspan_pci.c
+++ b/arch/ppc/syslib/qspan_pci.c
@@ -365,13 +365,13 @@ int qspan_pcibios_find_class(unsigned int class_code, unsigned short index,
 }
 
 void __init
-m8xx_pcibios_fixup(void))
+m8xx_pcibios_fixup(void)
 {
    /* Lots to do here, all board and configuration specific. */
 }
 
 void __init
-m8xx_setup_pci_ptrs(void))
+m8xx_setup_pci_ptrs(void)
 {
         set_config_access_method(qspan);
 
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/boards/landisk/gio.c b/arch/sh/boards/landisk/gio.c
index 50d38be62f01..a37643d002b2 100644
--- a/arch/sh/boards/landisk/gio.c
+++ b/arch/sh/boards/landisk/gio.c
@@ -69,7 +69,7 @@ static int gio_ioctl(struct inode *inode, struct file *filp,
         }
 
         switch (cmd) {
-        case GIODRV_IOCSGIOSETADDR:     /* addres set */
+        case GIODRV_IOCSGIOSETADDR:     /* address set */
                 addr = data;
                 break;
 
diff --git a/arch/sh/boards/landisk/setup.c b/arch/sh/boards/landisk/setup.c
index 4058b4f50d44..f953c7427769 100644
--- a/arch/sh/boards/landisk/setup.c
+++ b/arch/sh/boards/landisk/setup.c
@@ -44,8 +44,14 @@ static struct platform_device cf_ide_device = {
         },
 };
 
+static struct platform_device rtc_device = {
+        .name           = "rs5c313",
+        .id             = -1,
+};
+
 static struct platform_device *landisk_devices[] __initdata = {
         &cf_ide_device,
+        &rtc_device,
 };
 
 static int __init landisk_devices_setup(void)
diff --git a/arch/sh/boards/renesas/r7780rp/Makefile b/arch/sh/boards/renesas/r7780rp/Makefile
index 609e5d50dde8..b1d20afb4eb3 100644
--- a/arch/sh/boards/renesas/r7780rp/Makefile
+++ b/arch/sh/boards/renesas/r7780rp/Makefile
@@ -3,5 +3,8 @@
 #
 irqinit-y                       := irq-r7780rp.o
 irqinit-$(CONFIG_SH_R7785RP)    := irq-r7785rp.o
+obj-y                           := setup.o irq.o $(irqinit-y)
+
+ifneq ($(CONFIG_SH_R7785RP),y)
 obj-$(CONFIG_PUSH_SWITCH)       += psw.o
-obj-y                           := setup.o irq.o $(irqinit-y)
+endif
diff --git a/arch/sh/boards/se/73180/setup.c b/arch/sh/boards/se/73180/setup.c
index 911ce1cdbd7f..e143017c8975 100644
--- a/arch/sh/boards/se/73180/setup.c
+++ b/arch/sh/boards/se/73180/setup.c
@@ -38,8 +38,8 @@ static struct platform_device *se73180_devices[] __initdata = {
 
 static int __init se73180_devices_setup(void)
 {
-        return platform_add_devices(sh7343se_platform_devices,
-                                    ARRAY_SIZE(sh7343se_platform_devices));
+        return platform_add_devices(se73180_devices,
+                                    ARRAY_SIZE(se73180_devices));
 }
 __initcall(se73180_devices_setup);
 
diff --git a/arch/sh/boards/snapgear/rtc.c b/arch/sh/boards/snapgear/rtc.c
index 1659fdd6695a..edb3dd936cbb 100644
--- a/arch/sh/boards/snapgear/rtc.c
+++ b/arch/sh/boards/snapgear/rtc.c
@@ -108,7 +108,7 @@ static void ds1302_writebyte(unsigned int addr, unsigned int val)
 static void ds1302_reset(void)
 {
         unsigned long   flags;
-        /* Hardware dependant reset/init */
+        /* Hardware dependent reset/init */
         local_irq_save(flags);
         set_dirp(get_dirp() | RTC_RESET | RTC_IODATA | RTC_SCLK);
         set_dp(get_dp() & ~(RTC_RESET | RTC_IODATA | RTC_SCLK));
diff --git a/arch/sh/boards/superh/microdev/io.c b/arch/sh/boards/superh/microdev/io.c
index 83419bf4c834..b704e20d7e4d 100644
--- a/arch/sh/boards/superh/microdev/io.c
+++ b/arch/sh/boards/superh/microdev/io.c
@@ -198,12 +198,12 @@ void microdev_outb(unsigned char b, unsigned long port)
         /*
          *      There is a board feature with the current SH4-202 MicroDev in
          *      that the 2 byte enables (nBE0 and nBE1) are tied together (and
-         *      to the Chip Select Line (Ethernet_CS)). Due to this conectivity,
+         *      to the Chip Select Line (Ethernet_CS)). Due to this connectivity,
          *      it is not possible to safely perform 8-bit writes to the
          *      Ethernet registers, as 16-bits will be consumed from the Data
          *      lines (corrupting the other byte).  Hence, this function is
-         *      written to impliment 16-bit read/modify/write for all byte-wide
-         *      acceses.
+         *      written to implement 16-bit read/modify/write for all byte-wide
+         *      accesses.
          *
          *      Note: there is no problem with byte READS (even or odd).
          *
diff --git a/arch/sh/boards/superh/microdev/irq.c b/arch/sh/boards/superh/microdev/irq.c
index 8c64baa30364..4d335077a3ff 100644
--- a/arch/sh/boards/superh/microdev/irq.c
+++ b/arch/sh/boards/superh/microdev/irq.c
@@ -11,6 +11,7 @@
 
 #include <linux/init.h>
 #include <linux/irq.h>
+#include <linux/interrupt.h>
 #include <asm/system.h>
 #include <asm/io.h>
 #include <asm/microdev.h>
@@ -100,7 +101,7 @@ static void disable_microdev_irq(unsigned int irq)
 
         fpgaIrq = fpgaIrqTable[irq].fpgaIrq;
 
-        /* disable interupts on the FPGA INTC register */
+        /* disable interrupts on the FPGA INTC register */
         ctrl_outl(MICRODEV_FPGA_INTC_MASK(fpgaIrq), MICRODEV_FPGA_INTDSB_REG);
 }
 
@@ -125,7 +126,7 @@ static void enable_microdev_irq(unsigned int irq)
         priorities |= MICRODEV_FPGA_INTPRI_LEVEL(fpgaIrq, pri);
         ctrl_outl(priorities, priorityReg);
 
-        /* enable interupts on the FPGA INTC register */
+        /* enable interrupts on the FPGA INTC register */
         ctrl_outl(MICRODEV_FPGA_INTC_MASK(fpgaIrq), MICRODEV_FPGA_INTENB_REG);
 }
 
@@ -152,7 +153,7 @@ extern void __init init_microdev_irq(void)
 {
         int i;
 
-                /* disable interupts on the FPGA INTC register */
+                /* disable interrupts on the FPGA INTC register */
         ctrl_outl(~0ul, MICRODEV_FPGA_INTDSB_REG);
 
         for (i = 0; i < NUM_EXTERNAL_IRQS; i++)
diff --git a/arch/sh/boards/superh/microdev/setup.c b/arch/sh/boards/superh/microdev/setup.c
index 031c814e6e76..6396cea1c896 100644
--- a/arch/sh/boards/superh/microdev/setup.c
+++ b/arch/sh/boards/superh/microdev/setup.c
@@ -349,7 +349,7 @@ static int __init smsc_superio_setup(void)
         SMSC_WRITE_INDEXED(0x00, 0xc7); /* GP47 = nIOWOP */
         SMSC_WRITE_INDEXED(0x08, 0xe8); /* GP20 = nIDE2_OE */
 
-                /* Exit the configuraton state */
+                /* Exit the configuration state */
         outb(SMSC_EXIT_CONFIG_KEY, SMSC_CONFIG_PORT_ADDR);
 
         return 0;
diff --git a/arch/sh/boards/unknown/setup.c b/arch/sh/boards/unknown/setup.c
index 1c941370a2e3..bee4612de59b 100644
--- a/arch/sh/boards/unknown/setup.c
+++ b/arch/sh/boards/unknown/setup.c
@@ -6,7 +6,7 @@
  * May be copied or modified under the terms of the GNU General Public
  * License.  See linux/COPYING for more information.
  *
- * Setup code for an unknown machine (internal peripherials only)
+ * Setup code for an unknown machine (internal peripherals only)
  *
  * This is the simplest of all boards, and serves only as a quick and dirty
  * method to start debugging a new board during bring-up until proper board
diff --git a/arch/sh/cchips/voyagergx/irq.c b/arch/sh/cchips/voyagergx/irq.c
index 70f12907647f..d70e5c8461b5 100644
--- a/arch/sh/cchips/voyagergx/irq.c
+++ b/arch/sh/cchips/voyagergx/irq.c
@@ -28,7 +28,7 @@ static void disable_voyagergx_irq(unsigned int irq)
         unsigned long val;
         unsigned long mask = 1 << (irq - VOYAGER_IRQ_BASE);
 
-        pr_debug("disable_voyagergx_irq(%d): mask=%x\n", irq, mask);
+        pr_debug("disable_voyagergx_irq(%d): mask=%lx\n", irq, mask);
         val = readl((void __iomem *)VOYAGER_INT_MASK);
         val &= ~mask;
         writel(val, (void __iomem *)VOYAGER_INT_MASK);
@@ -39,7 +39,7 @@ static void enable_voyagergx_irq(unsigned int irq)
         unsigned long val;
         unsigned long mask = 1 << (irq - VOYAGER_IRQ_BASE);
 
-        pr_debug("disable_voyagergx_irq(%d): mask=%x\n", irq, mask);
+        pr_debug("disable_voyagergx_irq(%d): mask=%lx\n", irq, mask);
         val = readl((void __iomem *)VOYAGER_INT_MASK);
         val |= mask;
         writel(val, (void __iomem *)VOYAGER_INT_MASK);
@@ -125,11 +125,12 @@ int voyagergx_irq_demux(int irq)
                         i = 17;
                 else
                         printk("Unexpected IRQ irq = %d status = 0x%08lx\n", irq, val);
-                pr_debug("voyagergx_irq_demux %d \n", i);
-                if (i < VOYAGER_IRQ_NUM) {
+                pr_debug("voyagergx_irq_demux %ld \n", i);
+                if (i < VOYAGER_IRQ_NUM) {
                         irq = VOYAGER_IRQ_BASE + i;
-                        if (voyagergx_demux[i].func != 0)
-                                irq = voyagergx_demux[i].func(irq, voyagergx_demux[i].dev);
+                        if (voyagergx_demux[i].func != 0)
+                                irq = voyagergx_demux[i].func(irq,
+                                                voyagergx_demux[i].dev);
                 }
         }
         return irq;
diff --git a/arch/sh/drivers/dma/dma-api.c b/arch/sh/drivers/dma/dma-api.c
index e062067edd24..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);
@@ -115,7 +116,7 @@ static int search_cap(const char **haystack, const char *needle)
 /**
  * request_dma_bycap - Allocate a DMA channel based on its capabilities
  * @dmac: List of DMA controllers to search
- * @caps: List of capabilites
+ * @caps: List of capabilities
  *
  * Search all channels of all DMA controllers to find a channel which
  * matches the requested capabilities. The result is the channel
diff --git a/arch/sh/drivers/dma/dma-isa.c b/arch/sh/drivers/dma/dma-isa.c
index 05a74ffdb68d..5fb044b791c3 100644
--- a/arch/sh/drivers/dma/dma-isa.c
+++ b/arch/sh/drivers/dma/dma-isa.c
@@ -28,7 +28,7 @@
  * NOTE: ops->xfer() is the preferred way of doing things. However, there
  * are some users of the ISA DMA API that exist in common code that we
  * don't necessarily want to go out of our way to break, so we still
- * allow for some compatability at that level. Any new code is strongly
+ * allow for some compatibility at that level. Any new code is strongly
  * advised to run far away from the ISA DMA API and use the SH DMA API
  * directly.
  */
diff --git a/arch/sh/drivers/dma/dmabrg.c b/arch/sh/drivers/dma/dmabrg.c
index 9d0a29370f21..5e22689c2fcf 100644
--- a/arch/sh/drivers/dma/dmabrg.c
+++ b/arch/sh/drivers/dma/dmabrg.c
@@ -33,7 +33,7 @@
  *     9      | HAC1/SSI1 | rec | half done    | DMABRGI2
  *
  * all can be enabled/disabled in the DMABRGCR register,
- * as well as checked if they occured.
+ * as well as checked if they occurred.
  *
  * DMABRGI0 services  USB  DMA  Address  errors,  but it still must be
  * enabled/acked in the DMABRGCR register.  USB-DMA complete indicator
diff --git a/arch/sh/drivers/pci/ops-dreamcast.c b/arch/sh/drivers/pci/ops-dreamcast.c
index 381306cf5425..e1284fc69361 100644
--- a/arch/sh/drivers/pci/ops-dreamcast.c
+++ b/arch/sh/drivers/pci/ops-dreamcast.c
@@ -57,7 +57,7 @@ struct pci_channel board_pci_channels[] = {
  *
  * Also, we could very easily support both Type 0 and Type 1 configurations
  * here, but since it doesn't seem that there is any such implementation in
- * existance, we don't bother.
+ * existence, we don't bother.
  *
  * I suppose if someone actually gets around to ripping the chip out of
  * the BBA and hanging some more devices off of it, then this might be
diff --git a/arch/sh/drivers/pci/pci-st40.c b/arch/sh/drivers/pci/pci-st40.c
index d67656a44b15..543417ff8314 100644
--- a/arch/sh/drivers/pci/pci-st40.c
+++ b/arch/sh/drivers/pci/pci-st40.c
@@ -292,7 +292,7 @@ int __init st40pci_init(unsigned memStart, unsigned memSize)
                             PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER |
                             PCI_COMMAND_IO);
 
-        /* Accesse to the 0xb0000000 -> 0xb6000000 area will go through to 0x10000000 -> 0x16000000
+        /* Access to the 0xb0000000 -> 0xb6000000 area will go through to 0x10000000 -> 0x16000000
          * on the PCI bus. This allows a nice 1-1 bus to phys mapping.
          */
 
@@ -315,7 +315,7 @@ int __init st40pci_init(unsigned memStart, unsigned memSize)
         ST40PCI_WRITE(CSR_MBAR0, 0);
         ST40PCI_WRITE(LSR0, 0x0fff0001);
 
-        /* ... and set up the initial incomming window to expose all of RAM */
+        /* ... and set up the initial incoming window to expose all of RAM */
         pci_set_rbar_region(7, memStart, memStart, memSize);
 
         /* Maximise timeout values */
@@ -473,7 +473,7 @@ static void pci_set_rbar_region(unsigned int region,     unsigned long localAddr
 
         mask = r2p2(regionSize) - 0x10000;
 
-        /* Diable the region (in case currently in use, should never happen) */
+        /* Disable the region (in case currently in use, should never happen) */
         ST40PCI_WRITE_INDEXED(RSR, region, 0);
 
         /* Start of local address space to publish */
diff --git a/arch/sh/drivers/pci/pci-st40.h b/arch/sh/drivers/pci/pci-st40.h
index d729e0c2d5fe..cf0d35bd135c 100644
--- a/arch/sh/drivers/pci/pci-st40.h
+++ b/arch/sh/drivers/pci/pci-st40.h
@@ -4,7 +4,7 @@
  * May be copied or modified under the terms of the GNU General Public
  * License.  See linux/COPYING for more information.                            
  *
- * Defintions for the ST40 PCI hardware.
+ * Definitions for the ST40 PCI hardware.
  */
 
 #ifndef __PCI_ST40_H__
diff --git a/arch/sh/drivers/superhyway/ops-sh4-202.c b/arch/sh/drivers/superhyway/ops-sh4-202.c
index a55c98a9052b..3b14bf860db6 100644
--- a/arch/sh/drivers/superhyway/ops-sh4-202.c
+++ b/arch/sh/drivers/superhyway/ops-sh4-202.c
@@ -130,7 +130,7 @@ static int sh4202_read_vcr(unsigned long base, struct superhyway_vcr_info *vcr)
          * Some modules (PBR and ePBR for instance) also appear to have
          * VCRL/VCRH flipped in the documentation, but on the SH4-202
          * itself it appears that these are all consistently mapped with
-         * VCRH preceeding VCRL.
+         * VCRH preceding VCRL.
          *
          * Do not trust the documentation, for it is evil.
          */
diff --git a/arch/sh/kernel/cf-enabler.c b/arch/sh/kernel/cf-enabler.c
index 0758d48147a0..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>
 
@@ -31,7 +32,7 @@
  */
 #if defined(CONFIG_CPU_SH4)
 /* SH4 can't access PCMCIA interface through P2 area.
- * we must remap it with appropreate attribute bit of the page set.
+ * we must remap it with appropriate attribute bit of the page set.
  * this part is based on Greg Banks' hd64465_ss.c implementation - Masahiro Abe */
 
 #if defined(CONFIG_CF_AREA6)
@@ -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/clock.c b/arch/sh/kernel/cpu/clock.c
index 014f318f5a05..63251549e9a8 100644
--- a/arch/sh/kernel/cpu/clock.c
+++ b/arch/sh/kernel/cpu/clock.c
@@ -278,6 +278,11 @@ arch_init_clk_ops(struct clk_ops **ops, int type)
 {
 }
 
+void __init __attribute__ ((weak))
+arch_clk_init(void)
+{
+}
+
 static int show_clocks(char *buf, char **start, off_t off,
                        int len, int *eof, void *data)
 {
@@ -314,6 +319,8 @@ int __init clk_init(void)
                 ret |= clk_register(clk);
         }
 
+        arch_clk_init();
+
         /* Kick the child clocks.. */
         propagate_rate(&master_clk);
         propagate_rate(&bus_clk);
diff --git a/arch/sh/kernel/cpu/irq/maskreg.c b/arch/sh/kernel/cpu/irq/maskreg.c
index 492db31b3cab..978992e367a5 100644
--- a/arch/sh/kernel/cpu/irq/maskreg.c
+++ b/arch/sh/kernel/cpu/irq/maskreg.c
@@ -38,7 +38,7 @@ static struct hw_interrupt_type maskreg_irq_type = {
         .end = end_maskreg_irq
 };
 
-/* actual implementatin */
+/* actual implementation */
 static unsigned int startup_maskreg_irq(unsigned int irq)
 {
         enable_maskreg_irq(irq);
diff --git a/arch/sh/kernel/cpu/sh3/entry.S b/arch/sh/kernel/cpu/sh3/entry.S
index 832c0b4a1e6c..b0b59d4a33ca 100644
--- a/arch/sh/kernel/cpu/sh3/entry.S
+++ b/arch/sh/kernel/cpu/sh3/entry.S
@@ -320,6 +320,9 @@ skip_restore:
 
         .align  2
 5:      .long   0x00001000      ! DSP
+#ifdef CONFIG_KGDB_NMI
+6:      .long   in_nmi
+#endif
 7:      .long   0x30000000
 
 ! common exception handler
diff --git a/arch/sh/kernel/cpu/sh4/clock-sh4-202.c b/arch/sh/kernel/cpu/sh4/clock-sh4-202.c
index fcb2c41bc34e..a33429463e96 100644
--- a/arch/sh/kernel/cpu/sh4/clock-sh4-202.c
+++ b/arch/sh/kernel/cpu/sh4/clock-sh4-202.c
@@ -111,7 +111,7 @@ static int shoc_clk_verify_rate(struct clk *clk, unsigned long rate)
         return 0;
 }
 
-static int shoc_clk_set_rate(struct clk *clk, unsigned long rate)
+static int shoc_clk_set_rate(struct clk *clk, unsigned long rate, int algo_id)
 {
         unsigned long frqcr3;
         unsigned int tmp;
diff --git a/arch/sh/kernel/cpu/sh4/fpu.c b/arch/sh/kernel/cpu/sh4/fpu.c
index d61dd599169f..c5a4fc77fa06 100644
--- a/arch/sh/kernel/cpu/sh4/fpu.c
+++ b/arch/sh/kernel/cpu/sh4/fpu.c
@@ -138,7 +138,7 @@ restore_fpu(struct task_struct *tsk)
 /*
  * Load the FPU with signalling NANS.  This bit pattern we're using
  * has the property that no matter wether considered as single or as
- * double precission represents signaling NANS.  
+ * double precision represents signaling NANS.  
  */
 
 static void
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/cpu/sh4/setup-sh7750.c b/arch/sh/kernel/cpu/sh4/setup-sh7750.c
index 6f8f458912c7..03b14cf78ddf 100644
--- a/arch/sh/kernel/cpu/sh4/setup-sh7750.c
+++ b/arch/sh/kernel/cpu/sh4/setup-sh7750.c
@@ -106,6 +106,7 @@ static struct ipr_data sh7750_ipr_map[] = {
         { 38, 2,  8, 7 }, /* DMAC DMAE */
 };
 
