31 files changed, 248 insertions, 131 deletions

diff --git a/Documentation/x86_64/boot-options.txt b/Documentation/x86_64/boot-options.txt
index 153740f460a6..1921353259ae 100644
--- a/Documentation/x86_64/boot-options.txt
+++ b/Documentation/x86_64/boot-options.txt
@@ -52,6 +52,10 @@ APICs
                  apicmaintimer. Useful when your PIT timer is totally
                  broken.
 
+   disable_8254_timer / enable_8254_timer
+                 Enable interrupt 0 timer routing over the 8254 in addition to over
+                 the IO-APIC. The kernel tries to set a sensible default.
+
 Early Console
 
    syntax: earlyprintk=vga
diff --git a/Makefile b/Makefile
index 77a448c8e776..46eea76bc570 100644
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,7 @@
 VERSION = 2
 PATCHLEVEL = 6
 SUBLEVEL = 16
-EXTRAVERSION =-rc4
+EXTRAVERSION =-rc5
 NAME=Sliding Snow Leopard
 
 # *DOCUMENTATION*
diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index 0afec8566e7b..5b1a7d46d1d9 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -733,7 +733,7 @@ config PHYSICAL_START
 
 config HOTPLUG_CPU
         bool "Support for hot-pluggable CPUs (EXPERIMENTAL)"
-        depends on SMP && HOTPLUG && EXPERIMENTAL
+        depends on SMP && HOTPLUG && EXPERIMENTAL && !X86_VOYAGER
         ---help---
           Say Y here to experiment with turning CPUs off and on.  CPUs
           can be controlled through /sys/devices/system/cpu.
@@ -1060,6 +1060,7 @@ source "arch/i386/oprofile/Kconfig"
 
 config KPROBES
         bool "Kprobes (EXPERIMENTAL)"
+        depends on EXPERIMENTAL && MODULES
         help
           Kprobes allows you to trap at almost any kernel address and
           execute a callback function.  register_kprobe() establishes
diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index 79577f0ace98..8309a7b2cd63 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -44,9 +44,6 @@ extern void __init clustered_apic_check(void);
 extern int gsi_irq_sharing(int gsi);
 #include <asm/proto.h>
 
-static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id) { return 0; }
-
-
 #else                           /* X86 */
 
 #ifdef  CONFIG_X86_LOCAL_APIC
diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index f2dd218d88cb..235822b3f41b 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -2566,8 +2566,10 @@ int __init io_apic_get_unique_id (int ioapic, int apic_id)
                 spin_unlock_irqrestore(&ioapic_lock, flags);
 
                 /* Sanity check */
-                if (reg_00.bits.ID != apic_id)
-                        panic("IOAPIC[%d]: Unable change apic_id!\n", ioapic);
+                if (reg_00.bits.ID != apic_id) {
+                        printk("IOAPIC[%d]: Unable to change apic_id!\n", ioapic);
+                        return -1;
+                }
         }
 
         apic_printk(APIC_VERBOSE, KERN_INFO
diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index e7609abf3796..e6e2f43db85e 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -915,6 +915,7 @@ void __init mp_register_ioapic (
         u32                     gsi_base)
 {
         int                     idx = 0;
+        int                     tmpid;
 
         if (nr_ioapics >= MAX_IO_APICS) {
                 printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded "
@@ -935,9 +936,14 @@ void __init mp_register_ioapic (
 
         set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
         if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) && (boot_cpu_data.x86 < 15))
-                mp_ioapics[idx].mpc_apicid = io_apic_get_unique_id(idx, id);
+                tmpid = io_apic_get_unique_id(idx, id);
         else
-                mp_ioapics[idx].mpc_apicid = id;
+                tmpid = id;
+        if (tmpid == -1) {
+                nr_ioapics--;
+                return;
+        }
+        mp_ioapics[idx].mpc_apicid = tmpid;
         mp_ioapics[idx].mpc_apicver = io_apic_get_version(idx);
         
         /* 
diff --git a/arch/i386/mach-voyager/voyager_basic.c b/arch/i386/mach-voyager/voyager_basic.c
index 6761d294f260..b584060ec004 100644
--- a/arch/i386/mach-voyager/voyager_basic.c
+++ b/arch/i386/mach-voyager/voyager_basic.c
@@ -25,7 +25,6 @@
 #include <linux/sysrq.h>
 #include <linux/smp.h>
 #include <linux/nodemask.h>
-#include <asm/cpu.h>
 #include <asm/io.h>
 #include <asm/voyager.h>
 #include <asm/vic.h>
@@ -331,16 +330,3 @@ void machine_power_off(void)
         if (pm_power_off)
                 pm_power_off();
 }
-
-static struct i386_cpu cpu_devices[NR_CPUS];
-
-static int __init topology_init(void)
-{
-        int i;
-
-        for_each_present_cpu(i)
-                register_cpu(&cpu_devices[i].cpu, i, NULL);
-        return 0;
-}
-
-subsys_initcall(topology_init);
diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 845cd0902a50..a85ea9d37f05 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -453,6 +453,7 @@ source "arch/ia64/oprofile/Kconfig"
 
