Merge branch 'merge'

12 files changed, 110 insertions, 89 deletions

diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index b8eee4c7888b..e56c2adf57a4 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -17,6 +17,7 @@
 #include <linux/kexec.h>
 #include <linux/module.h>
 
+#include <asm/pgtable.h>
 #include <asm/page.h>
 #include <asm/e820.h>
 #include <asm/proto.h>
diff --git a/arch/x86_64/kernel/early_printk.c b/arch/x86_64/kernel/early_printk.c
index b93ef5b51980..140051e07fa6 100644
--- a/arch/x86_64/kernel/early_printk.c
+++ b/arch/x86_64/kernel/early_printk.c
@@ -2,7 +2,7 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/string.h>
-#include <linux/tty.h>
+#include <linux/screen_info.h>
 #include <asm/io.h>
 #include <asm/processor.h>
 #include <asm/fcntl.h>
diff --git a/arch/x86_64/kernel/machine_kexec.c b/arch/x86_64/kernel/machine_kexec.c
index 83fb24a02821..106076b370fc 100644
--- a/arch/x86_64/kernel/machine_kexec.c
+++ b/arch/x86_64/kernel/machine_kexec.c
@@ -207,14 +207,11 @@ NORET_TYPE void machine_kexec(struct kimage *image)
         __flush_tlb();
 
 
-        /* The segment registers are funny things, they are
-         * automatically loaded from a table, in memory wherever you
-         * set them to a specific selector, but this table is never
-         * accessed again unless you set the segment to a different selector.
-         *
-         * The more common model are caches where the behide
-         * the scenes work is done, but is also dropped at arbitrary
-         * times.
+        /* The segment registers are funny things, they have both a
+         * visible and an invisible part.  Whenever the visible part is
+         * set to a specific selector, the invisible part is loaded
+         * with from a table in memory.  At no other time is the
+         * descriptor table in memory accessed.
          *
          * I take advantage of this here by force loading the
          * segments, before I zap the gdt with an invalid value.
diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c
index 88845674c661..4e017fb30fb3 100644
--- a/arch/x86_64/kernel/mce.c
+++ b/arch/x86_64/kernel/mce.c
@@ -615,7 +615,7 @@ static __cpuinit int mce_create_device(unsigned int cpu)
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
-static __cpuinit void mce_remove_device(unsigned int cpu)
+static void mce_remove_device(unsigned int cpu)
 {
         int i;
 
@@ -626,10 +626,9 @@ static __cpuinit void mce_remove_device(unsigned int cpu)
         sysdev_remove_file(&per_cpu(device_mce,cpu), &attr_check_interval);
         sysdev_unregister(&per_cpu(device_mce,cpu));
 }
-#endif
 
 /* Get notified when a cpu comes on/off. Be hotplug friendly. */
-static __cpuinit int
+static int
 mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
 {
         unsigned int cpu = (unsigned long)hcpu;
@@ -638,18 +637,17 @@ mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
         case CPU_ONLINE:
                 mce_create_device(cpu);
                 break;
-#ifdef CONFIG_HOTPLUG_CPU
         case CPU_DEAD:
                 mce_remove_device(cpu);
                 break;
-#endif
         }
         return NOTIFY_OK;
 }
 
-static struct notifier_block __cpuinitdata mce_cpu_notifier = {
+static struct notifier_block mce_cpu_notifier = {
         .notifier_call = mce_cpu_callback,
 };
+#endif
 
 static __init int mce_init_device(void)
 {
@@ -664,7 +662,7 @@ static __init int mce_init_device(void)
                 mce_create_device(i);
         }
 
-        register_cpu_notifier(&mce_cpu_notifier);
+        register_hotcpu_notifier(&mce_cpu_notifier);
         misc_register(&mce_log_device);
         return err;
 }
diff --git a/arch/x86_64/kernel/mce_amd.c b/arch/x86_64/kernel/mce_amd.c
index 335200aa2737..883fe747f64c 100644
--- a/arch/x86_64/kernel/mce_amd.c
+++ b/arch/x86_64/kernel/mce_amd.c
@@ -558,7 +558,7 @@ out:
  *   of shared sysfs dir/files, and rest of the cores will be symlinked to it.
  */
 