+#ifdef CONFIG_CPU_SUBTYPE_SH7751
 static struct ipr_data sh7751_ipr_map[] = {
         { 44, 2,  8, 7 }, /* DMAC DMTE4 */
         { 45, 2,  8, 7 }, /* DMAC DMTE5 */
@@ -117,6 +118,7 @@ static struct ipr_data sh7751_ipr_map[] = {
         /*{ 72, INTPRI00,  8, ? },*/ /* TMU3 TUNI */
         /*{ 76, INTPRI00, 12, ? },*/ /* TMU4 TUNI */
 };
+#endif
 
 static unsigned long ipr_offsets[] = {
         0xffd00004UL,   /* 0: IPRA */
diff --git a/arch/sh/kernel/cpu/sh4a/clock-sh7722.c b/arch/sh/kernel/cpu/sh4a/clock-sh7722.c
index 29090035bc5b..51b386d454de 100644
--- a/arch/sh/kernel/cpu/sh4a/clock-sh7722.c
+++ b/arch/sh/kernel/cpu/sh4a/clock-sh7722.c
@@ -17,7 +17,6 @@
 #include <asm/clock.h>
 #include <asm/freq.h>
 
-#define SH7722_PLL_FREQ (32000000/8)
 #define N  (-1)
 #define NM (-2)
 #define ROUND_NEAREST 0
@@ -141,28 +140,36 @@ static void adjust_clocks(int originate, int *l, unsigned long v[],
 */
 static int divisors2[] = { 2, 3, 4, 5, 6, 8, 10, 12, 16, 20, 24, 32, 40 };
 
+static void master_clk_recalc(struct clk *clk)
+{
+        unsigned frqcr = ctrl_inl(FRQCR);
+
+        clk->rate = CONFIG_SH_PCLK_FREQ * (((frqcr >> 24) & 0x1f) + 1);
+}
+
 static void master_clk_init(struct clk *clk)
 {
-        clk_set_rate(clk, clk_get_rate(clk));
+        clk->parent = NULL;
+        clk->flags |= CLK_RATE_PROPAGATES;
+        clk->rate = CONFIG_SH_PCLK_FREQ;
+        master_clk_recalc(clk);
 }
 
-static void master_clk_recalc(struct clk *clk)
+
+static void module_clk_recalc(struct clk *clk)
 {
         unsigned long frqcr = ctrl_inl(FRQCR);
 
-        clk->rate = CONFIG_SH_PCLK_FREQ * (1 + (frqcr >> 24 & 0xF));
+        clk->rate = clk->parent->rate / (((frqcr >> 24) & 0x1f) + 1);
 }
 
 static int master_clk_setrate(struct clk *clk, unsigned long rate, int id)
 {
-        int div = rate / SH7722_PLL_FREQ;
+        int div = rate / clk->rate;
         int master_divs[] = { 2, 3, 4, 6, 8, 16 };
         int index;
         unsigned long frqcr;
 
-        if (rate < SH7722_PLL_FREQ * 2)
-                return -EINVAL;
-
         for (index = 1; index < ARRAY_SIZE(master_divs); index++)
                 if (div >= master_divs[index - 1] && div < master_divs[index])
                         break;
@@ -185,6 +192,10 @@ static struct clk_ops sh7722_master_clk_ops = {
         .set_rate = master_clk_setrate,
 };
 
+static struct clk_ops sh7722_module_clk_ops = {
+       .recalc = module_clk_recalc,
+};
+
 struct frqcr_context {
         unsigned mask;
         unsigned shift;
@@ -489,7 +500,7 @@ static void sh7722_siu_recalc(struct clk *clk)
 
         if (siu < 0)
                 return /* siu */ ;
-        BUG_ON(siu > 1);
+        BUG_ON(siu > 2);
         r = ctrl_inl(sh7722_siu_regs[siu]);
         clk->rate = clk->parent->rate * 2 / divisors2[r & 0xF];
 }
@@ -571,7 +582,7 @@ static struct clk *sh7722_clocks[] = {
  */
 struct clk_ops *onchip_ops[] = {
         &sh7722_master_clk_ops,
-        &sh7722_frqcr_clk_ops,
+        &sh7722_module_clk_ops,
         &sh7722_frqcr_clk_ops,
         &sh7722_frqcr_clk_ops,
 };
@@ -583,7 +594,7 @@ arch_init_clk_ops(struct clk_ops **ops, int type)
         *ops = onchip_ops[type];
 }
 
-int __init sh7722_clock_init(void)
+int __init arch_clk_init(void)
 {
         struct clk *master;
         int i;
@@ -597,4 +608,3 @@ int __init sh7722_clock_init(void)
         clk_put(master);
         return 0;
 }
-arch_initcall(sh7722_clock_init);
diff --git a/arch/sh/kernel/kgdb_stub.c b/arch/sh/kernel/kgdb_stub.c
index a5323364cbca..edd1ec214e6d 100644
--- a/arch/sh/kernel/kgdb_stub.c
+++ b/arch/sh/kernel/kgdb_stub.c
@@ -2,7 +2,7 @@
  * May be copied or modified under the terms of the GNU General Public
  * License.  See linux/COPYING for more information.
  *
- * Containes extracts from code by Glenn Engel, Jim Kingdon,
+ * Contains extracts from code by Glenn Engel, Jim Kingdon,
  * David Grothe <dave@gcom.com>, Tigran Aivazian <tigran@sco.com>,
  * Amit S. Kale <akale@veritas.com>,  William Gatliff <bgat@open-widgets.com>,
  * Ben Lee, Steve Chamberlain and Benoit Miller <fulg@iname.com>.
@@ -85,7 +85,7 @@
  *
  * Responses can be run-length encoded to save space.  A '*' means that
  * the next character is an ASCII encoding giving a repeat count which
- * stands for that many repititions of the character preceding the '*'.
+ * stands for that many repetitions of the character preceding the '*'.
  * The encoding is n+29, yielding a printable character where n >=3
  * (which is where RLE starts to win).  Don't use an n > 126.
  *
diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c
index 6b4f5748d0be..a11e2aa73cbc 100644
--- a/arch/sh/kernel/process.c
+++ b/arch/sh/kernel/process.c
@@ -26,8 +26,6 @@
 static int hlt_counter;
 int ubc_usercnt = 0;
 
-#define HARD_IDLE_TIMEOUT (HZ / 3)
-
 void (*pm_idle)(void);
 void (*pm_power_off)(void);
 EXPORT_SYMBOL(pm_power_off);
@@ -44,16 +42,39 @@ void enable_hlt(void)
 }
 EXPORT_SYMBOL(enable_hlt);
 
+static int __init nohlt_setup(char *__unused)
+{
+        hlt_counter = 1;
+        return 1;
+}
+__setup("nohlt", nohlt_setup);
+
+static int __init hlt_setup(char *__unused)
+{
+        hlt_counter = 0;
+        return 1;
+}
+__setup("hlt", hlt_setup);
+
 void default_idle(void)
 {
-        if (!hlt_counter)
-                cpu_sleep();
-        else
-                cpu_relax();
+        if (!hlt_counter) {
+                clear_thread_flag(TIF_POLLING_NRFLAG);
+                smp_mb__after_clear_bit();
+                set_bl_bit();
+                while (!need_resched())
+                        cpu_sleep();
+                clear_bl_bit();
+                set_thread_flag(TIF_POLLING_NRFLAG);
+        } else
+                while (!need_resched())
+                        cpu_relax();
 }
 
 void cpu_idle(void)
 {
+        set_thread_flag(TIF_POLLING_NRFLAG);
+
         /* endless idle loop with no priority at all */
         while (1) {
                 void (*idle)(void) = pm_idle;
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/syscalls.S b/arch/sh/kernel/syscalls.S
index 4357d1a6358f..7db1c2dc5992 100644
--- a/arch/sh/kernel/syscalls.S
+++ b/arch/sh/kernel/syscalls.S
@@ -355,3 +355,6 @@ ENTRY(sys_call_table)
         .long sys_getcpu
         .long sys_epoll_pwait
         .long sys_utimensat             /* 320 */
+        .long sys_signalfd
+        .long sys_timerfd
+        .long sys_eventfd
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/traps.c b/arch/sh/kernel/traps.c
index 3a197649cd83..5b75cb6f8f9b 100644
--- a/arch/sh/kernel/traps.c
+++ b/arch/sh/kernel/traps.c
@@ -21,6 +21,7 @@
 #include <linux/bug.h>
 #include <linux/debug_locks.h>
 #include <linux/kdebug.h>
+#include <linux/kexec.h>
 #include <linux/limits.h>
 #include <asm/system.h>
 #include <asm/uaccess.h>
@@ -101,6 +102,16 @@ void die(const char * str, struct pt_regs * regs, long err)
 
         bust_spinlocks(0);
         spin_unlock_irq(&die_lock);
+
+        if (kexec_should_crash(current))
+                crash_kexec(regs);
+
+        if (in_interrupt())
+                panic("Fatal exception in interrupt");
+
+        if (panic_on_oops)
+                panic("Fatal exception");
+
         do_exit(SIGSEGV);
 }
 
@@ -513,7 +524,7 @@ static int handle_unaligned_access(u16 instruction, struct pt_regs *regs)
  *       misaligned data access
  *       access to >= 0x80000000 is user mode
  * Unfortuntaly we can't distinguish between instruction address error
- * and data address errors caused by read acceses.
+ * and data address errors caused by read accesses.
  */
 asmlinkage void do_address_error(struct pt_regs *regs,
                                  unsigned long writeaccess,
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/sh/math-emu/math.c b/arch/sh/math-emu/math.c
index 1efbac15ff4e..a38e1eed9e77 100644
--- a/arch/sh/math-emu/math.c
+++ b/arch/sh/math-emu/math.c
@@ -148,7 +148,7 @@ fmac(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
         return 0;
 }
 
-// to process fmov's extention (odd n for DR access XD).
+// to process fmov's extension (odd n for DR access XD).
 #define FMOV_EXT(x) if(x&1) x+=16-1
 
 static int
diff --git a/arch/sh/mm/copy_page.S b/arch/sh/mm/copy_page.S
index 397c94c97315..ae039f2da162 100644
--- a/arch/sh/mm/copy_page.S
+++ b/arch/sh/mm/copy_page.S
@@ -129,6 +129,7 @@ ENTRY(__copy_user_page)
         rts
          nop
 #endif
+        .align 2
 .Lpsz:  .long   PAGE_SIZE
 /*
  * __kernel_size_t __copy_user(void *to, const void *from, __kernel_size_t n);
diff --git a/arch/sh/mm/fault.c b/arch/sh/mm/fault.c
index 9207da67ff8a..c878faa4ae46 100644
--- a/arch/sh/mm/fault.c
+++ b/arch/sh/mm/fault.c
@@ -15,43 +15,11 @@
 #include <linux/mm.h>
 #include <linux/hardirq.h>
 #include <linux/kprobes.h>
-#include <linux/kdebug.h>
 #include <asm/system.h>
 #include <asm/mmu_context.h>
 #include <asm/tlbflush.h>
 #include <asm/kgdb.h>
 
-#ifdef CONFIG_KPROBES
-ATOMIC_NOTIFIER_HEAD(notify_page_fault_chain);
-
-/* Hook to register for page fault notifications */
-int register_page_fault_notifier(struct notifier_block *nb)
-{
-        return atomic_notifier_chain_register(&notify_page_fault_chain, nb);
-}
-
-int unregister_page_fault_notifier(struct notifier_block *nb)
-{
-        return atomic_notifier_chain_unregister(&notify_page_fault_chain, nb);
-}
-
-static inline int notify_page_fault(enum die_val val, struct pt_regs *regs,
-                                    int trap, int sig)
-{
-        struct die_args args = {
-                .regs = regs,
-                .trapnr = trap,
-        };
-        return atomic_notifier_call_chain(&notify_page_fault_chain, val, &args);
-}
-#else
-static inline int notify_page_fault(enum die_val val, struct pt_regs *regs,
-                                    int trap, int sig)
-{
-        return NOTIFY_DONE;
-}
-#endif
-
 /*
  * This routine handles page faults.  It determines the address,
  * and the problem, and then passes it off to one of the appropriate
@@ -69,11 +37,6 @@ asmlinkage void __kprobes do_page_fault(struct pt_regs *regs,
         siginfo_t info;
 
         trace_hardirqs_on();
-
-        if (notify_page_fault(DIE_PAGE_FAULT, regs,
-                              writeaccess, SIGSEGV) == NOTIFY_STOP)
-                return;
-
         local_irq_enable();
 
 #ifdef CONFIG_SH_KGDB
@@ -285,7 +248,7 @@ asmlinkage int __kprobes __do_page_fault(struct pt_regs *regs,
         pte_t *pte;
         pte_t entry;
         struct mm_struct *mm = current->mm;
-        spinlock_t *ptl;
+        spinlock_t *ptl = NULL;
         int ret = 1;
 
 #ifdef CONFIG_SH_KGDB
diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index 8fe223a890ed..e0e644ff3204 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -12,6 +12,7 @@
 #include <linux/init.h>
 #include <linux/bootmem.h>
 #include <linux/proc_fs.h>
+#include <linux/pagemap.h>
 #include <linux/percpu.h>
 #include <linux/io.h>
 #include <asm/mmu_context.h>
@@ -112,7 +113,7 @@ static void set_pte_phys(unsigned long addr, unsigned long phys, pgprot_t prot)
  * As a performance optimization, other platforms preserve the fixmap mapping
  * across a context switch, we don't presently do this, but this could be done
  * in a similar fashion as to the wired TLB interface that sh64 uses (by way
- * of the memorry mapped UTLB configuration) -- this unfortunately forces us to
+ * of the memory mapped UTLB configuration) -- this unfortunately forces us to
  * give up a TLB entry for each mapping we want to preserve. While this may be
  * viable for a small number of fixmaps, it's not particularly useful for
  * everything and needs to be carefully evaluated. (ie, we may want this for
diff --git a/arch/sh/mm/ioremap.c b/arch/sh/mm/ioremap.c
index be03d74e99cb..0c7b7e33abdc 100644
--- a/arch/sh/mm/ioremap.c
+++ b/arch/sh/mm/ioremap.c
@@ -22,6 +22,7 @@
 #include <asm/addrspace.h>
 #include <asm/cacheflush.h>
 #include <asm/tlbflush.h>
+#include <asm/mmu.h>
 
 /*
  * Remap an arbitrary physical address space into the kernel virtual
diff --git a/arch/sh/mm/pmb.c b/arch/sh/mm/pmb.c
index 02aae06527dc..b6a5a338145b 100644
--- a/arch/sh/mm/pmb.c
+++ b/arch/sh/mm/pmb.c
@@ -3,7 +3,7 @@
  *
  * Privileged Space Mapping Buffer (PMB) Support.
  *
- * Copyright (C) 2005, 2006 Paul Mundt
+ * Copyright (C) 2005, 2006, 2007 Paul Mundt
  *
  * P1/P2 Section mapping definitions from map32.h, which was:
  *
@@ -68,6 +68,32 @@ static inline unsigned long mk_pmb_data(unsigned int entry)
         return mk_pmb_entry(entry) | PMB_DATA;
 }
 
+static DEFINE_SPINLOCK(pmb_list_lock);
+static struct pmb_entry *pmb_list;
+
+static inline void pmb_list_add(struct pmb_entry *pmbe)
+{
+        struct pmb_entry **p, *tmp;
+
+        p = &pmb_list;
+        while ((tmp = *p) != NULL)
+                p = &tmp->next;
+
+        pmbe->next = tmp;
+        *p = pmbe;
+}
+
+static inline void pmb_list_del(struct pmb_entry *pmbe)
+{
+        struct pmb_entry **p, *tmp;
+
+        for (p = &pmb_list; (tmp = *p); p = &tmp->next)
+                if (tmp == pmbe) {
+                        *p = tmp->next;
+                        return;
+                }
+}
+
 struct pmb_entry *pmb_alloc(unsigned long vpn, unsigned long ppn,
                             unsigned long flags)
 {
@@ -81,11 +107,19 @@ struct pmb_entry *pmb_alloc(unsigned long vpn, unsigned long ppn,
         pmbe->ppn       = ppn;
         pmbe->flags     = flags;
 
+        spin_lock_irq(&pmb_list_lock);
+        pmb_list_add(pmbe);
+        spin_unlock_irq(&pmb_list_lock);
+
         return pmbe;
 }
 
 void pmb_free(struct pmb_entry *pmbe)
 {
+        spin_lock_irq(&pmb_list_lock);
+        pmb_list_del(pmbe);
+        spin_unlock_irq(&pmb_list_lock);
+
         kmem_cache_free(pmb_cache, pmbe);
 }
 
@@ -167,31 +201,6 @@ void clear_pmb_entry(struct pmb_entry *pmbe)
         clear_bit(entry, &pmb_map);
 }
 
-static DEFINE_SPINLOCK(pmb_list_lock);
-static struct pmb_entry *pmb_list;
-
-static inline void pmb_list_add(struct pmb_entry *pmbe)
-{
-        struct pmb_entry **p, *tmp;
-
-        p = &pmb_list;
-        while ((tmp = *p) != NULL)
-                p = &tmp->next;
-
-        pmbe->next = tmp;
-        *p = pmbe;
-}
-
-static inline void pmb_list_del(struct pmb_entry *pmbe)
-{
-        struct pmb_entry **p, *tmp;
-
-        for (p = &pmb_list; (tmp = *p); p = &tmp->next)
-                if (tmp == pmbe) {
-                        *p = tmp->next;
-                        return;
-                }
-}
 
 static struct {
         unsigned long size;
@@ -283,25 +292,14 @@ void pmb_unmap(unsigned long addr)
         } while (pmbe);
 }
 
-static void pmb_cache_ctor(void *pmb, struct kmem_cache *cachep, unsigned long flags)
+static void pmb_cache_ctor(void *pmb, struct kmem_cache *cachep,
+                           unsigned long flags)
 {
         struct pmb_entry *pmbe = pmb;
 
         memset(pmb, 0, sizeof(struct pmb_entry));
 
-        spin_lock_irq(&pmb_list_lock);
-
         pmbe->entry = PMB_NO_ENTRY;
-        pmb_list_add(pmbe);
-
-        spin_unlock_irq(&pmb_list_lock);
-}
-
-static void pmb_cache_dtor(void *pmb, struct kmem_cache *cachep, unsigned long flags)
-{
-        spin_lock_irq(&pmb_list_lock);
-        pmb_list_del(pmb);
-        spin_unlock_irq(&pmb_list_lock);
 }
 
 static int __init pmb_init(void)
@@ -312,8 +310,7 @@ static int __init pmb_init(void)
         BUG_ON(unlikely(nr_entries >= NR_PMB_ENTRIES));
 
         pmb_cache = kmem_cache_create("pmb", sizeof(struct pmb_entry), 0,
-                                      SLAB_PANIC, pmb_cache_ctor,
-                                      pmb_cache_dtor);
+                                      SLAB_PANIC, pmb_cache_ctor, NULL);
 
         jump_to_P2();
 
diff --git a/arch/sh/tools/mach-types b/arch/sh/tools/mach-types
index 554f801db67b..fb40f188aff9 100644
--- a/arch/sh/tools/mach-types
+++ b/arch/sh/tools/mach-types
@@ -7,8 +7,11 @@
 #
 SE                      SH_SOLUTION_ENGINE
 7751SE                  SH_7751_SOLUTION_ENGINE         
+7722SE                  SH_7722_SOLUTION_ENGINE         
 7300SE                  SH_7300_SOLUTION_ENGINE
 7343SE                  SH_7343_SOLUTION_ENGINE
+7206SE                  SH_7206_SOLUTION_ENGINE
+7619SE                  SH_7619_SOLUTION_ENGINE
 7780SE                  SH_7780_SOLUTION_ENGINE
 73180SE                 SH_73180_SOLUTION_ENGINE
 7751SYSTEMH             SH_7751_SYSTEMH
@@ -31,5 +34,3 @@ R7785RP			SH_R7785RP
 TITAN                   SH_TITAN
 SHMIN                   SH_SHMIN
 7710VOIPGW              SH_7710VOIPGW
-7206SE                  SH_7206_SOLUTION_ENGINE
-7619SE                  SH_7619_SOLUTION_ENGINE
diff --git a/arch/sh64/kernel/pci_sh5.c b/arch/sh64/kernel/pci_sh5.c
index fb51660847c8..3334f99b5835 100644
--- a/arch/sh64/kernel/pci_sh5.c
+++ b/arch/sh64/kernel/pci_sh5.c
@@ -521,10 +521,10 @@ void __init pcibios_fixup_bus(struct pci_bus *bus)
                 bus->resource[0]->start = PCIBIOS_MIN_IO;
                 bus->resource[1]->start = PCIBIOS_MIN_MEM;
 #else
-                bus->resource[0]->end = 0
-                bus->resource[1]->end = 0
-                bus->resource[0]->start =0
-                  bus->resource[1]->start = 0;
+                bus->resource[0]->end = 0;
+                bus->resource[1]->end = 0;
+                bus->resource[0]->start =0;
+                bus->resource[1]->start = 0;
 #endif
                 /* Turn off downstream PF memory address range by default */
                 bus->resource[2]->start = 1024*1024;
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..89a1b469b93d 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)"
-        range 2 64
+        int "Maximum number of CPUs (2-1024)"
+        range 2 1024
         depends on SMP
-        default "32"
+        default "64"
 
 source "drivers/cpufreq/Kconfig"
 
@@ -396,6 +396,15 @@ config SCHED_SMT
           when dealing with UltraSPARC cpus at a cost of slightly increased
           overhead in some places. If unsure say N here.
 