 config KPROBES
         bool "Kprobes (EXPERIMENTAL)"
+        depends on EXPERIMENTAL && MODULES
         help
           Kprobes allows you to trap at almost any kernel address and
           execute a callback function.  register_kprobe() establishes
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 80d114a3a837..a834f9e0bbb3 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -963,7 +963,7 @@ source "arch/powerpc/oprofile/Kconfig"
 
 config KPROBES
         bool "Kprobes (EXPERIMENTAL)"
-        depends on PPC64
+        depends on PPC64 && EXPERIMENTAL && MODULES
         help
           Kprobes allows you to trap at almost any kernel address and
           execute a callback function.  register_kprobe() establishes
diff --git a/arch/sparc64/Kconfig b/arch/sparc64/Kconfig
index ab733be9af08..4c0a50a76554 100644
--- a/arch/sparc64/Kconfig
+++ b/arch/sparc64/Kconfig
@@ -383,6 +383,7 @@ source "arch/sparc64/oprofile/Kconfig"
 
 config KPROBES
         bool "Kprobes (EXPERIMENTAL)"
+        depends on EXPERIMENTAL && MODULES
         help
           Kprobes allows you to trap at almost any kernel address and
           execute a callback function.  register_kprobe() establishes
diff --git a/arch/sparc64/kernel/setup.c b/arch/sparc64/kernel/setup.c
index 054461e6946d..158bd31e15b7 100644
--- a/arch/sparc64/kernel/setup.c
+++ b/arch/sparc64/kernel/setup.c
@@ -542,6 +542,8 @@ void __init setup_arch(char **cmdline_p)
         }
 #endif
 
+        smp_setup_cpu_possible_map();
+
         paging_init();
 }
 
diff --git a/arch/sparc64/kernel/smp.c b/arch/sparc64/kernel/smp.c
index 1fb6323e65a4..1f7ad8a69052 100644
--- a/arch/sparc64/kernel/smp.c
+++ b/arch/sparc64/kernel/smp.c
@@ -1079,18 +1079,12 @@ int setup_profiling_timer(unsigned int multiplier)
         return 0;
 }
 
+/* Constrain the number of cpus to max_cpus.  */
 void __init smp_prepare_cpus(unsigned int max_cpus)
 {
-        int instance, mid;
-
-        instance = 0;
-        while (!cpu_find_by_instance(instance, NULL, &mid)) {
-                if (mid < max_cpus)
-                        cpu_set(mid, phys_cpu_present_map);
-                instance++;
-        }
-
         if (num_possible_cpus() > max_cpus) {
+                int instance, mid;
+
                 instance = 0;
                 while (!cpu_find_by_instance(instance, NULL, &mid)) {
                         if (mid != boot_cpu_id) {
@@ -1105,6 +1099,22 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
         smp_store_cpu_info(boot_cpu_id);
 }
 
+/* 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;
+
+        instance = 0;
+        while (!cpu_find_by_instance(instance, NULL, &mid)) {
+                if (mid < NR_CPUS)
+                        cpu_set(mid, phys_cpu_present_map);
+                instance++;
+        }
+}
+
 void __devinit smp_prepare_boot_cpu(void)
 {
         if (hard_smp_processor_id() >= NR_CPUS) {
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 2f9deca31cc9..e18eb79bf855 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -354,21 +354,6 @@ config HPET_TIMER
           as it is off-chip.  You can find the HPET spec at
           <http://www.intel.com/hardwaredesign/hpetspec.htm>.
 
-config X86_PM_TIMER
-        bool "PM timer" if EMBEDDED
-        depends on ACPI
-        default y
-        help
-          Support the ACPI PM timer for time keeping. This is slow,
-          but is useful on some chipsets without HPET on systems with more
-          than one CPU. On a single processor or single socket multi core
-          system it is normally not required.
-          When the PM timer is active 64bit vsyscalls are disabled
-          and should not be enabled (/proc/sys/kernel/vsyscall64 should
-          not be changed).
-          The kernel selects the PM timer only as a last resort, so it is
-          useful to enable just in case.
-
 config HPET_EMULATE_RTC
         bool "Provide RTC interrupt"
         depends on HPET_TIMER && RTC=y
@@ -592,6 +577,7 @@ source "arch/x86_64/oprofile/Kconfig"
 