-static __cpuinit void deallocate_threshold_block(unsigned int cpu,
+static void deallocate_threshold_block(unsigned int cpu,
                                                  unsigned int bank)
 {
         struct threshold_block *pos = NULL;
@@ -578,7 +578,7 @@ static __cpuinit void deallocate_threshold_block(unsigned int cpu,
         per_cpu(threshold_banks, cpu)[bank]->blocks = NULL;
 }
 
-static __cpuinit void threshold_remove_bank(unsigned int cpu, int bank)
+static void threshold_remove_bank(unsigned int cpu, int bank)
 {
         int i = 0;
         struct threshold_bank *b;
@@ -597,7 +597,7 @@ static __cpuinit void threshold_remove_bank(unsigned int cpu, int bank)
         /* sibling symlink */
         if (shared_bank[bank] && b->blocks->cpu != cpu) {
                 sysfs_remove_link(&per_cpu(device_mce, cpu).kobj, name);
-                per_cpu(threshold_banks, i)[bank] = NULL;
+                per_cpu(threshold_banks, cpu)[bank] = NULL;
                 return;
         }
 
@@ -618,7 +618,7 @@ free_out:
         per_cpu(threshold_banks, cpu)[bank] = NULL;
 }
 
-static __cpuinit void threshold_remove_device(unsigned int cpu)
+static void threshold_remove_device(unsigned int cpu)
 {
         unsigned int bank;
 
@@ -629,14 +629,8 @@ static __cpuinit void threshold_remove_device(unsigned int cpu)
         }
 }
 
-#else /* !CONFIG_HOTPLUG_CPU */
-static void threshold_remove_device(unsigned int cpu)
-{
-}
-#endif
-
 /* get notified when a cpu comes on/off */
-static int __cpuinit threshold_cpu_callback(struct notifier_block *nfb,
+static int threshold_cpu_callback(struct notifier_block *nfb,
                                             unsigned long action, void *hcpu)
 {
         /* cpu was unsigned int to begin with */
@@ -659,9 +653,10 @@ static int __cpuinit threshold_cpu_callback(struct notifier_block *nfb,
         return NOTIFY_OK;
 }
 
-static struct notifier_block threshold_cpu_notifier __cpuinitdata = {
+static struct notifier_block threshold_cpu_notifier = {
         .notifier_call = threshold_cpu_callback,
 };
+#endif /* CONFIG_HOTPLUG_CPU */
 
 static __init int threshold_init_device(void)
 {
@@ -673,7 +668,7 @@ static __init int threshold_init_device(void)
                 if (err)
                         return err;
         }
-        register_cpu_notifier(&threshold_cpu_notifier);
+        register_hotcpu_notifier(&threshold_cpu_notifier);
         return 0;
 }
 
diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c
index d91cb843f54d..146924ba5df5 100644
--- a/arch/x86_64/kernel/pci-calgary.c
+++ b/arch/x86_64/kernel/pci-calgary.c
@@ -1,9 +1,11 @@
 /*
  * Derived from arch/powerpc/kernel/iommu.c
  *
- * Copyright (C) 2006 Jon Mason <jdmason@us.ibm.com>, IBM Corporation
- * Copyright (C) 2006 Muli Ben-Yehuda <muli@il.ibm.com>, IBM Corporation
+ * Copyright (C) IBM Corporation, 2006
  *
+ * Author: Jon Mason <jdmason@us.ibm.com>
+ * Author: Muli Ben-Yehuda <muli@il.ibm.com>
+
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
@@ -83,7 +85,8 @@
 #define CSR_AGENT_MASK          0xffe0ffff
 
 #define MAX_NUM_OF_PHBS         8 /* how many PHBs in total? */
-#define MAX_PHB_BUS_NUM         (MAX_NUM_OF_PHBS * 2) /* max dev->bus->number */
+#define MAX_NUM_CHASSIS         8 /* max number of chassis */
+#define MAX_PHB_BUS_NUM         (MAX_NUM_OF_PHBS * MAX_NUM_CHASSIS * 2) /* max dev->bus->number */
 #define PHBS_PER_CALGARY        4
 
 /* register offsets in Calgary's internal register space */
@@ -108,7 +111,8 @@ static const unsigned long phb_offsets[] = {
         0xB000 /* PHB3 */
 };
 