+config SCHED_MC
+        bool "Multi-core scheduler support"
+        depends on SMP
+        default y
+        help
+          Multi-core scheduler support improves the CPU scheduler's decision
+          making when dealing with multi-core CPU chips at a cost of slightly
+          increased overhead in some places. If unsure say N here.
+
 source "kernel/Kconfig.preempt"
 
 config CMDLINE_BOOL
diff --git a/arch/sparc64/kernel/Makefile b/arch/sparc64/kernel/Makefile
index c749dccacc32..f964bf28d21a 100644
--- a/arch/sparc64/kernel/Makefile
+++ b/arch/sparc64/kernel/Makefile
@@ -1,4 +1,4 @@
-# $Id: Makefile,v 1.70 2002/02/09 19:49:30 davem Exp $
+#
 # Makefile for the linux kernel.
 #
 
@@ -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 sysfs.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..8059531bf0ac 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
-         * %o1: queue paddr
-         * %o2: num queue entries
+        /* %o0: cpuid
+         * %o1: pc
+         * %o2: rtba
+         * %o3: arg0
          *
          * returns %o0: status
          */
-        .globl  sun4v_cpu_qconf
-sun4v_cpu_qconf:
-        mov     HV_FAST_CPU_QCONF, %o5
+        .globl  sun4v_cpu_start
+        .type   sun4v_cpu_start,#function
+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 +2056,540 @@ 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: channel
+         * %o1: cookie
+         * %o2: 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_svc_send
+        .type   sun4v_svc_send,#function
+sun4v_svc_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_svc_send, .-sun4v_svc_send
+
+        .globl  sun4v_svc_recv
+        .type   sun4v_svc_recv,#function
+sun4v_svc_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_svc_recv, .-sun4v_svc_recv
+
+        .globl  sun4v_svc_getstatus
+        .type   sun4v_svc_getstatus,#function
+sun4v_svc_getstatus:
+        mov     HV_FAST_SVC_GETSTATUS, %o5
+        mov     %o1, %o4
+        ta      HV_FAST_TRAP
+        stx     %o1, [%o4]
+        retl
+         nop
+        .size   sun4v_svc_getstatus, .-sun4v_svc_getstatus
+
+        .globl  sun4v_svc_setstatus
+        .type   sun4v_svc_setstatus,#function
+sun4v_svc_setstatus:
+        mov     HV_FAST_SVC_SETSTATUS, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_svc_setstatus, .-sun4v_svc_setstatus
+
+        .globl  sun4v_svc_clrstatus
+        .type   sun4v_svc_clrstatus,#function
+sun4v_svc_clrstatus:
+        mov     HV_FAST_SVC_CLRSTATUS, %o5
+        ta      HV_FAST_TRAP
+        retl
+         nop
+        .size   sun4v_svc_clrstatus, .-sun4v_svc_clrstatus
+
+        .globl  sun4v_mmustat_conf
+        .type   sun4v_mmustat_conf,#function
+sun4v_mmustat_conf:
+        mov     %o1, %o4
+        mov     HV_FAST_MMUSTAT_CONF, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%o4]
+        retl
+         nop
+        .size   sun4v_mmustat_conf, .-sun4v_mmustat_conf
+
+        .globl  sun4v_mmustat_info
+        .type   sun4v_mmustat_info,#function
+sun4v_mmustat_info:
+        mov     %o0, %o4
+        mov     HV_FAST_MMUSTAT_INFO, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%o4]
+        retl
+         nop
+        .size   sun4v_mmustat_info, .-sun4v_mmustat_info
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..e60d283f60bc 100644
--- a/arch/sparc64/kernel/irq.c
+++ b/arch/sparc64/kernel/irq.c
@@ -1,7 +1,6 @@
-/* $Id: irq.c,v 1.114 2002/01/11 08:45:38 davem Exp $
- * irq.c: UltraSparc IRQ handling/init/registry.
+/* irq.c: UltraSparc IRQ handling/init/registry.
  *
- * Copyright (C) 1997  David S. Miller  (davem@caip.rutgers.edu)
+ * Copyright (C) 1997, 2007  David S. Miller  (davem@davemloft.net)
  * Copyright (C) 1998  Eddie C. Dost    (ecd@skynet.be)
  * Copyright (C) 1998  Jakub Jelinek    (jj@ultra.linux.cz)
  */
@@ -43,6 +42,7 @@
 #include <asm/cpudata.h>
 #include <asm/auxio.h>
 #include <asm/head.h>
+#include <asm/hypervisor.h>
 
 /* UPA nodes send interrupt packet to UltraSparc with first data reg
  * value low 5 (7 on Starfire) bits holding the IRQ identifier being
@@ -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;
@@ -382,6 +380,76 @@ static void sun4v_irq_end(unsigned int virt_irq)
         }
 }
 
+static void sun4v_virq_enable(unsigned int virt_irq)
+{
+        struct ino_bucket *bucket = virt_irq_to_bucket(virt_irq);
+        unsigned int ino = bucket - &ivector_table[0];
+
+        if (likely(bucket)) {
+                unsigned long cpuid, dev_handle, dev_ino;
+                int err;
+
+                cpuid = irq_choose_cpu(virt_irq);
+
+                dev_handle = ino & IMAP_IGN;
+                dev_ino = ino & IMAP_INO;
+
+                err = sun4v_vintr_set_target(dev_handle, dev_ino, cpuid);
+                if (err != HV_EOK)
+                        printk("sun4v_vintr_set_target(%lx,%lx,%lu): "
+                               "err(%d)\n",
+                               dev_handle, dev_ino, cpuid, err);
+                err = sun4v_vintr_set_state(dev_handle, dev_ino,
+                                            HV_INTR_ENABLED);
+                if (err != HV_EOK)
+                        printk("sun4v_vintr_set_state(%lx,%lx,"
+                               "HV_INTR_ENABLED): err(%d)\n",
+                               dev_handle, dev_ino, err);
+        }
+}
+
+static void sun4v_virq_disable(unsigned int virt_irq)
+{
+        struct ino_bucket *bucket = virt_irq_to_bucket(virt_irq);
+        unsigned int ino = bucket - &ivector_table[0];
+
+        if (likely(bucket)) {
+                unsigned long dev_handle, dev_ino;
+                int err;
+
+                dev_handle = ino & IMAP_IGN;
+                dev_ino = ino & IMAP_INO;
+
+                err = sun4v_vintr_set_state(dev_handle, dev_ino,
+                                            HV_INTR_DISABLED);
+                if (err != HV_EOK)
+                        printk("sun4v_vintr_set_state(%lx,%lx,"
+                               "HV_INTR_DISABLED): err(%d)\n",
+                               dev_handle, dev_ino, err);
+        }
+}
+
+static void sun4v_virq_end(unsigned int virt_irq)
+{
+        struct ino_bucket *bucket = virt_irq_to_bucket(virt_irq);
+        unsigned int ino = bucket - &ivector_table[0];
+
+        if (likely(bucket)) {
+                unsigned long dev_handle, dev_ino;
+                int err;
+
+                dev_handle = ino & IMAP_IGN;
+                dev_ino = ino & IMAP_INO;
+
+                err = sun4v_vintr_set_state(dev_handle, dev_ino,
+                                            HV_INTR_STATE_IDLE);
+                if (err != HV_EOK)
+                        printk("sun4v_vintr_set_state(%lx,%lx,"
+                                "HV_INTR_STATE_IDLE): err(%d)\n",
+                               dev_handle, dev_ino, err);
+        }
+}
+
 static void run_pre_handler(unsigned int virt_irq)
 {
         struct ino_bucket *bucket = virt_irq_to_bucket(virt_irq);
@@ -436,6 +504,21 @@ static struct irq_chip sun4v_msi = {
 };
 #endif
 
+static struct irq_chip sun4v_virq = {
+        .typename       = "vsun4v",
+        .enable         = sun4v_virq_enable,
+        .disable        = sun4v_virq_disable,
+        .end            = sun4v_virq_end,
+};
+
+static struct irq_chip sun4v_virq_ack = {
+        .typename       = "vsun4v+ack",
+        .enable         = sun4v_virq_enable,
+        .disable        = sun4v_virq_disable,
+        .ack            = run_pre_handler,
+        .end            = sun4v_virq_end,
+};
+
 void irq_install_pre_handler(int virt_irq,
                              void (*func)(unsigned int, void *, void *),
                              void *arg1, void *arg2)
@@ -449,7 +532,8 @@ void irq_install_pre_handler(int virt_irq,
 
         chip = get_irq_chip(virt_irq);
         if (chip == &sun4u_irq_ack ||
-            chip == &sun4v_irq_ack
+            chip == &sun4v_irq_ack ||
+            chip == &sun4v_virq_ack
 #ifdef CONFIG_PCI_MSI
             || chip == &sun4v_msi
 #endif
@@ -457,7 +541,9 @@ void irq_install_pre_handler(int virt_irq,
                 return;
 
         chip = (chip == &sun4u_irq ?
-                &sun4u_irq_ack : &sun4v_irq_ack);
+                &sun4u_irq_ack :
+                (chip == &sun4v_irq ?
+                 &sun4v_irq_ack : &sun4v_virq_ack));
         set_irq_chip(virt_irq, chip);
 }
 
@@ -494,19 +580,18 @@ out:
         return bucket->virt_irq;
 }
 
-unsigned int sun4v_build_irq(u32 devhandle, unsigned int devino)
+static unsigned int sun4v_build_common(unsigned long sysino,
+                                       struct irq_chip *chip)
 {
         struct ino_bucket *bucket;
         struct irq_handler_data *data;
-        unsigned long sysino;
 
         BUG_ON(tlb_type != hypervisor);
 
-        sysino = sun4v_devino_to_sysino(devhandle, devino);
         bucket = &ivector_table[sysino];
         if (!bucket->virt_irq) {
                 bucket->virt_irq = virt_irq_alloc(__irq(bucket));
-                set_irq_chip(bucket->virt_irq, &sun4v_irq);
+                set_irq_chip(bucket->virt_irq, chip);
         }
 
         data = get_irq_chip_data(bucket->virt_irq);
@@ -531,6 +616,32 @@ out:
         return bucket->virt_irq;
 }
 
+unsigned int sun4v_build_irq(u32 devhandle, unsigned int devino)
+{
+        unsigned long sysino = sun4v_devino_to_sysino(devhandle, devino);
+
+        return sun4v_build_common(sysino, &sun4v_irq);
+}
+
+unsigned int sun4v_build_virq(u32 devhandle, unsigned int devino)
+{
+        unsigned long sysino, hv_err;
+
+        BUG_ON(devhandle & ~IMAP_IGN);
+        BUG_ON(devino & ~IMAP_INO);
+
+        sysino = devhandle | devino;
+
+        hv_err = sun4v_vintr_set_cookie(devhandle, devino, sysino);
+        if (hv_err) {
+                prom_printf("IRQ: Fatal, cannot set cookie for [%x:%x] "
+                            "err=%lu\n", devhandle, devino, hv_err);
+                prom_halt();
+        }
+
+        return sun4v_build_common(sysino, &sun4v_virq);
+}
+
 #ifdef CONFIG_PCI_MSI
 unsigned int sun4v_build_msi(u32 devhandle, unsigned int *virt_irq_p,
                              unsigned int msi_start, unsigned int msi_end)
@@ -694,9 +805,20 @@ 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 +833,58 @@ 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->dev_mondo_pa, HV_CPU_QUEUE_DEVICE_MONDO);
-        register_one_mondo(tb->resum_mondo_pa, HV_CPU_QUEUE_RES_ERROR);
-        register_one_mondo(tb->nonresum_mondo_pa, HV_CPU_QUEUE_NONRES_ERROR);
+        register_one_mondo(tb->cpu_mondo_pa, HV_CPU_QUEUE_CPU_MONDO,
+                           tb->cpu_mondo_qmask);
+        register_one_mondo(tb->dev_mondo_pa, HV_CPU_QUEUE_DEVICE_MONDO,
+                           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)
-                page = alloc_bootmem_low_pages(PAGE_SIZE);
-        else
-                page = (void *) get_zeroed_page(GFP_ATOMIC);
+        if (use_bootmem) {
+                p = __alloc_bootmem_low(size, size, 0);
+        } else {
+                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)
-                page = alloc_bootmem_low_pages(PAGE_SIZE);
-        else
-                page = (void *) get_zeroed_page(GFP_ATOMIC);
+        if (use_bootmem) {
+                p = __alloc_bootmem_low(size, size, 0);
+        } else {
+                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 +915,12 @@ 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->dev_mondo_pa, use_bootmem);
-                alloc_one_mondo(&tb->resum_mondo_pa, use_bootmem);
-                alloc_one_kbuf(&tb->resum_kernel_buf_pa, use_bootmem);
-                alloc_one_mondo(&tb->nonresum_mondo_pa, use_bootmem);
-                alloc_one_kbuf(&tb->nonresum_kernel_buf_pa, use_bootmem);
+                alloc_one_mondo(&tb->cpu_mondo_pa, tb->cpu_mondo_qmask, use_bootmem);
+                alloc_one_mondo(&tb->dev_mondo_pa, tb->dev_mondo_qmask, use_bootmem);
+                alloc_one_mondo(&tb->resum_mondo_pa, tb->resum_qmask, use_bootmem);
+                alloc_one_kbuf(&tb->resum_kernel_buf_pa, tb->resum_qmask, use_bootmem);
+                alloc_one_mondo(&tb->nonresum_mondo_pa, tb->nonresum_qmask, 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..f0e16045fb16
--- /dev/null
+++ b/arch/sparc64/kernel/mdesc.c
@@ -0,0 +1,672 @@
+/* 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 mark_proc_ids(struct mdesc_node *mp, int proc_id)
+{
+        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"))
+                        continue;
+
+                id = md_get_property(t, "id", NULL);
+                if (*id < NR_CPUS)
+                        cpu_data(*id).proc_id = proc_id;
+        }
+}
+
+static void __init __set_proc_ids(const char *exec_unit_name)
+{
+        struct mdesc_node *mp;
+        int idx;
+
+        idx = 0;
+        md_for_each_node_by_name(mp, exec_unit_name) {
+                const char *type;
+                int len;
+
+                type = md_get_property(mp, "type", &len);
+                if (!find_in_proplist(type, "int", len) &&
+                    !find_in_proplist(type, "integer", len))
+                        continue;
+
+                mark_proc_ids(mp, idx);
+
+                idx++;
+        }
+}
+
+static void __init set_proc_ids(void)
+{
+        __set_proc_ids("exec_unit");
+        __set_proc_ids("exec-unit");
+}
+
+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;
+                c->proc_id = -1;
+        }
+
+#ifdef CONFIG_SMP
+        sparc64_multi_core = 1;
+#endif
+
+        set_core_ids();
+        set_proc_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/of_device.c b/arch/sparc64/kernel/of_device.c
index 16cc46a71872..6676b93219dc 100644
--- a/arch/sparc64/kernel/of_device.c
+++ b/arch/sparc64/kernel/of_device.c
@@ -343,6 +343,15 @@ static int of_bus_simba_match(struct device_node *np)
 
         if (model && !strcmp(model, "SUNW,simba"))
                 return 1;
+
+        /* Treat PCI busses lacking ranges property just like
+         * simba.
+         */
+        if (!strcmp(np->type, "pci") || !strcmp(np->type, "pciex")) {
+                if (!of_find_property(np, "ranges", NULL))
+                        return 1;
+        }
+
         return 0;
 }
 
@@ -549,8 +558,6 @@ static int __init build_one_resource(struct device_node *parent,
 
 static int __init use_1to1_mapping(struct device_node *pp)
 {
-        const char *model;
-
         /* If this is on the PMU bus, don't try to translate it even
          * if a ranges property exists.
          */
@@ -567,9 +574,11 @@ static int __init use_1to1_mapping(struct device_node *pp)
         if (!strcmp(pp->name, "dma"))
                 return 0;
 
-        /* Similarly for Simba PCI bridges.  */
-        model = of_get_property(pp, "model", NULL);
-        if (model && !strcmp(model, "SUNW,simba"))
+        /* Similarly for all PCI bridges, if we get this far
+         * it lacks a ranges property, and this will include
+         * cases like Simba.
+         */
+        if (!strcmp(pp->type, "pci") || !strcmp(pp->type, "pciex"))
                 return 0;
 
         return 1;
diff --git a/arch/sparc64/kernel/pci.c b/arch/sparc64/kernel/pci.c
index d4c077dc5e85..81f4a5ea05f7 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",
-                       op_res->start, op_res->end, i);
+                if (ofpci_verbose)
+                        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",
-               devfn, type, host_controller);
+        if (ofpci_verbose)
+                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",
-               dev->class, pci_name(dev));
+        if (ofpci_verbose)
+                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);
 
@@ -502,6 +522,89 @@ static void pci_resource_adjust(struct resource *res,
         res->end += root->start;
 }
 