 config KPROBES
         bool "Kprobes (EXPERIMENTAL)"
+        depends on EXPERIMENTAL && MODULES
         help
           Kprobes allows you to trap at almost any kernel address and
           execute a callback function.  register_kprobe() establishes
diff --git a/arch/x86_64/kernel/aperture.c b/arch/x86_64/kernel/aperture.c
index e4e2b7d01f89..a0f955b9995f 100644
--- a/arch/x86_64/kernel/aperture.c
+++ b/arch/x86_64/kernel/aperture.c
@@ -248,7 +248,7 @@ void __init iommu_hole_init(void)
                 /* Got the aperture from the AGP bridge */
         } else if (swiotlb && !valid_agp) {
                 /* Do nothing */
-        } else if ((!no_iommu && end_pfn >= MAX_DMA32_PFN) ||
+        } else if ((!no_iommu && end_pfn > MAX_DMA32_PFN) ||
                    force_iommu ||
                    valid_agp ||
                    fallback_aper_force) { 
diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index e5b14c57eaa0..d70605eda333 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -962,12 +962,14 @@ void smp_apic_timer_interrupt(struct pt_regs *regs)
         irq_exit();
 }
 
+int __initdata unsync_tsc_on_multicluster;
+
 /*
  * oem_force_hpet_timer -- force HPET mode for some boxes.
  *
  * Thus far, the major user of this is IBM's Summit2 series:
  *
- * Clustered boxes may have unsynced TSC problems if they are
+ * Some clustered boxes may have unsynced TSC problems if they are
  * multi-chassis. Use available data to take a good guess.
  * If in doubt, go HPET.
  */
@@ -977,6 +979,11 @@ __cpuinit int oem_force_hpet_timer(void)
         unsigned id;
         DECLARE_BITMAP(clustermap, NUM_APIC_CLUSTERS);
 
+        /* Only do this check on IBM machines - big Unisys systems
+           use multiple clusters too, but have synchronized TSC */
+        if (!unsync_tsc_on_multicluster)
+                return 0;
+
         bitmap_zero(clustermap, NUM_APIC_CLUSTERS);
 
         for (i = 0; i < NR_CPUS; i++) {
diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index 2585c1d92b26..ffed464e6b12 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -50,6 +50,8 @@ static int no_timer_check;
 
 int disable_timer_pin_1 __initdata;
 
+int timer_over_8254 __initdata = 1;
+
 /* Where if anywhere is the i8259 connect in external int mode */
 static struct { int pin, apic; } ioapic_i8259 = { -1, -1 };
 
@@ -251,6 +253,20 @@ static int __init enable_ioapic_setup(char *str)
 __setup("noapic", disable_ioapic_setup);
 __setup("apic", enable_ioapic_setup);
 
+static int __init setup_disable_8254_timer(char *s)
+{
+        timer_over_8254 = -1;
+        return 1;
+}
+static int __init setup_enable_8254_timer(char *s)
+{
+        timer_over_8254 = 2;
+        return 1;
+}
+
+__setup("disable_8254_timer", setup_disable_8254_timer);
+__setup("enable_8254_timer", setup_enable_8254_timer);
+
 #include <asm/pci-direct.h>
 #include <linux/pci_ids.h>
 #include <linux/pci.h>
@@ -309,27 +325,20 @@ void __init check_ioapic(void)
 #endif
                                         /* RED-PEN skip them on mptables too? */
                                         return;
+
+                                /* This should be actually default, but
+                                   for 2.6.16 let's do it for ATI only where
+                                   it's really needed. */
                                 case PCI_VENDOR_ID_ATI:
-                                        if (apic_runs_main_timer != 0)
-                                                break;
-#ifdef CONFIG_ACPI
-                                        /* Don't do this for laptops right
-                                           right now because their timer
-                                           doesn't necessarily tick in C2/3 */
-                                        if (acpi_fadt.revision >= 3 &&
-                        (acpi_fadt.plvl2_lat + acpi_fadt.plvl3_lat) < 1100) {
-                                                printk(KERN_INFO
-"ATI board detected, but seems to be a laptop. Timer might be shakey, sorry\n");
-                                                break;
-                                        }
-#endif                                  
+                                        if (timer_over_8254 == 1) {     
+                                                timer_over_8254 = 0;    
                                         printk(KERN_INFO
-             "ATI board detected. Using APIC/PM timer.\n");
-                                        apic_runs_main_timer = 1;
-                                        nohpet = 1;
+                "ATI board detected. Disabling timer routing over 8254.\n");
+                                        }       
                                         return;
                                 } 
 