-void* tce_table_kva[MAX_NUM_OF_PHBS * MAX_NUMNODES];
+static char bus_to_phb[MAX_PHB_BUS_NUM];
+void* tce_table_kva[MAX_PHB_BUS_NUM];
 unsigned int specified_table_size = TCE_TABLE_SIZE_UNSPECIFIED;
 static int translate_empty_slots __read_mostly = 0;
 static int calgary_detected __read_mostly = 0;
@@ -117,7 +121,7 @@ static int calgary_detected __read_mostly = 0;
  * the bitmap of PHBs the user requested that we disable
  * translation on.
  */
-static DECLARE_BITMAP(translation_disabled, MAX_NUMNODES * MAX_PHB_BUS_NUM);
+static DECLARE_BITMAP(translation_disabled, MAX_PHB_BUS_NUM);
 
 static void tce_cache_blast(struct iommu_table *tbl);
 
@@ -450,7 +454,7 @@ static struct dma_mapping_ops calgary_dma_ops = {
 
 static inline int busno_to_phbid(unsigned char num)
 {
-        return bus_to_phb(num) % PHBS_PER_CALGARY;
+        return bus_to_phb[num];
 }
 
 static inline unsigned long split_queue_offset(unsigned char num)
@@ -810,7 +814,7 @@ static int __init calgary_init(void)
         int i, ret = -ENODEV;
         struct pci_dev *dev = NULL;
 
-        for (i = 0; i <= num_online_nodes() * MAX_NUM_OF_PHBS; i++) {
+        for (i = 0; i < MAX_PHB_BUS_NUM; i++) {
                 dev = pci_get_device(PCI_VENDOR_ID_IBM,
                                      PCI_DEVICE_ID_IBM_CALGARY,
                                      dev);
@@ -820,7 +824,7 @@ static int __init calgary_init(void)
                         calgary_init_one_nontraslated(dev);
                         continue;
                 }
-                if (!tce_table_kva[i] && !translate_empty_slots) {
+                if (!tce_table_kva[dev->bus->number] && !translate_empty_slots) {
                         pci_dev_put(dev);
                         continue;
                 }
@@ -840,7 +844,7 @@ error:
                         pci_dev_put(dev);
                         continue;
                 }
-                if (!tce_table_kva[i] && !translate_empty_slots)
+                if (!tce_table_kva[dev->bus->number] && !translate_empty_slots)
                         continue;
                 calgary_disable_translation(dev);
                 calgary_free_tar(dev);
@@ -874,9 +878,10 @@ static inline int __init determine_tce_table_size(u64 ram)
 void __init detect_calgary(void)
 {
         u32 val;
-        int bus, table_idx;
+        int bus;
         void *tbl;
-        int detected = 0;
+        int calgary_found = 0;
+        int phb = -1;
 
         /*
          * if the user specified iommu=off or iommu=soft or we found
@@ -887,38 +892,46 @@ void __init detect_calgary(void)
 
         specified_table_size = determine_tce_table_size(end_pfn * PAGE_SIZE);
 
-        for (bus = 0, table_idx = 0;
-             bus <= num_online_nodes() * MAX_PHB_BUS_NUM;
-             bus++) {
-                BUG_ON(bus > MAX_NUMNODES * MAX_PHB_BUS_NUM);
+        for (bus = 0; bus < MAX_PHB_BUS_NUM; bus++) {
+                int dev;
+
+                tce_table_kva[bus] = NULL;
+                bus_to_phb[bus] = -1;
+
                 if (read_pci_config(bus, 0, 0, 0) != PCI_VENDOR_DEVICE_ID_CALGARY)
                         continue;
+
+                /*
+                 * There are 4 PHBs per Calgary chip.  Set phb to which phb (0-3)
+                 * it is connected to releative to the clagary chip.
+                 */
+                phb = (phb + 1) % PHBS_PER_CALGARY;
+
                 if (test_bit(bus, translation_disabled)) {
                         printk(KERN_INFO "Calgary: translation is disabled for "
                                "PHB 0x%x\n", bus);
                         /* skip this phb, don't allocate a tbl for it */
-                        tce_table_kva[table_idx] = NULL;
-                        table_idx++;
                         continue;
                 }
                 /*
-                 * scan the first slot of the PCI bus to see if there
-                 * are any devices present
+                 * Scan the slots of the PCI bus to see if there is a device present.
+                 * The parent bus will be the zero-ith device, so start at 1.
                  */
-                val = read_pci_config(bus, 1, 0, 0);
-                if (val != 0xffffffff || translate_empty_slots) {
-                        tbl = alloc_tce_table();
-                        if (!tbl)
-                                goto cleanup;
-                        detected = 1;
-                } else
-                        tbl = NULL;
-
-                tce_table_kva[table_idx] = tbl;
-                table_idx++;
+                for (dev = 1; dev < 8; dev++) {
+                        val = read_pci_config(bus, dev, 0, 0);
+                        if (val != 0xffffffff || translate_empty_slots) {
+                                tbl = alloc_tce_table();
+                                if (!tbl)
+                                        goto cleanup;
+                                tce_table_kva[bus] = tbl;
+                                bus_to_phb[bus] = phb;
+                                calgary_found = 1;
+                                break;
+                        }
+                }
         }
 