+/* For PCI bus devices which lack a 'ranges' property we interrogate
+ * the config space values to set the resources, just like the generic
+ * Linux PCI probing code does.
+ */
+static void __devinit pci_cfg_fake_ranges(struct pci_dev *dev,
+                                          struct pci_bus *bus,
+                                          struct pci_pbm_info *pbm)
+{
+        struct resource *res;
+        u8 io_base_lo, io_limit_lo;
+        u16 mem_base_lo, mem_limit_lo;
+        unsigned long base, limit;
+
+        pci_read_config_byte(dev, PCI_IO_BASE, &io_base_lo);
+        pci_read_config_byte(dev, PCI_IO_LIMIT, &io_limit_lo);
+        base = (io_base_lo & PCI_IO_RANGE_MASK) << 8;
+        limit = (io_limit_lo & PCI_IO_RANGE_MASK) << 8;
+
+        if ((io_base_lo & PCI_IO_RANGE_TYPE_MASK) == PCI_IO_RANGE_TYPE_32) {
+                u16 io_base_hi, io_limit_hi;
+
+                pci_read_config_word(dev, PCI_IO_BASE_UPPER16, &io_base_hi);
+                pci_read_config_word(dev, PCI_IO_LIMIT_UPPER16, &io_limit_hi);
+                base |= (io_base_hi << 16);
+                limit |= (io_limit_hi << 16);
+        }
+
+        res = bus->resource[0];
+        if (base <= limit) {
+                res->flags = (io_base_lo & PCI_IO_RANGE_TYPE_MASK) | IORESOURCE_IO;
+                if (!res->start)
+                        res->start = base;
+                if (!res->end)
+                        res->end = limit + 0xfff;
+                pci_resource_adjust(res, &pbm->io_space);
+        }
+
+        pci_read_config_word(dev, PCI_MEMORY_BASE, &mem_base_lo);
+        pci_read_config_word(dev, PCI_MEMORY_LIMIT, &mem_limit_lo);
+        base = (mem_base_lo & PCI_MEMORY_RANGE_MASK) << 16;
+        limit = (mem_limit_lo & PCI_MEMORY_RANGE_MASK) << 16;
+
+        res = bus->resource[1];
+        if (base <= limit) {
+                res->flags = ((mem_base_lo & PCI_MEMORY_RANGE_TYPE_MASK) |
+                              IORESOURCE_MEM);
+                res->start = base;
+                res->end = limit + 0xfffff;
+                pci_resource_adjust(res, &pbm->mem_space);
+        }
+
+        pci_read_config_word(dev, PCI_PREF_MEMORY_BASE, &mem_base_lo);
+        pci_read_config_word(dev, PCI_PREF_MEMORY_LIMIT, &mem_limit_lo);
+        base = (mem_base_lo & PCI_PREF_RANGE_MASK) << 16;
+        limit = (mem_limit_lo & PCI_PREF_RANGE_MASK) << 16;
+
+        if ((mem_base_lo & PCI_PREF_RANGE_TYPE_MASK) == PCI_PREF_RANGE_TYPE_64) {
+                u32 mem_base_hi, mem_limit_hi;
+
+                pci_read_config_dword(dev, PCI_PREF_BASE_UPPER32, &mem_base_hi);
+                pci_read_config_dword(dev, PCI_PREF_LIMIT_UPPER32, &mem_limit_hi);
+
+                /*
+                 * Some bridges set the base > limit by default, and some
+                 * (broken) BIOSes do not initialize them.  If we find
+                 * this, just assume they are not being used.
+                 */
+                if (mem_base_hi <= mem_limit_hi) {
+                        base |= ((long) mem_base_hi) << 32;
+                        limit |= ((long) mem_limit_hi) << 32;
+                }
+        }
+
+        res = bus->resource[2];
+        if (base <= limit) {
+                res->flags = ((mem_base_lo & PCI_MEMORY_RANGE_TYPE_MASK) |
+                              IORESOURCE_MEM | IORESOURCE_PREFETCH);
+                res->start = base;
+                res->end = limit + 0xfffff;
+                pci_resource_adjust(res, &pbm->mem_space);
+        }
+}
+
 /* Cook up fake bus resources for SUNW,simba PCI bridges which lack
  * a proper 'ranges' property.
  */
@@ -547,7 +650,8 @@ 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);
@@ -560,13 +664,8 @@ static void __devinit of_scan_pci_bridge(struct pci_pbm_info *pbm,
         simba = 0;
         if (ranges == NULL) {
                 const char *model = of_get_property(node, "model", NULL);
-                if (model && !strcmp(model, "SUNW,simba")) {
+                if (model && !strcmp(model, "SUNW,simba"))
                         simba = 1;
-                } else {
-                        printk(KERN_DEBUG "Can't get ranges for PCI-PCI bridge %s\n",
-                               node->full_name);
-                        return;
-                }
         }
 
         bus = pci_add_new_bus(dev->bus, dev, busrange[0]);
@@ -590,7 +689,10 @@ static void __devinit of_scan_pci_bridge(struct pci_pbm_info *pbm,
         }
         if (simba) {
                 apb_fake_ranges(dev, bus, pbm);
-                goto simba_cont;
+                goto after_ranges;
+        } else if (ranges == NULL) {
+                pci_cfg_fake_ranges(dev, bus, pbm);
+                goto after_ranges;
         }
         i = 1;
         for (; len >= 32; len -= 32, ranges += 8) {
@@ -629,10 +731,11 @@ static void __devinit of_scan_pci_bridge(struct pci_pbm_info *pbm,
                  */
                 pci_resource_adjust(res, root);
         }
-simba_cont:
+after_ranges:
         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 +749,14 @@ 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",
-               node->full_name, bus->number);
+        if (ofpci_verbose)
+                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 +766,9 @@ 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_common.c b/arch/sparc64/kernel/pci_common.c
index f974fefc3ebc..4249214608af 100644
--- a/arch/sparc64/kernel/pci_common.c
+++ b/arch/sparc64/kernel/pci_common.c
@@ -291,8 +291,9 @@ void pci_determine_mem_io_space(struct pci_pbm_info *pbm)
 
         for (i = 0; i < num_pbm_ranges; i++) {
                 const struct linux_prom_pci_ranges *pr = &pbm_ranges[i];
-                unsigned long a;
+                unsigned long a, size;
                 u32 parent_phys_hi, parent_phys_lo;
+                u32 size_hi, size_lo;
                 int type;
 
                 parent_phys_hi = pr->parent_phys_hi;
@@ -300,9 +301,14 @@ void pci_determine_mem_io_space(struct pci_pbm_info *pbm)
                 if (tlb_type == hypervisor)
                         parent_phys_hi &= 0x0fffffff;
 
+                size_hi = pr->size_hi;
+                size_lo = pr->size_lo;
+
                 type = (pr->child_phys_hi >> 24) & 0x3;
                 a = (((unsigned long)parent_phys_hi << 32UL) |
                      ((unsigned long)parent_phys_lo  <<  0UL));
+                size = (((unsigned long)size_hi << 32UL) |
+                        ((unsigned long)size_lo  <<  0UL));
 
                 switch (type) {
                 case 0:
@@ -313,7 +319,7 @@ void pci_determine_mem_io_space(struct pci_pbm_info *pbm)
                 case 1:
                         /* 16-bit IO space, 16MB */
                         pbm->io_space.start = a;
-                        pbm->io_space.end = a + ((16UL*1024UL*1024UL) - 1UL);
+                        pbm->io_space.end = a + size - 1UL;
                         pbm->io_space.flags = IORESOURCE_IO;
                         saw_io = 1;
                         break;
@@ -321,7 +327,7 @@ void pci_determine_mem_io_space(struct pci_pbm_info *pbm)
                 case 2:
                         /* 32-bit MEM space, 2GB */
                         pbm->mem_space.start = a;
-                        pbm->mem_space.end = a + (0x80000000UL - 1UL);
+                        pbm->mem_space.end = a + size - 1UL;
                         pbm->mem_space.flags = IORESOURCE_MEM;
                         saw_mem = 1;
                         break;
diff --git a/arch/sparc64/kernel/pci_sabre.c b/arch/sparc64/kernel/pci_sabre.c
index e2377796de89..22e1be5c7489 100644
--- a/arch/sparc64/kernel/pci_sabre.c
+++ b/arch/sparc64/kernel/pci_sabre.c
@@ -636,13 +636,18 @@ static void apb_init(struct pci_bus *sabre_bus)
 static void sabre_scan_bus(struct pci_pbm_info *pbm)
 {
         static int once;
-        struct pci_bus *pbus;
 
         /* The APB bridge speaks to the Sabre host PCI bridge
          * at 66Mhz, but the front side of APB runs at 33Mhz
          * for both segments.
+         *
+         * Hummingbird systems do not use APB, so they run
+         * at 66MHZ.
          */
-        pbm->is_66mhz_capable = 0;
+        if (hummingbird_p)
+                pbm->is_66mhz_capable = 1;
+        else
+                pbm->is_66mhz_capable = 0;
 
         /* This driver has not been verified to handle
          * multiple SABREs yet, so trap this.
@@ -656,13 +661,13 @@ static void sabre_scan_bus(struct pci_pbm_info *pbm)
         }
         once++;
 
-        pbus = pci_scan_one_pbm(pbm);
-        if (!pbus)
+        pbm->pci_bus = pci_scan_one_pbm(pbm);
+        if (!pbm->pci_bus)
                 return;
 
-        sabre_root_bus = pbus;
+        sabre_root_bus = pbm->pci_bus;
 
-        apb_init(pbus);
+        apb_init(pbm->pci_bus);
 
         sabre_register_error_handlers(pbm);
 }
@@ -762,9 +767,10 @@ 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);
-                        if (!strcmp(dp->name, "SUNW,UltraSPARC-IIe"))
-                                hummingbird_p = 1;
+                        for_each_node_by_type(dp, "cpu") {
+                                if (!strcmp(dp->name, "SUNW,UltraSPARC-IIe"))
+                                        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];
-        switch (vdma[1]) {
-                case 0x20000000:
-                        dma_mask |= 0x1fffffff;
-                        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();
+        if ((vdma[0] | vdma[1]) & ~IO_PAGE_MASK) {
+                prom_printf("PCI-SUN4V: strange virtual-dma[%08x:%08x].\n",
+                            vdma[0], vdma[1]);
+                prom_halt();
         };
 
-        tsbsize *= (8 * 1024);
-
-        num_tsb_entries = tsbsize / sizeof(iopte_t);
+        dma_mask = (roundup_pow_of_two(vdma[1]) - 1UL);
+        num_tsb_entries = vdma[1] / IO_PAGE_SIZE;
+        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..61036b346664 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;
 
@@ -932,29 +933,29 @@ static void __init fire_irq_trans_init(struct device_node *dp)
  * This should conform to both Sunfire/Wildfire server and Fusion
  * desktop designs.
  */
-#define SYSIO_IMAP_SLOT0        0x2c04UL
-#define SYSIO_IMAP_SLOT1        0x2c0cUL
-#define SYSIO_IMAP_SLOT2        0x2c14UL
-#define SYSIO_IMAP_SLOT3        0x2c1cUL
-#define SYSIO_IMAP_SCSI         0x3004UL
-#define SYSIO_IMAP_ETH          0x300cUL
-#define SYSIO_IMAP_BPP          0x3014UL
-#define SYSIO_IMAP_AUDIO        0x301cUL
-#define SYSIO_IMAP_PFAIL        0x3024UL
-#define SYSIO_IMAP_KMS          0x302cUL
-#define SYSIO_IMAP_FLPY         0x3034UL
-#define SYSIO_IMAP_SHW          0x303cUL
-#define SYSIO_IMAP_KBD          0x3044UL
-#define SYSIO_IMAP_MS           0x304cUL
-#define SYSIO_IMAP_SER          0x3054UL
-#define SYSIO_IMAP_TIM0         0x3064UL
-#define SYSIO_IMAP_TIM1         0x306cUL
-#define SYSIO_IMAP_UE           0x3074UL
-#define SYSIO_IMAP_CE           0x307cUL
-#define SYSIO_IMAP_SBERR        0x3084UL
-#define SYSIO_IMAP_PMGMT        0x308cUL
-#define SYSIO_IMAP_GFX          0x3094UL
-#define SYSIO_IMAP_EUPA         0x309cUL
+#define SYSIO_IMAP_SLOT0        0x2c00UL
+#define SYSIO_IMAP_SLOT1        0x2c08UL
+#define SYSIO_IMAP_SLOT2        0x2c10UL
+#define SYSIO_IMAP_SLOT3        0x2c18UL
+#define SYSIO_IMAP_SCSI         0x3000UL
+#define SYSIO_IMAP_ETH          0x3008UL
+#define SYSIO_IMAP_BPP          0x3010UL
+#define SYSIO_IMAP_AUDIO        0x3018UL
+#define SYSIO_IMAP_PFAIL        0x3020UL
+#define SYSIO_IMAP_KMS          0x3028UL
+#define SYSIO_IMAP_FLPY         0x3030UL
+#define SYSIO_IMAP_SHW          0x3038UL
+#define SYSIO_IMAP_KBD          0x3040UL
+#define SYSIO_IMAP_MS           0x3048UL
+#define SYSIO_IMAP_SER          0x3050UL
+#define SYSIO_IMAP_TIM0         0x3060UL
+#define SYSIO_IMAP_TIM1         0x3068UL
+#define SYSIO_IMAP_UE           0x3070UL
+#define SYSIO_IMAP_CE           0x3078UL
+#define SYSIO_IMAP_SBERR        0x3080UL
+#define SYSIO_IMAP_PMGMT        0x3088UL
+#define SYSIO_IMAP_GFX          0x3090UL
+#define SYSIO_IMAP_EUPA         0x3098UL
 
 #define bogon     ((unsigned long) -1)
 static unsigned long sysio_irq_offsets[] = {
@@ -1005,10 +1006,10 @@ static unsigned long sysio_irq_offsets[] = {
  * Interrupt Clear register pointer, SYSIO specific version.
  */
 #define SYSIO_ICLR_UNUSED0      0x3400UL
-#define SYSIO_ICLR_SLOT0        0x340cUL
-#define SYSIO_ICLR_SLOT1        0x344cUL
-#define SYSIO_ICLR_SLOT2        0x348cUL
-#define SYSIO_ICLR_SLOT3        0x34ccUL
+#define SYSIO_ICLR_SLOT0        0x3408UL
+#define SYSIO_ICLR_SLOT1        0x3448UL
+#define SYSIO_ICLR_SLOT2        0x3488UL
+#define SYSIO_ICLR_SLOT3        0x34c8UL
 static unsigned long sysio_imap_to_iclr(unsigned long imap)
 {
         unsigned long diff = SYSIO_ICLR_UNUSED0 - SYSIO_IMAP_SLOT0;
@@ -1665,6 +1666,155 @@ 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;
+                        cpu_data(cpuid).proc_id = portid;
+#ifdef CONFIG_SMP
+                        sparc64_multi_core = 1;
+#endif
+                } 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;
+                        cpu_data(cpuid).proc_id = -1;
+                }
+
+#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 +1829,7 @@ 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/sbus.c b/arch/sparc64/kernel/sbus.c
index 91f6e2a74ad5..a1fd9bcc0b87 100644
--- a/arch/sparc64/kernel/sbus.c
+++ b/arch/sparc64/kernel/sbus.c
@@ -629,29 +629,29 @@ void sbus_set_sbus64(struct sbus_dev *sdev, int bursts)
  * This should conform to both Sunfire/Wildfire server and Fusion
  * desktop designs.
  */
-#define SYSIO_IMAP_SLOT0        0x2c04UL
-#define SYSIO_IMAP_SLOT1        0x2c0cUL
-#define SYSIO_IMAP_SLOT2        0x2c14UL
-#define SYSIO_IMAP_SLOT3        0x2c1cUL
-#define SYSIO_IMAP_SCSI         0x3004UL
-#define SYSIO_IMAP_ETH          0x300cUL
-#define SYSIO_IMAP_BPP          0x3014UL
-#define SYSIO_IMAP_AUDIO        0x301cUL
-#define SYSIO_IMAP_PFAIL        0x3024UL
-#define SYSIO_IMAP_KMS          0x302cUL
-#define SYSIO_IMAP_FLPY         0x3034UL
-#define SYSIO_IMAP_SHW          0x303cUL
-#define SYSIO_IMAP_KBD          0x3044UL
-#define SYSIO_IMAP_MS           0x304cUL
-#define SYSIO_IMAP_SER          0x3054UL
-#define SYSIO_IMAP_TIM0         0x3064UL
-#define SYSIO_IMAP_TIM1         0x306cUL
-#define SYSIO_IMAP_UE           0x3074UL
-#define SYSIO_IMAP_CE           0x307cUL
-#define SYSIO_IMAP_SBERR        0x3084UL
-#define SYSIO_IMAP_PMGMT        0x308cUL
-#define SYSIO_IMAP_GFX          0x3094UL
-#define SYSIO_IMAP_EUPA         0x309cUL
+#define SYSIO_IMAP_SLOT0        0x2c00UL
+#define SYSIO_IMAP_SLOT1        0x2c08UL
+#define SYSIO_IMAP_SLOT2        0x2c10UL
+#define SYSIO_IMAP_SLOT3        0x2c18UL
+#define SYSIO_IMAP_SCSI         0x3000UL
+#define SYSIO_IMAP_ETH          0x3008UL
+#define SYSIO_IMAP_BPP          0x3010UL
+#define SYSIO_IMAP_AUDIO        0x3018UL
+#define SYSIO_IMAP_PFAIL        0x3020UL
+#define SYSIO_IMAP_KMS          0x3028UL
+#define SYSIO_IMAP_FLPY         0x3030UL
+#define SYSIO_IMAP_SHW          0x3038UL
+#define SYSIO_IMAP_KBD          0x3040UL
+#define SYSIO_IMAP_MS           0x3048UL
+#define SYSIO_IMAP_SER          0x3050UL
+#define SYSIO_IMAP_TIM0         0x3060UL
+#define SYSIO_IMAP_TIM1         0x3068UL
+#define SYSIO_IMAP_UE           0x3070UL
+#define SYSIO_IMAP_CE           0x3078UL
+#define SYSIO_IMAP_SBERR        0x3080UL
+#define SYSIO_IMAP_PMGMT        0x3088UL
+#define SYSIO_IMAP_GFX          0x3090UL
+#define SYSIO_IMAP_EUPA         0x3098UL
 
 #define bogon     ((unsigned long) -1)
 static unsigned long sysio_irq_offsets[] = {
@@ -700,10 +700,10 @@ static unsigned long sysio_irq_offsets[] = {
  * Interrupt Clear register pointer, SYSIO specific version.
  */
 #define SYSIO_ICLR_UNUSED0      0x3400UL
-#define SYSIO_ICLR_SLOT0        0x340cUL
-#define SYSIO_ICLR_SLOT1        0x344cUL
-#define SYSIO_ICLR_SLOT2        0x348cUL
-#define SYSIO_ICLR_SLOT3        0x34ccUL
+#define SYSIO_ICLR_SLOT0        0x3408UL
+#define SYSIO_ICLR_SLOT1        0x3448UL
+#define SYSIO_ICLR_SLOT2        0x3488UL
+#define SYSIO_ICLR_SLOT3        0x34c8UL
 static unsigned long sysio_imap_to_iclr(unsigned long imap)
 {
         unsigned long diff = SYSIO_ICLR_UNUSED0 - SYSIO_IMAP_SLOT0;
diff --git a/arch/sparc64/kernel/setup.c b/arch/sparc64/kernel/setup.c
index dea9c3c9ec5f..7490cc670a53 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)
 {
@@ -509,30 +513,3 @@ void sun_do_break(void)
 
 int serial_console = -1;
 int stop_a_enabled = 1;
-
-static int __init topology_init(void)
-{
-        int i, err;
-
-        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) {
-                        register_cpu(p, i);
-                        err = 0;
-                }
-        }
-
-        return err;
-}
-
-subsys_initcall(topology_init);
diff --git a/arch/sparc64/kernel/smp.c b/arch/sparc64/kernel/smp.c
index 24fdf1d0adc5..4dcd7d0b60f2 100644
--- a/arch/sparc64/kernel/smp.c
+++ b/arch/sparc64/kernel/smp.c
@@ -40,9 +40,12 @@
 #include <asm/tlb.h>
 #include <asm/sections.h>
 #include <asm/prom.h>
+#include <asm/mdesc.h>
 
 extern void calibrate_delay(void);
 
+int sparc64_multi_core __read_mostly;
+
 /* Please don't make this stuff initdata!!!  --DaveM */
 unsigned char boot_cpu_id;
 
@@ -50,6 +53,8 @@ cpumask_t cpu_online_map __read_mostly = CPU_MASK_NONE;
 cpumask_t phys_cpu_present_map __read_mostly = CPU_MASK_NONE;
 cpumask_t cpu_sibling_map[NR_CPUS] __read_mostly =
         { [0 ... NR_CPUS-1] = CPU_MASK_NONE };
+cpumask_t cpu_core_map[NR_CPUS] __read_mostly =
+        { [0 ... NR_CPUS-1] = CPU_MASK_NONE };
 static cpumask_t smp_commenced_mask;
 static cpumask_t cpu_callout_map;
 
@@ -75,53 +80,6 @@ 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 +103,7 @@ 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 +298,8 @@ 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 +404,7 @@ 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 +420,7 @@ 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 +449,10 @@ retry:
                                 : /* no outputs */
                                 : "r" (target), "i" (ASI_INTR_W));
                         nack_busy_id++;
+                        if (nack_busy_id == 32) {
+                                need_more = 1;
+                                break;
+                        }
                 }
         }
 
@@ -507,6 +469,16 @@ 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 +516,8 @@ 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 +1165,14 @@ int setup_profiling_timer(unsigned int multiplier)
 
 static void __init smp_tune_scheduling(void)
 {
-        struct device_node *dp;
-        int instance;
-        unsigned int def, smallest = ~0U;
-
-        def = ((tlb_type == hypervisor) ?
-               (3 * 1024 * 1024) :
-               (4 * 1024 * 1024));
+        unsigned int smallest = ~0U;
+        int i;
 
-        instance = 0;
-        while (!cpu_find_by_instance(instance, &dp, NULL)) {
-                unsigned int val;
+        for (i = 0; i < NR_CPUS; i++) {
+                unsigned int val = cpu_data(i).ecache_size;
 
-                val = of_getintprop_default(dp, "ecache-size", def);
-                if (val < smallest)
+                if (val && val < smallest)
                         smallest = val;
-
-                instance++;
         }
 
         /* Any value less than 256K is nonsense.  */
@@ -1230,60 +1195,59 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
         int i;
 
         if (num_possible_cpus() > max_cpus) {
-                int instance, mid;
-
-                instance = 0;
-                while (!cpu_find_by_instance(instance, NULL, &mid)) {
-                        if (mid != boot_cpu_id) {
-                                cpu_clear(mid, phys_cpu_present_map);
-                                cpu_clear(mid, cpu_present_map);
+                for_each_possible_cpu(i) {
+                        if (i != boot_cpu_id) {
+                                cpu_clear(i, phys_cpu_present_map);
+                                cpu_clear(i, cpu_present_map);
                                 if (num_possible_cpus() <= max_cpus)
                                         break;
                         }
-                        instance++;
                 }
         }
 
+        cpu_data(boot_cpu_id).udelay_val = loops_per_jiffy;
+        smp_tune_scheduling();
+}
+
+void __devinit smp_prepare_boot_cpu(void)
+{
+}
+
+void __devinit smp_fill_in_sib_core_maps(void)
+{
+        unsigned int i;
+
         for_each_possible_cpu(i) {
-                if (tlb_type == hypervisor) {
-                        int j;
+                unsigned 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]);
+                if (cpu_data(i).core_id == 0) {
+                        cpu_set(i, cpu_core_map[i]);
+                        continue;
+                }
+
+                for_each_possible_cpu(j) {
+                        if (cpu_data(i).core_id ==
+                            cpu_data(j).core_id)
+                                cpu_set(j, cpu_core_map[i]);
                 }
         }
 