+
                                 /* No multi-function device? */
                                 type = read_pci_config_byte(num,slot,func,
                                                             PCI_HEADER_TYPE);
@@ -1773,6 +1782,8 @@ static inline void unlock_ExtINT_logic(void)
  * a wide range of boards and BIOS bugs.  Fortunately only the timer IRQ
  * is so screwy.  Thanks to Brian Perkins for testing/hacking this beast
  * fanatically on his truly buggy board.
+ *
+ * FIXME: really need to revamp this for modern platforms only.
  */
 static inline void check_timer(void)
 {
@@ -1795,7 +1806,8 @@ static inline void check_timer(void)
          */
         apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
         init_8259A(1);
-        enable_8259A_irq(0);
+        if (timer_over_8254 > 0)
+                enable_8259A_irq(0);
 
         pin1  = find_isa_irq_pin(0, mp_INT);
         apic1 = find_isa_irq_apic(0, mp_INT);
@@ -1850,7 +1862,7 @@ static inline void check_timer(void)
         }
         printk(" failed.\n");
 
-        if (nmi_watchdog) {
+        if (nmi_watchdog == NMI_IO_APIC) {
                 printk(KERN_WARNING "timer doesn't work through the IO-APIC - disabling NMI Watchdog!\n");
                 nmi_watchdog = 0;
         }
diff --git a/arch/x86_64/kernel/pci-gart.c b/arch/x86_64/kernel/pci-gart.c
index dd0718dc178b..0c3f052ba6ce 100644
--- a/arch/x86_64/kernel/pci-gart.c
+++ b/arch/x86_64/kernel/pci-gart.c
@@ -228,11 +228,6 @@ static inline int need_iommu(struct device *dev, unsigned long addr, size_t size
         int mmu = high;
         if (force_iommu) 
                 mmu = 1; 
-        if (no_iommu) { 
-                if (high) 
-                        panic("PCI-DMA: high address but no IOMMU.\n"); 
-                mmu = 0; 
-        }       
         return mmu; 
 }
 
@@ -241,11 +236,6 @@ static inline int nonforced_iommu(struct device *dev, unsigned long addr, size_t
         u64 mask = *dev->dma_mask;
         int high = addr + size >= mask;
         int mmu = high;
-        if (no_iommu) { 
-                if (high) 
-                        panic("PCI-DMA: high address but no IOMMU.\n"); 
-                mmu = 0; 
-        }       
         return mmu; 
 }
 
@@ -310,7 +300,7 @@ void gart_unmap_sg(struct device *dev, struct scatterlist *sg, int nents, int di
 
         for (i = 0; i < nents; i++) {
                 struct scatterlist *s = &sg[i];
-                if (!s->dma_length)
+                if (!s->dma_length || !s->length)
                         break;
                 dma_unmap_single(dev, s->dma_address, s->dma_length, dir);
         }
@@ -364,6 +354,7 @@ static int __dma_map_cont(struct scatterlist *sg, int start, int stopat,
                 
                 BUG_ON(i > start && s->offset);
                 if (i == start) {
+                        *sout = *s; 
                         sout->dma_address = iommu_bus_base;
                         sout->dma_address += iommu_page*PAGE_SIZE + s->offset;
                         sout->dma_length = s->length;
@@ -390,6 +381,7 @@ static inline int dma_map_cont(struct scatterlist *sg, int start, int stopat,
 {
         if (!need) { 
                 BUG_ON(stopat - start != 1);
+                *sout = sg[start]; 
                 sout->dma_length = sg[start].length; 
                 return 0;
         } 
@@ -632,17 +624,13 @@ static int __init pci_iommu_init(void)
                 (agp_copy_info(agp_bridge, &info) < 0);
 #endif  
 
-        if (swiotlb) { 
-                no_iommu = 1;
+        if (swiotlb)
                 return -1; 
-        } 
-        
+
         if (no_iommu ||
             (!force_iommu && end_pfn <= MAX_DMA32_PFN) ||
             !iommu_aperture ||
             (no_agp && init_k8_gatt(&info) < 0)) {
-                no_iommu = 1;
-                no_iommu_init();
                 printk(KERN_INFO "PCI-DMA: Disabling IOMMU.\n");
                 if (end_pfn > MAX_DMA32_PFN) {
                         printk(KERN_ERR "WARNING more than 4GB of memory "
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 9435ab7d6fb8..aa55e3cec665 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -423,6 +423,12 @@ static __init void parse_cmdline_early (char ** cmdline_p)
                 else if(!memcmp(from, "elfcorehdr=", 11))
                         elfcorehdr_addr = memparse(from+11, &from);
 #endif
+
+#ifdef CONFIG_HOTPLUG_CPU
+                else if (!memcmp(from, "additional_cpus=", 16))
+                        setup_additional_cpus(from+16);
+#endif
+
         next_char:
                 c = *(from++);
                 if (!c)
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index 67e4e28f4df8..66e98659d077 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -1152,8 +1152,6 @@ void __init smp_cpus_done(unsigned int max_cpus)
         setup_ioapic_dest();
 #endif
 
-        time_init_gtod();
-
         check_nmi_watchdog();
 }
 
@@ -1244,7 +1242,7 @@ void __cpu_die(unsigned int cpu)
         printk(KERN_ERR "CPU %u didn't die...\n", cpu);
 }
 
-static __init int setup_additional_cpus(char *s)
+__init int setup_additional_cpus(char *s)
 {
         return get_option(&s, &additional_cpus);
 }
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index 67841d11ed1f..3080f84bf7b7 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -48,6 +48,8 @@ static void cpufreq_delayed_get(void);
 extern void i8254_timer_resume(void);
 extern int using_apic_timer;
 
+static char *time_init_gtod(void);
+
 DEFINE_SPINLOCK(rtc_lock);
 DEFINE_SPINLOCK(i8253_lock);
 
@@ -901,6 +903,7 @@ static struct irqaction irq0 = {
 void __init time_init(void)
 {
         char *timename;
+        char *gtod;
 
 #ifdef HPET_HACK_ENABLE_DANGEROUS
         if (!vxtime.hpet_address) {
@@ -945,21 +948,19 @@ void __init time_init(void)
                 timename = "PIT";
         }
 
-        printk(KERN_INFO "time.c: Using %ld.%06ld MHz %s timer.\n",
-               vxtime_hz / 1000000, vxtime_hz % 1000000, timename);
+        vxtime.mode = VXTIME_TSC;
+        gtod = time_init_gtod();
+
+        printk(KERN_INFO "time.c: Using %ld.%06ld MHz WALL %s GTOD %s timer.\n",
+               vxtime_hz / 1000000, vxtime_hz % 1000000, timename, gtod);
         printk(KERN_INFO "time.c: Detected %d.%03d MHz processor.\n",
                 cpu_khz / 1000, cpu_khz % 1000);
-        vxtime.mode = VXTIME_TSC;
         vxtime.quot = (1000000L << 32) / vxtime_hz;
         vxtime.tsc_quot = (1000L << 32) / cpu_khz;
         vxtime.last_tsc = get_cycles_sync();
         setup_irq(0, &irq0);
 
         set_cyc2ns_scale(cpu_khz);
-
-#ifndef CONFIG_SMP
-        time_init_gtod();
-#endif
 }
 
 /*
@@ -981,9 +982,9 @@ __cpuinit int unsynchronized_tsc(void)
 }
 
 /*
- * Decide after all CPUs are booted what mode gettimeofday should use.
+ * Decide what mode gettimeofday should use.
  */
-void __init time_init_gtod(void)
+__init static char *time_init_gtod(void)
 {
         char *timetype;
 
@@ -1011,8 +1012,7 @@ void __init time_init_gtod(void)
                 timetype = hpet_use_timer ? "HPET/TSC" : "PIT/TSC";
                 vxtime.mode = VXTIME_TSC;
         }
-
-        printk(KERN_INFO "time.c: Using %s based timekeeping.\n", timetype);
+        return timetype;
 }
 
 __setup("report_lost_ticks", time_setup);
diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig
index 0cce28c4025b..33e2ca847a26 100644
--- a/drivers/acpi/Kconfig
+++ b/drivers/acpi/Kconfig
@@ -247,7 +247,7 @@ config ACPI_CUSTOM_DSDT_FILE
           Enter the full path name to the file wich includes the AmlCode declaration.
 
 config ACPI_BLACKLIST_YEAR
-        int "Disable ACPI for systems before Jan 1st this year" if X86
+        int "Disable ACPI for systems before Jan 1st this year" if X86_32
         default 0
         help
           enter a 4-digit year, eg. 2001 to disable ACPI by default
@@ -285,9 +285,8 @@ config ACPI_SYSTEM
           dump your ACPI DSDT table using /proc/acpi/dsdt.
 
 config X86_PM_TIMER
-        bool "Power Management Timer Support"
+        bool "Power Management Timer Support" if EMBEDDED
         depends on X86
-        depends on !X86_64
         default y
         help
           The Power Management Timer is available on all ACPI-capable,
@@ -298,9 +297,8 @@ config X86_PM_TIMER
           voltage scaling, unlike the commonly used Time Stamp Counter
           (TSC) timing source.
 