-        smp_store_cpu_info(boot_cpu_id);
-        smp_tune_scheduling();
-}
+        for_each_possible_cpu(i) {
+                unsigned int j;
 
-/* Set this up early so that things like the scheduler can init
- * 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;
+                if (cpu_data(i).proc_id == -1) {
+                        cpu_set(i, cpu_sibling_map[i]);
+                        continue;
+                }
 
-        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);
+                for_each_possible_cpu(j) {
+                        if (cpu_data(i).proc_id ==
+                            cpu_data(j).proc_id)
+                                cpu_set(j, cpu_sibling_map[i]);
                 }
-                instance++;
         }
 }
 
-void __devinit smp_prepare_boot_cpu(void)
-{
-}
-
 int __cpuinit __cpu_up(unsigned int cpu)
 {
         int ret = smp_boot_one_cpu(cpu);
@@ -1337,7 +1301,7 @@ 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/sparc64_ksyms.c b/arch/sparc64/kernel/sparc64_ksyms.c
index d00f51a5683f..6fa761612899 100644
--- a/arch/sparc64/kernel/sparc64_ksyms.c
+++ b/arch/sparc64/kernel/sparc64_ksyms.c
@@ -24,6 +24,7 @@
 #include <linux/syscalls.h>
 #include <linux/percpu.h>
 #include <linux/init.h>
+#include <linux/rwsem.h>
 #include <net/compat.h>
 
 #include <asm/oplib.h>
@@ -58,7 +59,6 @@
 #include <asm/ns87303.h>
 #include <asm/timer.h>
 #include <asm/cpudata.h>
-#include <asm/rwsem.h>
 
 struct poll {
         int fd;
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.  */
-        ldxa    [%g0] ASI_SCRATCHPAD, %g3
-        sub     %g3, TRAP_PER_CPU_FAULT_INFO, %g3
+        /* Get &trap_block[smp_processor_id()] into %g4.  */
+        ldxa    [%g0] ASI_SCRATCHPAD, %g4
+        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
-        or      %g4, %lo(8192 - 1), %g4
+        lduw    [%g4 + TRAP_PER_CPU_CPU_MONDO_QMASK], %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.  */
-        ldxa    [%g0] ASI_SCRATCHPAD, %g3
-        sub     %g3, TRAP_PER_CPU_FAULT_INFO, %g3
+        /* Get &trap_block[smp_processor_id()] into %g4.  */
+        ldxa    [%g0] ASI_SCRATCHPAD, %g4
+        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
-        or      %g4, %lo(8192 - 1), %g4
+        lduw    [%g4 + TRAP_PER_CPU_DEV_MONDO_QMASK], %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/sysfs.c b/arch/sparc64/kernel/sysfs.c
new file mode 100644
index 000000000000..cdb1477af89f
--- /dev/null
+++ b/arch/sparc64/kernel/sysfs.c
@@ -0,0 +1,297 @@
+/* sysfs.c: Toplogy sysfs support code for sparc64.
+ *
+ * Copyright (C) 2007 David S. Miller <davem@davemloft.net>
+ */
+#include <linux/sysdev.h>
+#include <linux/cpu.h>
+#include <linux/smp.h>
+#include <linux/percpu.h>
+#include <linux/init.h>
+
+#include <asm/hypervisor.h>
+#include <asm/spitfire.h>
+
+static DEFINE_PER_CPU(struct hv_mmu_statistics, mmu_stats) __attribute__((aligned(64)));
+
+#define SHOW_MMUSTAT_ULONG(NAME) \
+static ssize_t show_##NAME(struct sys_device *dev, char *buf) \
+{ \
+        struct hv_mmu_statistics *p = &per_cpu(mmu_stats, dev->id); \
+        return sprintf(buf, "%lu\n", p->NAME); \
+} \
+static SYSDEV_ATTR(NAME, 0444, show_##NAME, NULL)
+
+SHOW_MMUSTAT_ULONG(immu_tsb_hits_ctx0_8k_tte);
+SHOW_MMUSTAT_ULONG(immu_tsb_ticks_ctx0_8k_tte);
+SHOW_MMUSTAT_ULONG(immu_tsb_hits_ctx0_64k_tte);
+SHOW_MMUSTAT_ULONG(immu_tsb_ticks_ctx0_64k_tte);
+SHOW_MMUSTAT_ULONG(immu_tsb_hits_ctx0_4mb_tte);
+SHOW_MMUSTAT_ULONG(immu_tsb_ticks_ctx0_4mb_tte);
+SHOW_MMUSTAT_ULONG(immu_tsb_hits_ctx0_256mb_tte);
+SHOW_MMUSTAT_ULONG(immu_tsb_ticks_ctx0_256mb_tte);
+SHOW_MMUSTAT_ULONG(immu_tsb_hits_ctxnon0_8k_tte);
+SHOW_MMUSTAT_ULONG(immu_tsb_ticks_ctxnon0_8k_tte);
+SHOW_MMUSTAT_ULONG(immu_tsb_hits_ctxnon0_64k_tte);
+SHOW_MMUSTAT_ULONG(immu_tsb_ticks_ctxnon0_64k_tte);
+SHOW_MMUSTAT_ULONG(immu_tsb_hits_ctxnon0_4mb_tte);
+SHOW_MMUSTAT_ULONG(immu_tsb_ticks_ctxnon0_4mb_tte);
+SHOW_MMUSTAT_ULONG(immu_tsb_hits_ctxnon0_256mb_tte);
+SHOW_MMUSTAT_ULONG(immu_tsb_ticks_ctxnon0_256mb_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_hits_ctx0_8k_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_ticks_ctx0_8k_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_hits_ctx0_64k_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_ticks_ctx0_64k_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_hits_ctx0_4mb_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_ticks_ctx0_4mb_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_hits_ctx0_256mb_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_ticks_ctx0_256mb_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_hits_ctxnon0_8k_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_ticks_ctxnon0_8k_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_hits_ctxnon0_64k_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_ticks_ctxnon0_64k_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_hits_ctxnon0_4mb_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_ticks_ctxnon0_4mb_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_hits_ctxnon0_256mb_tte);
+SHOW_MMUSTAT_ULONG(dmmu_tsb_ticks_ctxnon0_256mb_tte);
+
+static struct attribute *mmu_stat_attrs[] = {
+        &attr_immu_tsb_hits_ctx0_8k_tte.attr,
+        &attr_immu_tsb_ticks_ctx0_8k_tte.attr,
+        &attr_immu_tsb_hits_ctx0_64k_tte.attr,
+        &attr_immu_tsb_ticks_ctx0_64k_tte.attr,
+        &attr_immu_tsb_hits_ctx0_4mb_tte.attr,
+        &attr_immu_tsb_ticks_ctx0_4mb_tte.attr,
+        &attr_immu_tsb_hits_ctx0_256mb_tte.attr,
+        &attr_immu_tsb_ticks_ctx0_256mb_tte.attr,
+        &attr_immu_tsb_hits_ctxnon0_8k_tte.attr,
+        &attr_immu_tsb_ticks_ctxnon0_8k_tte.attr,
+        &attr_immu_tsb_hits_ctxnon0_64k_tte.attr,
+        &attr_immu_tsb_ticks_ctxnon0_64k_tte.attr,
+        &attr_immu_tsb_hits_ctxnon0_4mb_tte.attr,
+        &attr_immu_tsb_ticks_ctxnon0_4mb_tte.attr,
+        &attr_immu_tsb_hits_ctxnon0_256mb_tte.attr,
+        &attr_immu_tsb_ticks_ctxnon0_256mb_tte.attr,
+        &attr_dmmu_tsb_hits_ctx0_8k_tte.attr,
+        &attr_dmmu_tsb_ticks_ctx0_8k_tte.attr,
+        &attr_dmmu_tsb_hits_ctx0_64k_tte.attr,
+        &attr_dmmu_tsb_ticks_ctx0_64k_tte.attr,
+        &attr_dmmu_tsb_hits_ctx0_4mb_tte.attr,
+        &attr_dmmu_tsb_ticks_ctx0_4mb_tte.attr,
+        &attr_dmmu_tsb_hits_ctx0_256mb_tte.attr,
+        &attr_dmmu_tsb_ticks_ctx0_256mb_tte.attr,
+        &attr_dmmu_tsb_hits_ctxnon0_8k_tte.attr,
+        &attr_dmmu_tsb_ticks_ctxnon0_8k_tte.attr,
+        &attr_dmmu_tsb_hits_ctxnon0_64k_tte.attr,
+        &attr_dmmu_tsb_ticks_ctxnon0_64k_tte.attr,
+        &attr_dmmu_tsb_hits_ctxnon0_4mb_tte.attr,
+        &attr_dmmu_tsb_ticks_ctxnon0_4mb_tte.attr,
+        &attr_dmmu_tsb_hits_ctxnon0_256mb_tte.attr,
+        &attr_dmmu_tsb_ticks_ctxnon0_256mb_tte.attr,
+        NULL,
+};
+
+static struct attribute_group mmu_stat_group = {
+        .attrs = mmu_stat_attrs,
+        .name = "mmu_stats",
+};
+
+/* XXX convert to rusty's on_one_cpu */
+static unsigned long run_on_cpu(unsigned long cpu,
+                                unsigned long (*func)(unsigned long),
+                                unsigned long arg)
+{
+        cpumask_t old_affinity = current->cpus_allowed;
+        unsigned long ret;
+
+        /* should return -EINVAL to userspace */
+        if (set_cpus_allowed(current, cpumask_of_cpu(cpu)))
+                return 0;
+
+        ret = func(arg);
+
+        set_cpus_allowed(current, old_affinity);
+
+        return ret;
+}
+
+static unsigned long read_mmustat_enable(unsigned long junk)
+{
+        unsigned long ra = 0;
+
+        sun4v_mmustat_info(&ra);
+
+        return ra != 0;
+}
+
+static unsigned long write_mmustat_enable(unsigned long val)
+{
+        unsigned long ra, orig_ra;
+
+        if (val)
+                ra = __pa(&per_cpu(mmu_stats, smp_processor_id()));
+        else
+                ra = 0UL;
+
+        return sun4v_mmustat_conf(ra, &orig_ra);
+}
+
+static ssize_t show_mmustat_enable(struct sys_device *s, char *buf)
+{
+        unsigned long val = run_on_cpu(s->id, read_mmustat_enable, 0);
+        return sprintf(buf, "%lx\n", val);
+}
+
+static ssize_t store_mmustat_enable(struct sys_device *s, const char *buf, size_t count)
+{
+        unsigned long val, err;
+        int ret = sscanf(buf, "%ld", &val);
+
+        if (ret != 1)
+                return -EINVAL;
+
+        err = run_on_cpu(s->id, write_mmustat_enable, val);
+        if (err)
+                return -EIO;
+
+        return count;
+}
+
+static SYSDEV_ATTR(mmustat_enable, 0644, show_mmustat_enable, store_mmustat_enable);
+
+static int mmu_stats_supported;
+
+static int register_mmu_stats(struct sys_device *s)
+{
+        if (!mmu_stats_supported)
+                return 0;
+        sysdev_create_file(s, &attr_mmustat_enable);
+        return sysfs_create_group(&s->kobj, &mmu_stat_group);
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+static void unregister_mmu_stats(struct sys_device *s)
+{
+        if (!mmu_stats_supported)
+                return;
+        sysfs_remove_group(&s->kobj, &mmu_stat_group);
+        sysdev_remove_file(s, &attr_mmustat_enable);
+}
+#endif
+
+#define SHOW_CPUDATA_ULONG_NAME(NAME, MEMBER) \
+static ssize_t show_##NAME(struct sys_device *dev, char *buf) \
+{ \
+        cpuinfo_sparc *c = &cpu_data(dev->id); \
+        return sprintf(buf, "%lu\n", c->MEMBER); \
+}
+
+#define SHOW_CPUDATA_UINT_NAME(NAME, MEMBER) \
+static ssize_t show_##NAME(struct sys_device *dev, char *buf) \
+{ \
+        cpuinfo_sparc *c = &cpu_data(dev->id); \
+        return sprintf(buf, "%u\n", c->MEMBER); \
+}
+
+SHOW_CPUDATA_ULONG_NAME(clock_tick, clock_tick);
+SHOW_CPUDATA_ULONG_NAME(udelay_val, udelay_val);
+SHOW_CPUDATA_UINT_NAME(l1_dcache_size, dcache_size);
+SHOW_CPUDATA_UINT_NAME(l1_dcache_line_size, dcache_line_size);
+SHOW_CPUDATA_UINT_NAME(l1_icache_size, icache_size);
+SHOW_CPUDATA_UINT_NAME(l1_icache_line_size, icache_line_size);
+SHOW_CPUDATA_UINT_NAME(l2_cache_size, ecache_size);
+SHOW_CPUDATA_UINT_NAME(l2_cache_line_size, ecache_line_size);
+
+static struct sysdev_attribute cpu_core_attrs[] = {
+        _SYSDEV_ATTR(clock_tick,          0444, show_clock_tick, NULL),
+        _SYSDEV_ATTR(udelay_val,          0444, show_udelay_val, NULL),
+        _SYSDEV_ATTR(l1_dcache_size,      0444, show_l1_dcache_size, NULL),
+        _SYSDEV_ATTR(l1_dcache_line_size, 0444, show_l1_dcache_line_size, NULL),
+        _SYSDEV_ATTR(l1_icache_size,      0444, show_l1_icache_size, NULL),
+        _SYSDEV_ATTR(l1_icache_line_size, 0444, show_l1_icache_line_size, NULL),
+        _SYSDEV_ATTR(l2_cache_size,       0444, show_l2_cache_size, NULL),
+        _SYSDEV_ATTR(l2_cache_line_size,  0444, show_l2_cache_line_size, NULL),
+};
+
+static DEFINE_PER_CPU(struct cpu, cpu_devices);
+
+static void register_cpu_online(unsigned int cpu)
+{
+        struct cpu *c = &per_cpu(cpu_devices, cpu);
+        struct sys_device *s = &c->sysdev;
+        int i;
+
+        for (i = 0; i < ARRAY_SIZE(cpu_core_attrs); i++)
+                sysdev_create_file(s, &cpu_core_attrs[i]);
+
+        register_mmu_stats(s);
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+static void unregister_cpu_online(unsigned int cpu)
+{
+        struct cpu *c = &per_cpu(cpu_devices, cpu);
+        struct sys_device *s = &c->sysdev;
+        int i;
+
+        unregister_mmu_stats(s);
+        for (i = 0; i < ARRAY_SIZE(cpu_core_attrs); i++)
+                sysdev_remove_file(s, &cpu_core_attrs[i]);
+}
+#endif
+
+static int __cpuinit sysfs_cpu_notify(struct notifier_block *self,
+                                      unsigned long action, void *hcpu)
+{
+        unsigned int cpu = (unsigned int)(long)hcpu;
+
+        switch (action) {
+        case CPU_ONLINE:
+        case CPU_ONLINE_FROZEN:
+                register_cpu_online(cpu);
+                break;
+#ifdef CONFIG_HOTPLUG_CPU
+        case CPU_DEAD:
+        case CPU_DEAD_FROZEN:
+                unregister_cpu_online(cpu);
+                break;
+#endif
+        }
+        return NOTIFY_OK;
+}
+
+static struct notifier_block __cpuinitdata sysfs_cpu_nb = {
+        .notifier_call  = sysfs_cpu_notify,
+};
+
+static void __init check_mmu_stats(void)
+{
+        unsigned long dummy1, err;
+
+        if (tlb_type != hypervisor)
+                return;
+
+        err = sun4v_mmustat_info(&dummy1);
+        if (!err)
+                mmu_stats_supported = 1;
+}
+
+static int __init topology_init(void)
+{
+        int cpu;
+
+        check_mmu_stats();
+
+        register_cpu_notifier(&sysfs_cpu_nb);
+
+        for_each_possible_cpu(cpu) {
+                struct cpu *c = &per_cpu(cpu_devices, cpu);
+
+                register_cpu(c, cpu);
+                if (cpu_online(cpu))
+                        register_cpu_online(cpu);
+        }
+
+        return 0;
+}
+
+subsys_initcall(topology_init);
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");
-        register unsigned long arg0 asm("%o0");
-        register unsigned long arg1 asm("%o1");
+        unsigned long ret, time;
         int retries = 10000;
 
 retry:
-        func = HV_FAST_TOD_GET;
-        arg0 = 0;
-        arg1 = 0;
-        __asm__ __volatile__("ta        %6"
-                             : "=&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) {
+        ret = sun4v_tod_get(&time);
+        if (ret == HV_EOK)
+                return time;
+        if (ret == 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");
-        register unsigned long arg0 asm("%o0");
+        unsigned long ret;
         int retries = 10000;
 
 retry:
-        func = HV_FAST_TOD_SET;
-        arg0 = secs;
-        __asm__ __volatile__("ta        %4"
-                             : "=&r" (func), "=&r" (arg0)
-                             : "0" (func), "1" (arg0),
-                               "i" (HV_FAST_TRAP));
-        if (arg0 == HV_EOK)
+        ret = sun4v_tod_set(secs);
+        if (ret == 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);
-                        prop = of_find_property(dp, "clock-frequency", NULL);
+                        clock = local_cpu_data().clock_tick;
                 }
         } 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, instance, sz;
+        int i, 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;
-        while (!cpu_find_by_instance(instance, &dp, NULL)) {
+        for (i = 0; i < NR_CPUS; i++) {
                 unsigned long val;
 
-                val = of_getintprop_default(dp, "ecache-size",
-                                            (2 * 1024 * 1024));
+                val = cpu_data(i).ecache_size;
+                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>
-
-extern void device_scan(void);
+#include <asm/sstate.h>
+#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.  */
-struct tsb swapper_4m_tsb[KERNEL_TSB4M_NENTRIES];
+/* A special kernel TSB for 4MB and 256MB linear mappings.
+ * 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_mask      (256UL - 1UL)
-
-#if NR_CPUS > 256
-#error D-cache dirty tracking and thread_info->cpu need fixing for > 256 cpus
-#endif
+#define PG_dcache_cpu_shift     32UL
+#define PG_dcache_cpu_mask      \
+        ((1UL<<ilog2(roundup_pow_of_two(NR_CPUS)))-1UL)
 
 #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");
-        register unsigned long arg0 asm("%o0");
-        register unsigned long arg1 asm("%o1");
-        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) {
+        unsigned long ret = sun4v_mmu_map_perm_addr(vaddr, 0, pte, mmu);
+
+        if (ret != 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");
-        register unsigned long arg0 asm("%o0");
-        register unsigned long arg1 asm("%o1");
-        unsigned long pa;
+        unsigned long pa, ret;
 
         pa = kern_base + ((unsigned long)&ktsb_descr[0] - KERNBASE);
 
-        func = HV_FAST_MMU_TSB_CTX0;
-        arg0 = NUM_KTSB_DESCR;
-        arg1 = pa;
-        __asm__ __volatile__("ta        %6"
-                             : "=&r" (func), "=&r" (arg0), "=&r" (arg1)
-                             : "0" (func), "1" (arg0), "2" (arg1),
-                               "i" (HV_FAST_TRAP));
+        ret = sun4v_mmu_tsb_ctx0(NUM_KTSB_DESCR, pa);
+        if (ret != 0) {
+                prom_printf("hypervisor_mmu_tsb_ctx0[%lx]: "
+                            "errors with %lx\n", pa, ret);
+                prom_halt();
+        }
 }
 
 /* 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/um/Kconfig b/arch/um/Kconfig
index c504312219b4..e6ff30266542 100644
--- a/arch/um/Kconfig
+++ b/arch/um/Kconfig
@@ -278,6 +278,7 @@ config HIGHMEM
 config KERNEL_STACK_ORDER
         int "Kernel stack size order"
         default 1 if 64BIT
+        range 1 10 if 64BIT
         default 0 if !64BIT
         help
         This option determines the size of UML kernel stacks.  They will
diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c
index ced99106f798..4bd40bb43ec2 100644
--- a/arch/um/drivers/line.c
+++ b/arch/um/drivers/line.c
@@ -3,6 +3,7 @@
  * Licensed under the GPL
  */
 
+#include "linux/kernel.h"
 #include "linux/sched.h"
 #include "linux/slab.h"
 #include "linux/list.h"
diff --git a/arch/um/drivers/stderr_console.c b/arch/um/drivers/stderr_console.c
index 911539293871..4739dd527b43 100644
--- a/arch/um/drivers/stderr_console.c
+++ b/arch/um/drivers/stderr_console.c
@@ -1,3 +1,4 @@
+#include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/console.h>
 
diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c
index 70509ddaac03..2e09f162c42f 100644
--- a/arch/um/drivers/ubd_kern.c
+++ b/arch/um/drivers/ubd_kern.c
@@ -20,6 +20,7 @@
 #define MAJOR_NR UBD_MAJOR
 #define UBD_SHIFT 4
 