-          So, if you see messages like 'Losing too many ticks!' in the
-          kernel logs, and/or you are using this on a notebook which
-          does not yet have an HPET, you should say "Y" here.
+          You should nearly always say Y here because many modern
+          systems require this timer. 
 
 config ACPI_CONTAINER
         tristate "ACPI0004,PNP0A05 and PNP0A06 Container Driver (EXPERIMENTAL)"
diff --git a/drivers/char/sx.c b/drivers/char/sx.c
index c2490e270f1f..588e75ec1630 100644
--- a/drivers/char/sx.c
+++ b/drivers/char/sx.c
@@ -2173,15 +2173,17 @@ static int probe_si (struct sx_board *board)
         if ( IS_SI1_BOARD(board)) {
                 /* This should be an SI1 board, which has this
                    location writable... */
-                if (read_sx_byte (board, SI2_ISA_ID_BASE) != 0x10)
+                if (read_sx_byte (board, SI2_ISA_ID_BASE) != 0x10) {
                         func_exit ();
                         return 0; 
+                }
         } else {
                 /* This should be an SI2 board, which has the bottom
                    3 bits non-writable... */
-                if (read_sx_byte (board, SI2_ISA_ID_BASE) == 0x10)
+                if (read_sx_byte (board, SI2_ISA_ID_BASE) == 0x10) {
                         func_exit ();
                         return 0; 
+                }
         }
 
         /* Now we're pretty much convinced that there is an SI board here, 
@@ -2192,15 +2194,17 @@ static int probe_si (struct sx_board *board)
         if ( IS_SI1_BOARD(board)) {
                 /* This should be an SI1 board, which has this
                    location writable... */
-                if (read_sx_byte (board, SI2_ISA_ID_BASE) != 0x10)
+                if (read_sx_byte (board, SI2_ISA_ID_BASE) != 0x10) {
                         func_exit();
                         return 0; 
+                }
         } else {
                 /* This should be an SI2 board, which has the bottom
                    3 bits non-writable... */
-                if (read_sx_byte (board, SI2_ISA_ID_BASE) == 0x10)
+                if (read_sx_byte (board, SI2_ISA_ID_BASE) == 0x10) {
                         func_exit ();
                         return 0; 
+                }
         }
 
         printheader ();
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index 930db398d107..9d9872347f56 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -89,6 +89,11 @@
 #define SD_MAX_RETRIES          5
 #define SD_PASSTHROUGH_RETRIES  1
 
+/*
+ * Size of the initial data buffer for mode and read capacity data
+ */
+#define SD_BUF_SIZE             512
+
 static void scsi_disk_release(struct kref *kref);
 
 struct scsi_disk {
@@ -1239,7 +1244,7 @@ sd_do_mode_sense(struct scsi_device *sdp, int dbd, int modepage,
 
 /*
  * read write protect setting, if possible - called only in sd_revalidate_disk()
- * called with buffer of length 512
+ * called with buffer of length SD_BUF_SIZE
  */
 static void
 sd_read_write_protect_flag(struct scsi_disk *sdkp, char *diskname,
@@ -1297,7 +1302,7 @@ sd_read_write_protect_flag(struct scsi_disk *sdkp, char *diskname,
 
 /*
  * sd_read_cache_type - called only from sd_revalidate_disk()
- * called with buffer of length 512
+ * called with buffer of length SD_BUF_SIZE
  */
 static void
 sd_read_cache_type(struct scsi_disk *sdkp, char *diskname,
@@ -1342,6 +1347,8 @@ sd_read_cache_type(struct scsi_disk *sdkp, char *diskname,
 
         /* Take headers and block descriptors into account */
         len += data.header_length + data.block_descriptor_length;
+        if (len > SD_BUF_SIZE)
+                goto bad_sense;
 
         /* Get the data */
         res = sd_do_mode_sense(sdp, dbd, modepage, buffer, len, &data, &sshdr);
@@ -1354,6 +1361,12 @@ sd_read_cache_type(struct scsi_disk *sdkp, char *diskname,
                 int ct = 0;
                 int offset = data.header_length + data.block_descriptor_length;
 
+                if (offset >= SD_BUF_SIZE - 2) {
+                        printk(KERN_ERR "%s: malformed MODE SENSE response",
+                                diskname);
+                        goto defaults;
+                }
+
                 if ((buffer[offset] & 0x3f) != modepage) {
                         printk(KERN_ERR "%s: got wrong page\n", diskname);
                         goto defaults;
@@ -1398,6 +1411,7 @@ defaults:
                diskname);
         sdkp->WCE = 0;
         sdkp->RCD = 0;
+        sdkp->DPOFUA = 0;
 }
 
 /**
@@ -1421,7 +1435,7 @@ static int sd_revalidate_disk(struct gendisk *disk)
         if (!scsi_device_online(sdp))
                 goto out;
 
-        buffer = kmalloc(512, GFP_KERNEL | __GFP_DMA);
+        buffer = kmalloc(SD_BUF_SIZE, GFP_KERNEL | __GFP_DMA);
         if (!buffer) {
                 printk(KERN_WARNING "(sd_revalidate_disk:) Memory allocation "
                        "failure.\n");
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 1b117a441298..c2eac2a50bd2 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -938,6 +938,11 @@ static int load_elf_binary(struct linux_binprm * bprm, struct pt_regs * regs)
                 kfree(elf_interpreter);
         } else {
                 elf_entry = loc->elf_ex.e_entry;
+                if (BAD_ADDR(elf_entry)) {
+                        send_sig(SIGSEGV, current, 0);
+                        retval = -ENOEXEC; /* Nobody gets to see this, but.. */
+                        goto out_free_dentry;
+                }
         }
 
         kfree(elf_phdata);
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 057e60217fc5..537ac70edfe5 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -2531,18 +2531,9 @@ static int rtc_ioctl(unsigned fd, unsigned cmd, unsigned long arg)
                 val32 = kval;
                 return put_user(val32, (unsigned int __user *)arg);
         case RTC_IRQP_SET32:
+                return sys_ioctl(fd, RTC_IRQP_SET, arg); 
         case RTC_EPOCH_SET32:
-                ret = get_user(val32, (unsigned int __user *)arg);
-                if (ret)
-                        return ret;
-                kval = val32;
-
-                set_fs(KERNEL_DS);
-                ret = sys_ioctl(fd, (cmd == RTC_IRQP_SET32) ?
-                                RTC_IRQP_SET : RTC_EPOCH_SET,
-                                (unsigned long)&kval);
-                set_fs(oldfs);
-                return ret;
+                return sys_ioctl(fd, RTC_EPOCH_SET, arg);
         default:
                 /* unreached */
                 return -ENOIOCTLCMD;
diff --git a/include/asm-sparc64/futex.h b/include/asm-sparc64/futex.h
index 6a332a9f099c..0caf60147e97 100644
--- a/include/asm-sparc64/futex.h
+++ b/include/asm-sparc64/futex.h
@@ -1,6 +1,86 @@
-#ifndef _ASM_FUTEX_H
-#define _ASM_FUTEX_H
+#ifndef _SPARC64_FUTEX_H
+#define _SPARC64_FUTEX_H
 
-#include <asm-generic/futex.h>
+#include <linux/futex.h>
+#include <asm/errno.h>
+#include <asm/system.h>
+#include <asm/uaccess.h>
 
-#endif
+#define __futex_cas_op(insn, ret, oldval, uaddr, oparg) \
+        __asm__ __volatile__(                           \
+        "\n1:   lduwa   [%3] %%asi, %2\n"               \
+        "       " insn "\n"                             \
+        "2:     casa    [%3] %%asi, %2, %1\n"           \
+        "       cmp     %2, %1\n"                       \
+        "       bne,pn  %%icc, 1b\n"                    \
+        "        mov    0, %0\n"                        \
+        "3:\n"                                          \
+        "       .section .fixup,#alloc,#execinstr\n"    \
+        "       .align  4\n"                            \
+        "4:     ba      3b\n"                           \
+        "        mov    %5, %0\n"                       \
+        "       .previous\n"                            \
+        "       .section __ex_table,#alloc\n"           \
+        "       .align  4\n"                            \
+        "       .word   1b, 4b\n"                       \
+        "       .word   2b, 4b\n"                       \
+        "       .previous\n"                            \
+        : "=&r" (ret), "=&r" (oldval), "=&r" (tem)      \
+        : "r" (uaddr), "r" (oparg), "i" (-EFAULT)       \
+        : "memory")
+
+static inline int futex_atomic_op_inuser(int encoded_op, int __user *uaddr)
+{
+        int op = (encoded_op >> 28) & 7;
+        int cmp = (encoded_op >> 24) & 15;
+        int oparg = (encoded_op << 8) >> 20;
+        int cmparg = (encoded_op << 20) >> 20;
+        int oldval = 0, ret, tem;
+
+        if (unlikely(!access_ok(VERIFY_WRITE, uaddr, sizeof(int))))
+                return -EFAULT;
+        if (unlikely((((unsigned long) uaddr) & 0x3UL)))
+                return -EINVAL;
+
+        if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28))
+                oparg = 1 << oparg;
+
+        inc_preempt_count();
+
+        switch (op) {
+        case FUTEX_OP_SET:
+                __futex_cas_op("mov\t%4, %1", ret, oldval, uaddr, oparg);
+                break;
+        case FUTEX_OP_ADD:
+                __futex_cas_op("add\t%2, %4, %1", ret, oldval, uaddr, oparg);
+                break;
+        case FUTEX_OP_OR:
+                __futex_cas_op("or\t%2, %4, %1", ret, oldval, uaddr, oparg);
+                break;
+        case FUTEX_OP_ANDN:
+                __futex_cas_op("and\t%2, %4, %1", ret, oldval, uaddr, oparg);
+                break;
+        case FUTEX_OP_XOR:
+                __futex_cas_op("xor\t%2, %4, %1", ret, oldval, uaddr, oparg);
+                break;
+        default:
+                ret = -ENOSYS;
+        }
+
+        dec_preempt_count();
+
+        if (!ret) {
+                switch (cmp) {
+                case FUTEX_OP_CMP_EQ: ret = (oldval == cmparg); break;
+                case FUTEX_OP_CMP_NE: ret = (oldval != cmparg); break;
+                case FUTEX_OP_CMP_LT: ret = (oldval < cmparg); break;
+                case FUTEX_OP_CMP_GE: ret = (oldval >= cmparg); break;
+                case FUTEX_OP_CMP_LE: ret = (oldval <= cmparg); break;
+                case FUTEX_OP_CMP_GT: ret = (oldval > cmparg); break;
+                default: ret = -ENOSYS;
+                }
+        }
+        return ret;
+}
+
+#endif /* !(_SPARC64_FUTEX_H) */
diff --git a/include/asm-sparc64/smp.h b/include/asm-sparc64/smp.h
index 110a2de89123..473edb2603ec 100644
--- a/include/asm-sparc64/smp.h
+++ b/include/asm-sparc64/smp.h
@@ -66,8 +66,14 @@ static __inline__ int hard_smp_processor_id(void)
 