+#include "linux/kernel.h"
 #include "linux/module.h"
 #include "linux/blkdev.h"
 #include "linux/hdreg.h"
diff --git a/arch/um/kernel/exitcode.c b/arch/um/kernel/exitcode.c
index 8b7f2cdedf94..c716b5a6db13 100644
--- a/arch/um/kernel/exitcode.c
+++ b/arch/um/kernel/exitcode.c
@@ -1,8 +1,9 @@
-/* 
+/*
  * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
  * Licensed under the GPL
  */
 
+#include "linux/kernel.h"
 #include "linux/init.h"
 #include "linux/ctype.h"
 #include "linux/proc_fs.h"
@@ -24,11 +25,14 @@ static int read_proc_exitcode(char *page, char **start, off_t off,
         val = uml_exitcode;
         len = sprintf(page, "%d\n", val);
         len -= off;
-        if(len <= off+count) *eof = 1;
+        if(len <= off+count)
+                *eof = 1;
         *start = page + off;
-        if(len > count) len = count;
-        if(len < 0) len = 0;
-        return(len);
+        if(len > count)
+                len = count;
+        if(len < 0)
+                len = 0;
+        return len;
 }
 
 static int write_proc_exitcode(struct file *file, const char __user *buffer,
@@ -38,12 +42,14 @@ static int write_proc_exitcode(struct file *file, const char __user *buffer,
         int tmp;
 
         if(copy_from_user(buf, buffer, count))
-                return(-EFAULT);
+                return -EFAULT;
+
         tmp = simple_strtol(buf, &end, 0);
         if((*end != '\0') && !isspace(*end))
-                return(-EINVAL);
+                return -EINVAL;
+
         uml_exitcode = tmp;
-        return(count);
+        return count;
 }
 
 static int make_proc_exitcode(void)
@@ -54,24 +60,13 @@ static int make_proc_exitcode(void)
         if(ent == NULL){
                 printk(KERN_WARNING "make_proc_exitcode : Failed to register "
                        "/proc/exitcode\n");
-                return(0);
+                return 0;
         }
 
         ent->read_proc = read_proc_exitcode;
         ent->write_proc = write_proc_exitcode;
-        
-        return(0);
+
+        return 0;
 }
 
 __initcall(make_proc_exitcode);
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
diff --git a/arch/um/os-Linux/start_up.c b/arch/um/os-Linux/start_up.c
index 79471f85eb89..3fc13fa8729d 100644
--- a/arch/um/os-Linux/start_up.c
+++ b/arch/um/os-Linux/start_up.c
@@ -144,9 +144,7 @@ static int stop_ptraced_child(int pid, void *stack, int exitcode,
                 int exit_with = WEXITSTATUS(status);
                 if (exit_with == 2)
                         non_fatal("check_ptrace : child exited with status 2. "
-                                  "Serious trouble happening! Try updating "
-                                  "your host skas patch!\nDisabling SYSEMU "
-                                  "support.");
+                                  "\nDisabling SYSEMU support.\n");
                 non_fatal("check_ptrace : child exited with exitcode %d, while "
                           "expecting %d; status 0x%x\n", exit_with,
                           exitcode, status);
@@ -209,6 +207,7 @@ __uml_setup("nosysemu", nosysemu_cmd_param,
 static void __init check_sysemu(void)
 {
         void *stack;
+        unsigned long regs[MAX_REG_NR];
         int pid, n, status, count=0;
 
         non_fatal("Checking syscall emulation patch for ptrace...");
@@ -225,11 +224,20 @@ static void __init check_sysemu(void)
                 fatal("check_sysemu : expected SIGTRAP, got status = %d",
                       status);
 
-        n = ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_RET_OFFSET,
-                   os_getpid());
-        if(n < 0)
-                fatal_perror("check_sysemu : failed to modify system call "
-                             "return");
+        if(ptrace(PTRACE_GETREGS, pid, 0, regs) < 0)
+                fatal_perror("check_sysemu : PTRACE_GETREGS failed");
+        if(PT_SYSCALL_NR(regs) != __NR_getpid){
+                non_fatal("check_sysemu got system call number %d, "
+                          "expected %d...", PT_SYSCALL_NR(regs), __NR_getpid);
+                goto fail;
+        }
+
+        n = ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_RET_OFFSET, os_getpid());
+        if(n < 0){
+                non_fatal("check_sysemu : failed to modify system call "
+                          "return");
+                goto fail;
+        }
 
         if (stop_ptraced_child(pid, stack, 0, 0) < 0)
                 goto fail_stopped;
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 145bb824b2a8..5ce94430c019 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -428,12 +428,15 @@ config NR_CPUS
           memory in the static kernel configuration.
 
 config HOTPLUG_CPU
-        bool "Support for hot-pluggable CPUs (EXPERIMENTAL)"
+        bool "Support for suspend on SMP and hot-pluggable CPUs (EXPERIMENTAL)"
         depends on SMP && HOTPLUG && EXPERIMENTAL
         help
                 Say Y here to experiment with turning CPUs off and on.  CPUs
                 can be controlled through /sys/devices/system/cpu/cpu#.
-                Say N if you want to disable CPU hotplug.
+                This is also required for suspend/hibernation on SMP systems.
+
+                Say N if you want to disable CPU hotplug and don't need to
+                suspend.
 
 config ARCH_ENABLE_MEMORY_HOTPLUG
         def_bool y
diff --git a/arch/x86_64/kernel/k8.c b/arch/x86_64/kernel/k8.c
index bc11b32e8b4d..7377ccb21335 100644
--- a/arch/x86_64/kernel/k8.c
+++ b/arch/x86_64/kernel/k8.c
@@ -39,10 +39,10 @@ int cache_k8_northbridges(void)
 {
         int i;
         struct pci_dev *dev;
+
         if (num_k8_northbridges)
                 return 0;
 
-        num_k8_northbridges = 0;
         dev = NULL;
         while ((dev = next_k8_northbridge(dev)) != NULL)
                 num_k8_northbridges++;
@@ -52,6 +52,11 @@ int cache_k8_northbridges(void)
         if (!k8_northbridges)
                 return -ENOMEM;
 
+        if (!num_k8_northbridges) {
+                k8_northbridges[0] = NULL;
+                return 0;
+        }
+
         flush_words = kmalloc(num_k8_northbridges * sizeof(u32), GFP_KERNEL);
         if (!flush_words) {
                 kfree(k8_northbridges);
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index cb29fb96948d..aac1c0be54c6 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -465,13 +465,14 @@ static unsigned int die_nest_count;
 
 unsigned __kprobes long oops_begin(void)
 {
-        int cpu = smp_processor_id();
+        int cpu;
         unsigned long flags;
 
         oops_enter();
 
         /* racy, but better than risking deadlock. */
         local_irq_save(flags);
+        cpu = smp_processor_id();
         if (!spin_trylock(&die_lock)) { 
                 if (cpu == die_owner) 
                         /* nested oops. should stop eventually */;
diff --git a/arch/x86_64/mm/fault.c b/arch/x86_64/mm/fault.c
index bfb62a13d7ee..635e58d443d7 100644
--- a/arch/x86_64/mm/fault.c
+++ b/arch/x86_64/mm/fault.c
@@ -476,6 +476,12 @@ bad_area:
 bad_area_nosemaphore:
         /* User mode accesses just cause a SIGSEGV */
         if (error_code & PF_USER) {
+
+                /*
+                 * It's possible to have interrupts off here.
+                 */
+                local_irq_enable();
+
                 if (is_prefetch(regs, address, error_code))
                         return;
 
diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index 1336da8bdee1..efb6e845114e 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -79,6 +79,8 @@ void show_mem(void)
                         if (unlikely(i % MAX_ORDER_NR_PAGES == 0)) {
                                 touch_nmi_watchdog();
                         }
+                        if (!pfn_valid(pgdat->node_start_pfn + i))
+                                continue;
                         page = pfn_to_page(pgdat->node_start_pfn + i);
                         total++;
                         if (PageReserved(page))
@@ -761,3 +763,9 @@ 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);
+}
diff --git a/arch/xtensa/kernel/asm-offsets.c b/arch/xtensa/kernel/asm-offsets.c
index 698079b3a336..d0323cd6a2ea 100644
--- a/arch/xtensa/kernel/asm-offsets.c
+++ b/arch/xtensa/kernel/asm-offsets.c
@@ -39,6 +39,7 @@ int main(void)
         DEFINE(PT_LEND, offsetof (struct pt_regs, lend));
         DEFINE(PT_LCOUNT, offsetof (struct pt_regs, lcount));
         DEFINE(PT_SAR, offsetof (struct pt_regs, sar));
+        DEFINE(PT_ICOUNTLEVEL, offsetof (struct pt_regs, icountlevel));
         DEFINE(PT_SYSCALL, offsetof (struct pt_regs, syscall));
         DEFINE(PT_AREG, offsetof (struct pt_regs, areg[0]));
         DEFINE(PT_AREG0, offsetof (struct pt_regs, areg[0]));
diff --git a/arch/xtensa/kernel/entry.S b/arch/xtensa/kernel/entry.S
index 9e271ba009bf..8dc7a2c26ff9 100644
--- a/arch/xtensa/kernel/entry.S
+++ b/arch/xtensa/kernel/entry.S
@@ -125,8 +125,9 @@ _user_exception:
 
         movi    a2, 0
         rsr     a3, SAR
-        wsr     a2, ICOUNTLEVEL
+        xsr     a2, ICOUNTLEVEL
         s32i    a3, a1, PT_SAR
+        s32i    a2, a1, PT_ICOUNTLEVEL
 
         /* Rotate ws so that the current windowbase is at bit0. */
         /* Assume ws = xxwww1yyyy. Rotate ws right, so that a2 = yyyyxxwww1 */
@@ -276,8 +277,9 @@ _kernel_exception:
 
         movi    a2, 0
         rsr     a3, SAR
-        wsr     a2, ICOUNTLEVEL
+        xsr     a2, ICOUNTLEVEL
         s32i    a3, a1, PT_SAR
+        s32i    a2, a1, PT_ICOUNTLEVEL
 
         /* Rotate ws so that the current windowbase is at bit0. */
         /* Assume ws = xxwww1yyyy. Rotate ws right, so that a2 = yyyyxxwww1 */
@@ -330,14 +332,16 @@ _kernel_exception:
 
 common_exception:
 
-        /* Save EXCVADDR, DEBUGCAUSE, and PC, and clear LCOUNT */
+        /* Save some registers, disable loops and clear the syscall flag. */
 
         rsr     a2, DEBUGCAUSE
         rsr     a3, EPC_1
         s32i    a2, a1, PT_DEBUGCAUSE
         s32i    a3, a1, PT_PC
 
+        movi    a2, -1
         rsr     a3, EXCVADDR
+        s32i    a2, a1, PT_SYSCALL
         movi    a2, 0
         s32i    a3, a1, PT_EXCVADDR
         xsr     a2, LCOUNT
@@ -450,27 +454,8 @@ common_exception_return:
 
         /* Restore the state of the task and return from the exception. */
 
-
-        /* If we are returning from a user exception, and the process
-         * to run next has PT_SINGLESTEP set, we want to setup
-         * ICOUNT and ICOUNTLEVEL to step one instruction.
-         * PT_SINGLESTEP is set by sys_ptrace (ptrace.c)
-         */
-
 4:      /* a2 holds GET_CURRENT(a2,a1)  */
 
-        l32i    a3, a2, TI_TASK
-        l32i    a3, a3, TASK_PTRACE
-        bbci.l  a3, PT_SINGLESTEP_BIT, 1f # jump if single-step flag is not set
-
-        movi    a3, -2                  # PT_SINGLESTEP flag is set,
-        movi    a4, 1                   # icountlevel of 1 means it won't
-        wsr     a3, ICOUNT              # start counting until after rfe
-        wsr     a4, ICOUNTLEVEL         # so setup icount & icountlevel.
-        isync
-
-1:
-
 #if XCHAL_EXTRA_SA_SIZE
 
         /* For user exceptions, restore the extra state from the user's TCB. */
@@ -665,6 +650,13 @@ common_exception_exit:
         wsr     a3, LEND
         wsr     a2, LCOUNT
 
+        /* We control single stepping through the ICOUNTLEVEL register. */
+
+        l32i    a2, a1, PT_ICOUNTLEVEL
+        movi    a3, -2
+        wsr     a2, ICOUNTLEVEL
+        wsr     a3, ICOUNT
+
         /* Check if it was double exception. */
 
         l32i    a0, a1, PT_DEPC
diff --git a/arch/xtensa/kernel/head.S b/arch/xtensa/kernel/head.S
index ea89910efa44..67e69139520b 100644
--- a/arch/xtensa/kernel/head.S
+++ b/arch/xtensa/kernel/head.S
@@ -19,6 +19,8 @@
 #include <asm/page.h>
 #include <asm/cacheasm.h>
 
+#include <linux/linkage.h>
+
 /*
  * This module contains the entry code for kernel images. It performs the
  * minimal setup needed to call the generic C routines.
@@ -227,13 +229,14 @@ _startup:
 should_never_return:
         j       should_never_return
 
-        /* Define some common data structures here.  We define them
-         * here in this assembly file due to their unusual alignment
-         * requirements.
-         */
 
-        .comm   swapper_pg_dir,PAGE_SIZE,PAGE_SIZE
-        .comm   empty_bad_page_table,PAGE_SIZE,PAGE_SIZE
-        .comm   empty_bad_page,PAGE_SIZE,PAGE_SIZE
-        .comm   empty_zero_page,PAGE_SIZE,PAGE_SIZE
+/*
+ * BSS section
+ */
+        
+.section ".bss.page_aligned", "w"
+ENTRY(swapper_pg_dir)
+        .fill   PAGE_SIZE, 1, 0
+ENTRY(empty_zero_page)
+        .fill   PAGE_SIZE, 1, 0
 
diff --git a/arch/xtensa/kernel/pci.c b/arch/xtensa/kernel/pci.c
index 45571ccb72d6..77deae5290f0 100644
--- a/arch/xtensa/kernel/pci.c
+++ b/arch/xtensa/kernel/pci.c
@@ -401,7 +401,7 @@ int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
  * Also, think for a moment about likes of floppy.c that
  * include architecture specific parts. They may want to redefine ins/outs.
  *
- * We do not use horroble macroses here because we want to
+ * We do not use horrible macros here because we want to
  * advance pointer by sizeof(size).
  */
 void outsb(unsigned long addr, const void *src, unsigned long count) {
diff --git a/arch/xtensa/kernel/setup.c b/arch/xtensa/kernel/setup.c
index 1ecf6716c327..2e8d398cf196 100644
--- a/arch/xtensa/kernel/setup.c
+++ b/arch/xtensa/kernel/setup.c
@@ -41,6 +41,7 @@
 #include <asm/platform.h>
 #include <asm/page.h>
 #include <asm/setup.h>
+#include <asm/param.h>
 
 #if defined(CONFIG_VGA_CONSOLE) || defined(CONFIG_DUMMY_CONSOLE)
 struct screen_info screen_info = { 0, 24, 0, 0, 0, 80, 0, 0, 0, 24, 1, 16};
diff --git a/arch/xtensa/kernel/signal.c b/arch/xtensa/kernel/signal.c
index 58107672a619..033aae0336d2 100644
--- a/arch/xtensa/kernel/signal.c
+++ b/arch/xtensa/kernel/signal.c
@@ -1,397 +1,239 @@
-// TODO coprocessor stuff
 /*
- *  linux/arch/xtensa/kernel/signal.c
+ * arch/xtensa/kernel/signal.c
  *
- *  Copyright (C) 1991, 1992  Linus Torvalds
- *  1997-11-28  Modified for POSIX.1b signals by Richard Henderson
- *
- *  Joe Taylor <joe@tensilica.com>
- *  Chris Zankel <chris@zankel.net>
+ * Default platform functions.
  *
+ * 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.
  *
+ * Copyright (C) 2005, 2006 Tensilica Inc.
+ * Copyright (C) 1991, 1992  Linus Torvalds
+ * 1997-11-28  Modified for POSIX.1b signals by Richard Henderson
  *
+ * Chris Zankel <chris@zankel.net>
+ * Joe Taylor <joe@tensilica.com>
  */
 
-#include <asm/variant/core.h>
-#include <asm/coprocessor.h>
-#include <linux/sched.h>
-#include <linux/mm.h>
-#include <linux/smp.h>
-#include <linux/kernel.h>
 #include <linux/signal.h>
 #include <linux/errno.h>
-#include <linux/wait.h>
 #include <linux/ptrace.h>
-#include <linux/unistd.h>
-#include <linux/stddef.h>
 #include <linux/personality.h>
+#include <linux/freezer.h>
+
 #include <asm/ucontext.h>
 #include <asm/uaccess.h>
-#include <asm/pgtable.h>
 #include <asm/cacheflush.h>
+#include <asm/coprocessor.h>
+#include <asm/unistd.h>
 
 #define DEBUG_SIG  0
 
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 
-asmlinkage long sys_wait4(pid_t pid,unsigned int * stat_addr, int options,
-                          struct rusage * ru);
 asmlinkage int do_signal(struct pt_regs *regs, sigset_t *oldset);
 
 extern struct task_struct *coproc_owners[];
 
+extern void release_all_cp (struct task_struct *);
 
-/*
- * Atomically swap in the new signal mask, and wait for a signal.
+struct rt_sigframe
+{
+        struct siginfo info;
+        struct ucontext uc;
+        cp_state_t cpstate;
+        unsigned char retcode[6];
+        unsigned int window[4];
+};
+
+/* 
+ * Flush register windows stored in pt_regs to stack.
+ * Returns 1 for errors.
+ *
+ * Note that windowbase, windowstart, and wmask are not updated!
  */
 
-int xtensa_sigsuspend(struct pt_regs *regs)
+int
+flush_window_regs_user(struct pt_regs *regs)
 {
-        old_sigset_t mask = (old_sigset_t) regs->areg[3];
-        sigset_t saveset;
+        const unsigned long ws = regs->windowstart;
+        const unsigned long wb = regs->windowbase;
+        unsigned long sp = 0;
+        unsigned long wm;
+        int err = 1;
+        int base;
 
-        mask &= _BLOCKABLE;
-        spin_lock_irq(&current->sighand->siglock);
-        saveset = current->blocked;
-        siginitset(&current->blocked, mask);
-        recalc_sigpending();
-        spin_unlock_irq(&current->sighand->siglock);
+        /* Return if no other frames. */
 
-        regs->areg[2] = -EINTR;
-        while (1) {
-                current->state = TASK_INTERRUPTIBLE;
-                schedule();
-                if (do_signal(regs, &saveset))
-                        return -EINTR;
-        }
-}
+        if (regs->wmask == 1)
+                return 0;
 
-asmlinkage int
-xtensa_rt_sigsuspend(struct pt_regs *regs)
-{
-        sigset_t *unewset = (sigset_t *) regs->areg[4];
-        size_t sigsetsize = (size_t) regs->areg[3];
-        sigset_t saveset, newset;
-        /* XXX: Don't preclude handling different sized sigset_t's.  */
-        if (sigsetsize != sizeof(sigset_t))
-                return -EINVAL;
+        /* Rotate windowmask and skip empty frames. */
 
-        if (copy_from_user(&newset, unewset, sizeof(newset)))
-                return -EFAULT;
-        sigdelsetmask(&newset, ~_BLOCKABLE);
-        spin_lock_irq(&current->sighand->siglock);
-        saveset = current->blocked;
-        current->blocked = newset;
-        recalc_sigpending();
-        spin_unlock_irq(&current->sighand->siglock);
+        wm = (ws >> wb) | (ws << (XCHAL_NUM_AREGS / 4 - wb));
+        base = (XCHAL_NUM_AREGS / 4) - (regs->wmask >> 4);
+                
+        /* For call8 or call12 frames, we need the previous stack pointer. */
 
-        regs->areg[2] = -EINTR;
-        while (1) {
-                current->state = TASK_INTERRUPTIBLE;
-                schedule();
-                if (do_signal(regs, &saveset))
-                        return -EINTR;
-        }
-}
+        if ((regs->wmask & 2) == 0)
+                if (__get_user(sp, (int*)(regs->areg[base * 4 + 1] - 12)))
+                        goto errout;
 
-asmlinkage int
-xtensa_sigaction(int sig, const struct old_sigaction *act,
-              struct old_sigaction *oact)
-{
-        struct k_sigaction new_ka, old_ka;
-        int ret;
+        /* Spill frames to stack. */
 