 #define raw_smp_processor_id() (current_thread_info()->cpu)
 
+extern void smp_setup_cpu_possible_map(void);
+
 #endif /* !(__ASSEMBLY__) */
 
+#else
+
+#define smp_setup_cpu_possible_map() do { } while (0)
+
 #endif /* !(CONFIG_SMP) */
 
 #define NO_PROC_ID              0xFF
diff --git a/include/asm-x86_64/acpi.h b/include/asm-x86_64/acpi.h
index aa1c7b2e438c..e2b9923189a0 100644
--- a/include/asm-x86_64/acpi.h
+++ b/include/asm-x86_64/acpi.h
@@ -164,6 +164,20 @@ extern u8 x86_acpiid_to_apicid[];
 
 extern int acpi_skip_timer_override;
 
+extern int unsync_tsc_on_multicluster;
+
+static inline int acpi_madt_oem_check(char *oem, char *productid) 
+{ 
+        /* Copied from i386. Probably has too many entries. */
+        if (!strncmp(oem, "IBM ENSW", 8) && 
+                (!strncmp(productid, "VIGIL SMP", 9) 
+                        || !strncmp(productid, "EXA", 3)
+                        || !strncmp(productid, "RUTHLESS SMP", 12))) {
+                unsync_tsc_on_multicluster = 1;
+        }
+        return 0;
+}
+
 #endif /*__KERNEL__*/
 
 #endif /*_ASM_ACPI_H*/
diff --git a/include/asm-x86_64/pci.h b/include/asm-x86_64/pci.h
index fd03e15d7ea6..8a05af264d18 100644
--- a/include/asm-x86_64/pci.h
+++ b/include/asm-x86_64/pci.h
@@ -19,8 +19,6 @@ extern unsigned int pcibios_assign_all_busses(void);
 #endif
 #define pcibios_scan_all_fns(a, b)      0
 
-extern int no_iommu, force_iommu;
-
 extern unsigned long pci_mem_start;
 #define PCIBIOS_MIN_IO          0x1000
 #define PCIBIOS_MIN_MEM         (pci_mem_start)
diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h
index 8fbf4dd72115..715fd94cf577 100644
--- a/include/asm-x86_64/pgtable.h
+++ b/include/asm-x86_64/pgtable.h
@@ -131,7 +131,7 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long
 #define PGDIR_SIZE      (1UL << PGDIR_SHIFT)
 #define PGDIR_MASK      (~(PGDIR_SIZE-1))
 
-#define USER_PTRS_PER_PGD       (TASK_SIZE/PGDIR_SIZE)
+#define USER_PTRS_PER_PGD       ((TASK_SIZE-1)/PGDIR_SIZE+1)
 #define FIRST_USER_ADDRESS      0
 
 #ifndef __ASSEMBLY__
diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h
index eca3f2d633db..3ba8fd45fcb3 100644
--- a/include/asm-x86_64/proto.h
+++ b/include/asm-x86_64/proto.h
@@ -39,7 +39,6 @@ extern void config_acpi_tables(void);
 extern void ia32_syscall(void);
 extern void iommu_hole_init(void);
 
-extern void time_init_gtod(void);
 extern int pmtimer_mark_offset(void);
 extern void pmtimer_resume(void);
 extern void pmtimer_wait(unsigned);
@@ -134,6 +133,7 @@ extern int force_iommu;
 
 extern int reboot_force;
 extern int notsc_setup(char *);
+extern int setup_additional_cpus(char *);
 
 extern void smp_local_timer_interrupt(struct pt_regs * regs);