-        if (act) {
-                old_sigset_t mask;
-                if (!access_ok(VERIFY_READ, act, sizeof(*act)) ||
-                    __get_user(new_ka.sa.sa_handler, &act->sa_handler) ||
-                    __get_user(new_ka.sa.sa_restorer, &act->sa_restorer))
-                        return -EFAULT;
-                __get_user(new_ka.sa.sa_flags, &act->sa_flags);
-                __get_user(mask, &act->sa_mask);
-                siginitset(&new_ka.sa.sa_mask, mask);
-        }
+        while (base < XCHAL_NUM_AREGS / 4) {
 
-        ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
+                int m = (wm >> base);
+                int inc = 0;
 
-        if (!ret && oact) {
-                if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) ||
-                    __put_user(old_ka.sa.sa_handler, &oact->sa_handler) ||
-                    __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer))
-                        return -EFAULT;
-                __put_user(old_ka.sa.sa_flags, &oact->sa_flags);
-                __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask);
-        }
+                /* Save registers a4..a7 (call8) or a4...a11 (call12) */
 
-        return ret;
-}
+                if (m & 2) {                    /* call4 */
+                        inc = 1;
 
-asmlinkage int
-xtensa_sigaltstack(struct pt_regs *regs)
-{
-        const stack_t *uss = (stack_t *) regs->areg[4];
-        stack_t *uoss = (stack_t *) regs->areg[3];
+                } else if (m & 4) {             /* call8 */
+                        if (copy_to_user((void*)(sp - 32),
+                                           &regs->areg[(base + 1) * 4], 16))
+                                goto errout;
+                        inc = 2;
 
-        if (regs->depc > 64)
-                panic ("Double exception sys_sigreturn\n");
+                } else if (m & 8) {     /* call12 */
+                        if (copy_to_user((void*)(sp - 48),
+                                           &regs->areg[(base + 1) * 4], 32))
+                                goto errout;
+                        inc = 3;
+                }
 
+                /* Save current frame a0..a3 under next SP */
 
-        return do_sigaltstack(uss, uoss, regs->areg[1]);
-}
+                sp = regs->areg[((base + inc) * 4 + 1) % XCHAL_NUM_AREGS];
+                if (copy_to_user((void*)(sp - 16), &regs->areg[base * 4], 16))
+                        goto errout;
+
+                /* Get current stack pointer for next loop iteration. */
+
+                sp = regs->areg[base * 4 + 1];
+                base += inc;
+        }
+
+        return 0;
 
+errout:
+        return err;
+}
 
 /*
- * Do a signal return; undo the signal stack.
+ * Note: We don't copy double exception 'regs', we have to finish double exc. 
+ * first before we return to signal handler! This dbl.exc.handler might cause 
+ * another double exception, but I think we are fine as the situation is the 
+ * same as if we had returned to the signal handerl and got an interrupt 
+ * immediately...
  */
 
-struct sigframe
-{
-        struct sigcontext sc;
-        struct _cpstate cpstate;
-        unsigned long extramask[_NSIG_WORDS-1];
-        unsigned char retcode[6];
-        unsigned int reserved[4]; /* Reserved area for chaining */
-        unsigned int window[4]; /* Window of 4 registers for initial context */
-};
-
-struct rt_sigframe
+static int
+setup_sigcontext(struct sigcontext __user *sc, cp_state_t *cpstate,
+                 struct pt_regs *regs, unsigned long mask)
 {
-        struct siginfo info;
-        struct ucontext uc;
-        struct _cpstate cpstate;
-        unsigned char retcode[6];
-        unsigned int reserved[4]; /* Reserved area for chaining */
-        unsigned int window[4]; /* Window of 4 registers for initial context */
-};
+        int err = 0;
 
-extern void release_all_cp (struct task_struct *);
+#define COPY(x) err |= __put_user(regs->x, &sc->sc_##x)
+        COPY(pc);
+        COPY(ps);
+        COPY(lbeg);
+        COPY(lend);
+        COPY(lcount);
+        COPY(sar);
+#undef COPY
 
+        err |= flush_window_regs_user(regs);
+        err |= __copy_to_user (sc->sc_a, regs->areg, 16 * 4);
 
-// FIXME restore_cpextra
-static inline int
-restore_cpextra (struct _cpstate *buf)
-{
-#if 0
-        /* The signal handler may have used coprocessors in which
-         * case they are still enabled.  We disable them to force a
-         * reloading of the original task's CP state by the lazy
-         * context-switching mechanisms of CP exception handling.
-         * Also, we essentially discard any coprocessor state that the
-         * signal handler created. */
+        // err |= __copy_to_user (sc->sc_a, regs->areg, XCHAL_NUM_AREGS * 4)
 
-        struct task_struct *tsk = current;
-        release_all_cp(tsk);
-        return __copy_from_user(tsk->thread.cpextra, buf, XTENSA_CP_EXTRA_SIZE);
+#if XCHAL_HAVE_CP
+# error Coprocessors unsupported
+        err |= save_cpextra(cpstate);
+        err |= __put_user(err ? NULL : cpstate, &sc->sc_cpstate);
 #endif
-        return 0;
-}
-
-/* Note: We don't copy double exception 'tregs', we have to finish double exc. first before we return to signal handler! This dbl.exc.handler might cause another double exception, but I think we are fine as the situation is the same as if we had returned to the signal handerl and got an interrupt immediately...
- */
+        /* non-iBCS2 extensions.. */
+        err |= __put_user(mask, &sc->oldmask);
 
+        return err;
+}
 
 static int
-restore_sigcontext(struct pt_regs *regs, struct sigcontext *sc)
+restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc)
 {
-        struct thread_struct *thread;
         unsigned int err = 0;
         unsigned long ps;
-        struct _cpstate *buf;
 
 #define COPY(x) err |= __get_user(regs->x, &sc->sc_##x)
         COPY(pc);
-        COPY(depc);
-        COPY(wmask);
         COPY(lbeg);
         COPY(lend);
         COPY(lcount);
         COPY(sar);
-        COPY(windowbase);
-        COPY(windowstart);
 #undef COPY
 
+        /* All registers were flushed to stack. Start with a prestine frame. */
+
+        regs->wmask = 1;
+        regs->windowbase = 0;
+        regs->windowstart = 1;
+
         /* For PS, restore only PS.CALLINC.
          * Assume that all other bits are either the same as for the signal
          * handler, or the user mode value doesn't matter (e.g. PS.OWB).
          */
         err |= __get_user(ps, &sc->sc_ps);
-        regs->ps = (regs->ps & ~PS_CALLINC_MASK)
-                | (ps & PS_CALLINC_MASK);
+        regs->ps = (regs->ps & ~PS_CALLINC_MASK) | (ps & PS_CALLINC_MASK);
 
         /* Additional corruption checks */
 
-        if ((regs->windowbase >= (XCHAL_NUM_AREGS/4))
-        || ((regs->windowstart & ~((1<<(XCHAL_NUM_AREGS/4)) - 1)) != 0) )
-                err = 1;
         if ((regs->lcount > 0)
-        && ((regs->lbeg > TASK_SIZE) || (regs->lend > TASK_SIZE)) )
+            && ((regs->lbeg > TASK_SIZE) || (regs->lend > TASK_SIZE)) )
                 err = 1;
 
-        /* Restore extended register state.
-         * See struct thread_struct in processor.h.
-         */
-        thread = &current->thread;
-
-        err |= __copy_from_user (regs->areg, sc->sc_areg, XCHAL_NUM_AREGS*4);
-        err |= __get_user(buf, &sc->sc_cpstate);
-        if (buf) {
-                if (!access_ok(VERIFY_READ, buf, sizeof(*buf)))
-                        goto badframe;
-                err |= restore_cpextra(buf);
-        }
-
-        regs->syscall = -1;             /* disable syscall checks */
-        return err;
-
-badframe:
-        return 1;
-}
-
-static inline void
-flush_my_cpstate(struct task_struct *tsk)
-{
-        unsigned long flags;
-        local_irq_save(flags);
-
-#if 0   // FIXME
-        for (i = 0; i < XCHAL_CP_NUM; i++) {
-                if (tsk == coproc_owners[i]) {
-                        xthal_validate_cp(i);
-                        xthal_save_cpregs(tsk->thread.cpregs_ptr[i], i);
+        err |= __copy_from_user(regs->areg, sc->sc_a, 16 * 4);
 
-                        /* Invalidate and "disown" the cp to allow
-                         * callers the chance to reset cp state in the
-                         * task_struct. */
+#if XCHAL_HAVE_CP
+# error Coprocessors unsupported
+        /* The signal handler may have used coprocessors in which
+         * case they are still enabled.  We disable them to force a
+         * reloading of the original task's CP state by the lazy
+         * context-switching mechanisms of CP exception handling.
+         * Also, we essentially discard any coprocessor state that the
+         * signal handler created. */
 
-                        xthal_invalidate_cp(i);
-                        coproc_owners[i] = 0;
-                }
+        if (!err) {
+          struct task_struct *tsk = current;
+          release_all_cp(tsk);
+          err |= __copy_from_user(tsk->thread.cpextra, sc->sc_cpstate, 
+                                  XTENSA_CP_EXTRA_SIZE);
         }
 #endif
-        local_irq_restore(flags);
-}
-
-/* Return codes:
-        0:  nothing saved
-        1:  stuff to save, successful
-       -1:  stuff to save, error happened
-*/
-static int
-save_cpextra (struct _cpstate *buf)
-{
-#if XCHAL_CP_NUM == 0
-        return 0;
-#else
-
-        /* FIXME: If a task has never used a coprocessor, there is
-         * no need to save and restore anything.  Tracking this
-         * information would allow us to optimize this section.
-         * Perhaps we can use current->used_math or (current->flags &
-         * PF_USEDFPU) or define a new field in the thread
-         * structure. */
-
-        /* We flush any live, task-owned cp state to the task_struct,
-         * then copy it all to the sigframe.  Then we clear all
-         * cp/extra state in the task_struct, effectively
-         * clearing/resetting all cp/extra state for the signal
-         * handler (cp-exception handling will load these new values
-         * into the cp/extra registers.)  This step is important for
-         * things like a floating-point cp, where the OS must reset
-         * the FCR to the default rounding mode. */
-
-        int err = 0;
-        struct task_struct *tsk = current;
-
-        flush_my_cpstate(tsk);
-        /* Note that we just copy everything: 'extra' and 'cp' state together.*/
-        err |= __copy_to_user(buf, tsk->thread.cp_save, XTENSA_CP_EXTRA_SIZE);
-        memset(tsk->thread.cp_save, 0, XTENSA_CP_EXTRA_SIZE);
-
-#if (XTENSA_CP_EXTRA_SIZE == 0)
-#error Sanity check on memset above, cpextra_size should not be zero.
-#endif
-
-        return err ? -1 : 1;
-#endif
-}
-
-static int
-setup_sigcontext(struct sigcontext *sc, struct _cpstate *cpstate,
-                 struct pt_regs *regs, unsigned long mask)
-{
-        struct thread_struct *thread;
-        int err = 0;
-
-//printk("setup_sigcontext\n");
-#define COPY(x) err |= __put_user(regs->x, &sc->sc_##x)
-        COPY(pc);
-        COPY(ps);
-        COPY(depc);
-        COPY(wmask);
-        COPY(lbeg);
-        COPY(lend);
-        COPY(lcount);
-        COPY(sar);
-        COPY(windowbase);
-        COPY(windowstart);
-#undef COPY
-
-        /* Save extended register state.
-         * See struct thread_struct in processor.h.
-         */
-        thread = &current->thread;
-        err |= __copy_to_user (sc->sc_areg, regs->areg, XCHAL_NUM_AREGS * 4);
-        err |= save_cpextra(cpstate);
-        err |= __put_user(err ? NULL : cpstate, &sc->sc_cpstate);
-        /* non-iBCS2 extensions.. */
-        err |= __put_user(mask, &sc->oldmask);
 
+        regs->syscall = -1;             /* disable syscall checks */
         return err;
 }
 
-asmlinkage int xtensa_sigreturn(struct pt_regs *regs)
-{
-        struct sigframe *frame = (struct sigframe *)regs->areg[1];
-        sigset_t set;
-        if (regs->depc > 64)
-                panic ("Double exception sys_sigreturn\n");
-
-        if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
-                goto badframe;
-
-        if (__get_user(set.sig[0], &frame->sc.oldmask)
-            || (_NSIG_WORDS > 1
-                && __copy_from_user(&set.sig[1], &frame->extramask,
-                                    sizeof(frame->extramask))))
-                goto badframe;
-
-        sigdelsetmask(&set, ~_BLOCKABLE);
-
-        spin_lock_irq(&current->sighand->siglock);
-        current->blocked = set;
-        recalc_sigpending();
-        spin_unlock_irq(&current->sighand->siglock);
 
-        if (restore_sigcontext(regs, &frame->sc))
-                goto badframe;
-        return regs->areg[2];
 
-badframe:
-        force_sig(SIGSEGV, current);
-        return 0;
-}
+/*
+ * Do a signal return; undo the signal stack.
+ */
 
-asmlinkage int xtensa_rt_sigreturn(struct pt_regs *regs)
+asmlinkage long xtensa_rt_sigreturn(long a0, long a1, long a2, long a3,
+                                    long a4, long a5, struct pt_regs *regs)
 {
-        struct rt_sigframe *frame = (struct rt_sigframe *)regs->areg[1];
+        struct rt_sigframe __user *frame;
         sigset_t set;
-        stack_t st;
         int ret;
+
         if (regs->depc > 64)
-        {
-                printk("!!!!!!! DEPC !!!!!!!\n");
-                return 0;
-        }
+                panic("rt_sigreturn in double exception!\n");
+
+        frame = (struct rt_sigframe __user *) regs->areg[1];
 
         if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
                 goto badframe;
@@ -407,13 +249,11 @@ asmlinkage int xtensa_rt_sigreturn(struct pt_regs *regs)
 
         if (restore_sigcontext(regs, &frame->uc.uc_mcontext))
                 goto badframe;
+
         ret = regs->areg[2];
 
-        if (__copy_from_user(&st, &frame->uc.uc_stack, sizeof(st)))
+        if (do_sigaltstack(&frame->uc.uc_stack, NULL, regs->areg[1]) == -EFAULT)
                 goto badframe;
-        /* It is more difficult to avoid calling this function than to
-           call it and ignore errors.  */
-        do_sigaltstack(&st, NULL, regs->areg[1]);
 
         return ret;
 
@@ -422,77 +262,50 @@ badframe:
         return 0;
 }
 
-/*
- * Set up a signal frame.
- */
+
 
 /*
- * Determine which stack to use..
+ * Set up a signal frame.
  */
-static inline void *
-get_sigframe(struct k_sigaction *ka, unsigned long sp, size_t frame_size)
-{
-        if ((ka->sa.sa_flags & SA_ONSTACK) != 0 && ! sas_ss_flags(sp))
-                sp = current->sas_ss_sp + current->sas_ss_size;
-
-        return (void *)((sp - frame_size) & -16ul);
-}
-
-#define USE_SIGRETURN           0
-#define USE_RT_SIGRETURN        1
 
 static int
-gen_return_code(unsigned char *codemem, unsigned int use_rt_sigreturn)
+gen_return_code(unsigned char *codemem)
 {
-        unsigned int retcall;
         int err = 0;
 
-#if 0
-        /* Ignoring SA_RESTORER for now; it's supposed to be obsolete,
-         * and the xtensa glibc doesn't use it.
+        /*
+         * The 12-bit immediate is really split up within the 24-bit MOVI
+         * instruction.  As long as the above system call numbers fit within
+         * 8-bits, the following code works fine. See the Xtensa ISA for
+         * details.
          */
-        if (ka->sa.sa_flags & SA_RESTORER) {
-                regs->pr = (unsigned long) ka->sa.sa_restorer;
-        } else
-#endif /* 0 */
-        {
-
-#if (__NR_sigreturn > 255) || (__NR_rt_sigreturn > 255)
-
-/* The 12-bit immediate is really split up within the 24-bit MOVI
- * instruction.  As long as the above system call numbers fit within
- * 8-bits, the following code works fine. See the Xtensa ISA for
- * details.
- */
 
-#error Generating the MOVI instruction below breaks!
+#if __NR_rt_sigreturn > 255
+# error Generating the MOVI instruction below breaks!
 #endif
 
-                retcall = use_rt_sigreturn ? __NR_rt_sigreturn : __NR_sigreturn;
-
 #ifdef __XTENSA_EB__   /* Big Endian version */
-                /* Generate instruction:  MOVI a2, retcall */
-                err |= __put_user(0x22, &codemem[0]);
-                err |= __put_user(0x0a, &codemem[1]);
-                err |= __put_user(retcall, &codemem[2]);
-                /* Generate instruction:  SYSCALL */
-                err |= __put_user(0x00, &codemem[3]);
-                err |= __put_user(0x05, &codemem[4]);
-                err |= __put_user(0x00, &codemem[5]);
+        /* Generate instruction:  MOVI a2, __NR_rt_sigreturn */
+        err |= __put_user(0x22, &codemem[0]);
+        err |= __put_user(0x0a, &codemem[1]);
+        err |= __put_user(__NR_rt_sigreturn, &codemem[2]);
+        /* Generate instruction:  SYSCALL */
+        err |= __put_user(0x00, &codemem[3]);
+        err |= __put_user(0x05, &codemem[4]);
+        err |= __put_user(0x00, &codemem[5]);
 
 #elif defined __XTENSA_EL__   /* Little Endian version */
-                /* Generate instruction:  MOVI a2, retcall */
-                err |= __put_user(0x22, &codemem[0]);
-                err |= __put_user(0xa0, &codemem[1]);
-                err |= __put_user(retcall, &codemem[2]);
-                /* Generate instruction:  SYSCALL */
-                err |= __put_user(0x00, &codemem[3]);
-                err |= __put_user(0x50, &codemem[4]);
-                err |= __put_user(0x00, &codemem[5]);
+        /* Generate instruction:  MOVI a2, __NR_rt_sigreturn */
+        err |= __put_user(0x22, &codemem[0]);
+        err |= __put_user(0xa0, &codemem[1]);
+        err |= __put_user(__NR_rt_sigreturn, &codemem[2]);
+        /* Generate instruction:  SYSCALL */
+        err |= __put_user(0x00, &codemem[3]);
+        err |= __put_user(0x50, &codemem[4]);
+        err |= __put_user(0x00, &codemem[5]);
 #else
-#error Must use compiler for Xtensa processors.
+# error Must use compiler for Xtensa processors.
 #endif
-        }
 
         /* Flush generated code out of the data cache */
 
@@ -504,97 +317,29 @@ gen_return_code(unsigned char *codemem, unsigned int use_rt_sigreturn)
         return err;
 }
 
-static void
-set_thread_state(struct pt_regs *regs, void *stack, unsigned char *retaddr,
-        void *handler, unsigned long arg1, void *arg2, void *arg3)
-{
-        /* Set up registers for signal handler */
-        start_thread(regs, (unsigned long) handler, (unsigned long) stack);
-
-        /* Set up a stack frame for a call4
-         * Note: PS.CALLINC is set to one by start_thread
-         */
-        regs->areg[4] = (((unsigned long) retaddr) & 0x3fffffff) | 0x40000000;
-        regs->areg[6] = arg1;
-        regs->areg[7] = (unsigned long) arg2;
-        regs->areg[8] = (unsigned long) arg3;
-}
 
-static void setup_frame(int sig, struct k_sigaction *ka,
+static void setup_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
                         sigset_t *set, struct pt_regs *regs)
 {
-        struct sigframe *frame;
+        struct rt_sigframe *frame;
         int err = 0;
         int signal;
+        unsigned long sp, ra;
 
-        frame = get_sigframe(ka, regs->areg[1], sizeof(*frame));
-        if (regs->depc > 64)
-        {
-                printk("!!!!!!! DEPC !!!!!!!\n");
-                return;
-        }
-
-
-        if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
-                goto give_sigsegv;
-
-        signal = current_thread_info()->exec_domain
-                && current_thread_info()->exec_domain->signal_invmap
-                && sig < 32
-                ? current_thread_info()->exec_domain->signal_invmap[sig]
-                : sig;
-
-        err |= setup_sigcontext(&frame->sc, &frame->cpstate, regs, set->sig[0]);
+        sp = regs->areg[1];
 
-        if (_NSIG_WORDS > 1) {
-                err |= __copy_to_user(frame->extramask, &set->sig[1],
-                                      sizeof(frame->extramask));
+        if ((ka->sa.sa_flags & SA_ONSTACK) != 0 && ! on_sig_stack(sp)) {
+                sp = current->sas_ss_sp + current->sas_ss_size;
         }
 
-        /* Create sys_sigreturn syscall in stack frame */
-        err |= gen_return_code(frame->retcode, USE_SIGRETURN);
-
-        if (err)
-                goto give_sigsegv;
-
-        /* Create signal handler execution context.
-         * Return context not modified until this point.
-         */
-        set_thread_state(regs, frame, frame->retcode,
-                ka->sa.sa_handler, signal, &frame->sc, NULL);
-
-        /* Set access mode to USER_DS.  Nomenclature is outdated, but
-         * functionality is used in uaccess.h
-         */
-        set_fs(USER_DS);
-
-
-#if DEBUG_SIG
-        printk("SIG deliver (%s:%d): signal=%d sp=%p pc=%08x\n",
-                current->comm, current->pid, signal, frame, regs->pc);
-#endif
-
-        return;
-
-give_sigsegv:
-        if (sig == SIGSEGV)
-                ka->sa.sa_handler = SIG_DFL;
-        force_sig(SIGSEGV, current);
-}
-
-static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
-                           sigset_t *set, struct pt_regs *regs)
-{
-        struct rt_sigframe *frame;
-        int err = 0;
-        int signal;
+        frame = (void *)((sp - sizeof(*frame)) & -16ul);
 
-        frame = get_sigframe(ka, regs->areg[1], sizeof(*frame));
         if (regs->depc > 64)
                 panic ("Double exception sys_sigreturn\n");
 
-        if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
+        if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) {
                 goto give_sigsegv;
+        }
 
         signal = current_thread_info()->exec_domain
                 && current_thread_info()->exec_domain->signal_invmap
@@ -602,9 +347,12 @@ static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
                 ? current_thread_info()->exec_domain->signal_invmap[sig]
                 : sig;
 
-        err |= copy_siginfo_to_user(&frame->info, info);
+        if (ka->sa.sa_flags & SA_SIGINFO) {
+                err |= copy_siginfo_to_user(&frame->info, info);
+        }
+
+        /* Create the user context.  */
 
-        /* Create the ucontext.  */
         err |= __put_user(0, &frame->uc.uc_flags);
         err |= __put_user(0, &frame->uc.uc_link);
         err |= __put_user((void *)current->sas_ss_sp,
@@ -617,16 +365,31 @@ static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
         err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
 
         /* Create sys_rt_sigreturn syscall in stack frame */
-        err |= gen_return_code(frame->retcode, USE_RT_SIGRETURN);
 
-        if (err)
+        err |= gen_return_code(frame->retcode);
+
+        if (err) {
                 goto give_sigsegv;
+        }
+                
 
-        /* Create signal handler execution context.
+        /* 
+         * Create signal handler execution context.
          * Return context not modified until this point.
          */
-        set_thread_state(regs, frame, frame->retcode,
-                ka->sa.sa_handler, signal, &frame->info, &frame->uc);
+
+        /* Set up registers for signal handler */
+        start_thread(regs, (unsigned long) ka->sa.sa_handler, 
+                     (unsigned long) frame);
+
+        /* Set up a stack frame for a call4
+         * Note: PS.CALLINC is set to one by start_thread
+         */
+        ra = (unsigned long) frame->retcode;
+        regs->areg[4] = (((unsigned long) ra) & 0x3fffffff) | 0x40000000;
+        regs->areg[6] = (unsigned long) signal;
+        regs->areg[7] = (unsigned long) &frame->info;
+        regs->areg[8] = (unsigned long) &frame->uc;
 
         /* Set access mode to USER_DS.  Nomenclature is outdated, but
          * functionality is used in uaccess.h
@@ -646,6 +409,48 @@ give_sigsegv:
         force_sig(SIGSEGV, current);
 }
 
+/*
+ * Atomically swap in the new signal mask, and wait for a signal.
+ */
+
+asmlinkage long xtensa_rt_sigsuspend(sigset_t __user *unewset, 
+                                     size_t sigsetsize,
+                                     long a2, long a3, long a4, long a5, 
+                                     struct pt_regs *regs)
+{
+        sigset_t saveset, newset;
+
+        /* XXX: Don't preclude handling different sized sigset_t's.  */
+        if (sigsetsize != sizeof(sigset_t))
+                return -EINVAL;
+
+        if (copy_from_user(&newset, unewset, sizeof(newset)))
+                return -EFAULT;
+
+        sigdelsetmask(&newset, ~_BLOCKABLE);
+        spin_lock_irq(&current->sighand->siglock);
+        saveset = current->blocked;
+        current->blocked = newset;
+        recalc_sigpending();
+        spin_unlock_irq(&current->sighand->siglock);
+
+        regs->areg[2] = -EINTR;
+        while (1) {
+                current->state = TASK_INTERRUPTIBLE;
+                schedule();
+                if (do_signal(regs, &saveset))
+                        return -EINTR;
+        }
+}
+
+asmlinkage long xtensa_sigaltstack(const stack_t __user *uss, 
+                                   stack_t __user *uoss,
+                                   long a2, long a3, long a4, long a5,
+                                   struct pt_regs *regs)
+{
+        return do_sigaltstack(uss, uoss, regs->areg[1]);
+}
+
 
 
 /*
@@ -663,51 +468,89 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
         int signr;
         struct k_sigaction ka;
 
+        if (!user_mode(regs))
+                return 0;
+
+        if (try_to_freeze())
+                goto no_signal;
+
         if (!oldset)
                 oldset = &current->blocked;
 
+        task_pt_regs(current)->icountlevel = 0;
+
         signr = get_signal_to_deliver(&info, &ka, regs, NULL);
 
-        /* Are we from a system call? */
-        if (regs->syscall >= 0) {
-                /* If so, check system call restarting.. */
-                switch (regs->areg[2]) {
-                        case ERESTARTNOHAND:
-                        case ERESTART_RESTARTBLOCK:
-                                regs->areg[2] = -EINTR;
-                                break;
+        if (signr > 0) {
+
+                /* Are we from a system call? */
+
+                if ((signed)regs->syscall >= 0) {
 
-                        case ERESTARTSYS:
-                                if (!(ka.sa.sa_flags & SA_RESTART)) {
+                        /* If so, check system call restarting.. */
+
+                        switch (regs->areg[2]) {
+                                case -ERESTARTNOHAND:
+                                case -ERESTART_RESTARTBLOCK:
                                         regs->areg[2] = -EINTR;
                                         break;
-                                }
-                        /* fallthrough */
-                        case ERESTARTNOINTR:
-                                regs->areg[2] = regs->syscall;
-                                regs->pc -= 3;
+
+                                case -ERESTARTSYS:
+                                        if (!(ka.sa.sa_flags & SA_RESTART)) {
+                                                regs->areg[2] = -EINTR;
+                                                break;
+                                        }
+                                        /* fallthrough */
+                                case -ERESTARTNOINTR:
+                                        regs->areg[2] = regs->syscall;
+                                        regs->pc -= 3;
+                                        break;
+
+                                default:
+                                        /* nothing to do */
+                                        if (regs->areg[2] != 0)
+                                        break;
+                        }
                 }
-        }
 
-        if (signr == 0)
-                return 0;               /* no signals delivered */
+                /* Whee!  Actually deliver the signal.  */
+                /* Set up the stack frame */
+                setup_frame(signr, &ka, &info, oldset, regs);
 
-        /* Whee!  Actually deliver the signal.  */
+                if (ka.sa.sa_flags & SA_ONESHOT)
+                        ka.sa.sa_handler = SIG_DFL;
 
-        /* Set up the stack frame */
-        if (ka.sa.sa_flags & SA_SIGINFO)
-                setup_rt_frame(signr, &ka, &info, oldset, regs);
-        else
-                setup_frame(signr, &ka, oldset, regs);
+                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, signr);
+                recalc_sigpending();
+                spin_unlock_irq(&current->sighand->siglock);
+                if (current->ptrace & PT_SINGLESTEP)
+                        task_pt_regs(current)->icountlevel = 1;
 
-        if (ka.sa.sa_flags & SA_ONESHOT)
-                ka.sa.sa_handler = SIG_DFL;
+                return 1;
+        }
 
-        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, signr);
-        recalc_sigpending();
-        spin_unlock_irq(&current->sighand->siglock);
-        return 1;
+no_signal:
+        /* Did we come from a system call? */
+        if ((signed) regs->syscall >= 0) {
+                /* Restart the system call - no handlers present */
+                switch (regs->areg[2]) {
+                case -ERESTARTNOHAND:
+                case -ERESTARTSYS:
+                case -ERESTARTNOINTR:
+                        regs->areg[2] = regs->syscall;
+                        regs->pc -= 3;
+                        break;
+                case -ERESTART_RESTARTBLOCK:
+                        regs->areg[2] = __NR_restart_syscall;
+                        regs->pc -= 3;
+                        break;
+                }
+        }
+        if (current->ptrace & PT_SINGLESTEP)
+                task_pt_regs(current)->icountlevel = 1;
+        return 0;
 }
+
diff --git a/arch/xtensa/kernel/vmlinux.lds.S b/arch/xtensa/kernel/vmlinux.lds.S
index 4b7b4ff79973..b0582c3c5f8d 100644
--- a/arch/xtensa/kernel/vmlinux.lds.S
+++ b/arch/xtensa/kernel/vmlinux.lds.S
@@ -84,9 +84,7 @@ SECTIONS
   {
     /* The .head.text section must be the first section! */
     *(.head.text)
-    *(.literal)
-    TEXT_TEXT
-    *(.srom.text)
+    *(.literal .text)
     VMLINUX_SYMBOL(__sched_text_start) = .;
     *(.sched.literal .sched.text)
     VMLINUX_SYMBOL(__sched_text_end) = .;
@@ -96,6 +94,7 @@ SECTIONS
 
   }
   _etext = .;
+  PROVIDE (etext = .);
 
   . = ALIGN(16);
 
@@ -103,32 +102,6 @@ SECTIONS
 
   /*  Relocation table */
 
-  . = ALIGN(16);
-  __boot_reloc_table_start = ABSOLUTE(.);
-
-  __relocate : {
-
-    RELOCATE_ENTRY(_WindowVectors_text,
-                   .WindowVectors.text);
-#if 0
-    RELOCATE_ENTRY(_KernelExceptionVector_literal,
-                   .KernelExceptionVector.literal);
-#endif
-    RELOCATE_ENTRY(_KernelExceptionVector_text,
-                   .KernelExceptionVector.text);
-#if 0
-    RELOCATE_ENTRY(_UserExceptionVector_literal,
-                   .UserExceptionVector.literal);
-#endif
-    RELOCATE_ENTRY(_UserExceptionVector_text,
-                   .UserExceptionVector.text);
-    RELOCATE_ENTRY(_DoubleExceptionVector_literal,
-                   .DoubleExceptionVector.literal);
-    RELOCATE_ENTRY(_DoubleExceptionVector_text,
-                   .DoubleExceptionVector.text);
-  }
-  __boot_reloc_table_end = ABSOLUTE(.) ;
-
   .fixup   : { *(.fixup) }
 
   . = ALIGN(16);
@@ -145,8 +118,7 @@ SECTIONS
   _fdata = .;
   .data :
   {
-    DATA_DATA
-    CONSTRUCTORS
+    *(.data) CONSTRUCTORS
     . = ALIGN(XCHAL_ICACHE_LINESIZE);
     *(.data.cacheline_aligned)
   }
@@ -174,6 +146,22 @@ SECTIONS
     __tagtable_begin = .;
     *(.taglist)
     __tagtable_end = .;
+
+    . = ALIGN(16);
+    __boot_reloc_table_start = ABSOLUTE(.);
+
+    RELOCATE_ENTRY(_WindowVectors_text,
+                   .WindowVectors.text);
+    RELOCATE_ENTRY(_KernelExceptionVector_text,
+                   .KernelExceptionVector.text);
+    RELOCATE_ENTRY(_UserExceptionVector_text,
+                   .UserExceptionVector.text);
+    RELOCATE_ENTRY(_DoubleExceptionVector_literal,
+                   .DoubleExceptionVector.literal);
+    RELOCATE_ENTRY(_DoubleExceptionVector_text,
+                   .DoubleExceptionVector.text);
+  
+    __boot_reloc_table_end = ABSOLUTE(.) ;
   }
 
   . = ALIGN(XCHAL_ICACHE_LINESIZE);
@@ -194,16 +182,6 @@ SECTIONS
 
   SECURITY_INIT
 
-  . = ALIGN(4);
-
-  __start___ftr_fixup = .;
-  __ftr_fixup : { *(__ftr_fixup) }
-  __stop___ftr_fixup = .;
-
-  . = ALIGN(4096);
-  __per_cpu_start = .;
-  .data.percpu  : { *(.data.percpu) }
-  __per_cpu_end = .;
 
 #ifdef CONFIG_BLK_DEV_INITRD
   . = ALIGN(4096);
@@ -212,6 +190,12 @@ SECTIONS
   __initramfs_end = .;
 #endif
 
+  . = ALIGN(4096);
+  __per_cpu_start = .;
+  .data.percpu  : { *(.data.percpu) }
+  __per_cpu_end = .;
+
+
   /* We need this dummy segment here */
 
   . = ALIGN(4);
@@ -273,9 +257,9 @@ SECTIONS
 
   /* BSS section */
   _bss_start = .;
-  .sbss : { *(.sbss) *(.scommon) }
-  .bss : { *(COMMON) *(.bss) }
+  .bss : { *(.bss.page_aligned) *(.bss) }
   _bss_end = .;
+
   _end = .;
 
   /* only used by the boot loader  */
@@ -293,16 +277,16 @@ SECTIONS
     *(.ResetVector.text)
   }
 
-
   /* Sections to be discarded */
   /DISCARD/ :
   {
-        *(.text.exit)
-        *(.text.exit.literal)
-        *(.data.exit)
+        *(.exit.literal .exit.text)
+        *(.exit.data)
         *(.exitcall.exit)
   }
 
+  .xt.lit : { *(.xt.lit) }
+  .xt.prop : { *(.xt.prop) }
 
   .debug  0 :  { *(.debug) }
   .line  0 :  { *(.line) }
diff --git a/arch/xtensa/kernel/xtensa_ksyms.c b/arch/xtensa/kernel/xtensa_ksyms.c
index cd7e6a020602..60dbdb43fb4c 100644
--- a/arch/xtensa/kernel/xtensa_ksyms.c
+++ b/arch/xtensa/kernel/xtensa_ksyms.c
@@ -38,21 +38,10 @@
 /*
  * String functions
  */
-EXPORT_SYMBOL(memcmp);
 EXPORT_SYMBOL(memset);
 EXPORT_SYMBOL(memcpy);
 EXPORT_SYMBOL(memmove);
-EXPORT_SYMBOL(memchr);
-EXPORT_SYMBOL(strcat);
-EXPORT_SYMBOL(strchr);
-EXPORT_SYMBOL(strlen);
-EXPORT_SYMBOL(strncat);
-EXPORT_SYMBOL(strnlen);
-EXPORT_SYMBOL(strrchr);
-EXPORT_SYMBOL(strstr);
 
-EXPORT_SYMBOL(enable_irq);
-EXPORT_SYMBOL(disable_irq);
 EXPORT_SYMBOL(kernel_thread);
 
 /*
diff --git a/arch/xtensa/lib/strncpy_user.S b/arch/xtensa/lib/strncpy_user.S
index a834057bda6b..b2655d94558d 100644
--- a/arch/xtensa/lib/strncpy_user.S
+++ b/arch/xtensa/lib/strncpy_user.S
@@ -25,18 +25,18 @@
 /*
  * char *__strncpy_user(char *dst, const char *src, size_t len)
  */
-.text
-.begin literal
-.align  4
-.Lmask0:
-        .byte   0xff, 0x00, 0x00, 0x00
-.Lmask1:
-        .byte   0x00, 0xff, 0x00, 0x00
-.Lmask2:
-        .byte   0x00, 0x00, 0xff, 0x00
-.Lmask3:
-        .byte   0x00, 0x00, 0x00, 0xff
-.end literal
+
+#ifdef __XTENSA_EB__
+# define MASK0 0xff000000
+# define MASK1 0x00ff0000
+# define MASK2 0x0000ff00
+# define MASK3 0x000000ff
+#else
+# define MASK0 0x000000ff
+# define MASK1 0x0000ff00
+# define MASK2 0x00ff0000
+# define MASK3 0xff000000
+#endif
 
 # Register use
 #   a0/ return address
@@ -53,6 +53,7 @@
 #   a11/ dst
 #   a12/ tmp
 
+.text
 .align  4
 .global __strncpy_user
 .type   __strncpy_user,@function
@@ -61,10 +62,10 @@ __strncpy_user:
         # a2/ dst, a3/ src, a4/ len
         mov     a11, a2         # leave dst in return value register
         beqz    a4, .Lret       # if len is zero
-        l32r    a5, .Lmask0     # mask for byte 0
-        l32r    a6, .Lmask1     # mask for byte 1
-        l32r    a7, .Lmask2     # mask for byte 2
-        l32r    a8, .Lmask3     # mask for byte 3
+        movi    a5, MASK0       # mask for byte 0
+        movi    a6, MASK1       # mask for byte 1
+        movi    a7, MASK2       # mask for byte 2
+        movi    a8, MASK3       # mask for byte 3
         bbsi.l  a3, 0, .Lsrc1mod2 # if only  8-bit aligned
         bbsi.l  a3, 1, .Lsrc2mod4 # if only 16-bit aligned
 .Lsrcaligned:   # return here when src is word-aligned
diff --git a/arch/xtensa/lib/strnlen_user.S b/arch/xtensa/lib/strnlen_user.S
index 5e9c1e709b2e..ad3f616322ca 100644
--- a/arch/xtensa/lib/strnlen_user.S
+++ b/arch/xtensa/lib/strnlen_user.S
@@ -24,18 +24,18 @@
 /*
  * size_t __strnlen_user(const char *s, size_t len)
  */
-.text
-.begin literal
-.align  4
-.Lmask0:
-        .byte   0xff, 0x00, 0x00, 0x00
-.Lmask1:
-        .byte   0x00, 0xff, 0x00, 0x00
-.Lmask2:
-        .byte   0x00, 0x00, 0xff, 0x00
-.Lmask3:
-        .byte   0x00, 0x00, 0x00, 0xff
-.end literal
+
+#ifdef __XTENSA_EB__
+# define MASK0 0xff000000
+# define MASK1 0x00ff0000
+# define MASK2 0x0000ff00
+# define MASK3 0x000000ff
+#else
+# define MASK0 0x000000ff
+# define MASK1 0x0000ff00
+# define MASK2 0x00ff0000
+# define MASK3 0xff000000
+#endif
 
 # Register use:
 #   a2/ src
@@ -48,6 +48,7 @@
 #   a9/ tmp
 #   a10/ tmp
 
+.text
 .align  4
 .global __strnlen_user
 .type   __strnlen_user,@function
@@ -56,10 +57,10 @@ __strnlen_user:
         # a2/ s, a3/ len
         addi    a4, a2, -4      # because we overincrement at the end;
                                 # we compensate with load offsets of 4
-        l32r    a5, .Lmask0     # mask for byte 0
-        l32r    a6, .Lmask1     # mask for byte 1
-        l32r    a7, .Lmask2     # mask for byte 2
-        l32r    a8, .Lmask3     # mask for byte 3
+        movi    a5, MASK0       # mask for byte 0
+        movi    a6, MASK1       # mask for byte 1
+        movi    a7, MASK2       # mask for byte 2
+        movi    a8, MASK3       # mask for byte 3
         bbsi.l  a2, 0, .L1mod2  # if only  8-bit aligned
         bbsi.l  a2, 1, .L2mod4  # if only 16-bit aligned
 
diff --git a/arch/xtensa/mm/init.c b/arch/xtensa/mm/init.c
index e1ec2d1e8189..8415c76f11c2 100644
--- a/arch/xtensa/mm/init.c
+++ b/arch/xtensa/mm/init.c
@@ -205,7 +205,7 @@ void __init init_mmu (void)
         /* Writing zeros to the <t>TLBCFG special registers ensure
          * that valid values exist in the register.  For existing
          * PGSZID<w> fields, zero selects the first element of the
-         * page-size array.  For nonexistant PGSZID<w> fields, zero is
+         * page-size array.  For nonexistent PGSZID<w> fields, zero is
          * the best value to write.  Also, when changing PGSZID<w>
          * fields, the corresponding TLB must be flushed.
          */
diff --git a/arch/xtensa/platform-iss/network.c b/arch/xtensa/platform-iss/network.c
index 4bfe333be229..f09962fa98c0 100644
--- a/arch/xtensa/platform-iss/network.c
+++ b/arch/xtensa/platform-iss/network.c
@@ -473,7 +473,7 @@ static int iss_net_open(struct net_device *dev)
         netif_start_queue(dev);
 
         /* clear buffer - it can happen that the host side of the interface
-         * is full when we gethere. In this case, new data is never queued,
+         * is full when we get here. In this case, new data is never queued,
          * SIGIOs never arrive, and the net never works.
          */
         while ((err = iss_net_rx(dev)) > 0)