-        if (detected) {
+        if (calgary_found) {
                 iommu_detected = 1;
                 calgary_detected = 1;
                 printk(KERN_INFO "PCI-DMA: Calgary IOMMU detected. "
@@ -927,9 +940,9 @@ void __init detect_calgary(void)
         return;
 
 cleanup:
-        for (--table_idx; table_idx >= 0; --table_idx)
-                if (tce_table_kva[table_idx])
-                        free_tce_table(tce_table_kva[table_idx]);
+        for (--bus; bus >= 0; --bus)
+                if (tce_table_kva[bus])
+                        free_tce_table(tce_table_kva[bus]);
 }
 
 int __init calgary_iommu_init(void)
@@ -1000,7 +1013,7 @@ static int __init calgary_parse_options(char *p)
                         if (p == endp)
                                 break;
 
-                        if (bridge <= (num_online_nodes() * MAX_PHB_BUS_NUM)) {
+                        if (bridge < MAX_PHB_BUS_NUM) {
                                 printk(KERN_INFO "Calgary: disabling "
                                        "translation for PHB 0x%x\n", bridge);
                                 set_bit(bridge, translation_disabled);
diff --git a/arch/x86_64/kernel/pci-swiotlb.c b/arch/x86_64/kernel/pci-swiotlb.c
index ebdb77fe2057..6a55f87ba97f 100644
--- a/arch/x86_64/kernel/pci-swiotlb.c
+++ b/arch/x86_64/kernel/pci-swiotlb.c
@@ -31,9 +31,10 @@ struct dma_mapping_ops swiotlb_dma_ops = {
 void pci_swiotlb_init(void)
 {
         /* don't initialize swiotlb if iommu=off (no_iommu=1) */
-        if (!iommu_detected && !no_iommu &&
-            (end_pfn > MAX_DMA32_PFN || force_iommu))
+        if (!iommu_detected && !no_iommu && end_pfn > MAX_DMA32_PFN)
                swiotlb = 1;
+        if (swiotlb_force)
+                swiotlb = 1;
         if (swiotlb) {
                 printk(KERN_INFO "PCI-DMA: Using software bounce buffering for IO (SWIOTLB)\n");
                 swiotlb_init();
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 0925518b58d0..8a099ff1f8bc 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -21,7 +21,7 @@
 #include <linux/slab.h>
 #include <linux/user.h>
 #include <linux/a.out.h>
-#include <linux/tty.h>
+#include <linux/screen_info.h>
 #include <linux/ioport.h>
 #include <linux/delay.h>
 #include <linux/init.h>
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index b7c705969791..975380207b46 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -771,12 +771,10 @@ static int __cpuinit do_boot_cpu(int cpu, int apicid)
         unsigned long start_rip;
         struct create_idle c_idle = {
                 .cpu = cpu,
-                .done = COMPLETION_INITIALIZER(c_idle.done),
+                .done = COMPLETION_INITIALIZER_ONSTACK(c_idle.done),
         };
         DECLARE_WORK(work, do_fork_idle, &c_idle);
 
-        lockdep_set_class(&c_idle.done.wait.lock, &waitqueue_lock_key);
-
         /* allocate memory for gdts of secondary cpus. Hotplug is considered */
         if (!cpu_gdt_descr[cpu].address &&
                 !(cpu_gdt_descr[cpu].address = get_zeroed_page(GFP_KERNEL))) {
diff --git a/arch/x86_64/kernel/tce.c b/arch/x86_64/kernel/tce.c
index 8d4c67f61b8e..5530dda3f27a 100644
--- a/arch/x86_64/kernel/tce.c
+++ b/arch/x86_64/kernel/tce.c
@@ -1,8 +1,10 @@
 /*
  * Derived from arch/powerpc/platforms/pseries/iommu.c
  *
- * Copyright (C) 2006 Jon Mason <jdmason@us.ibm.com>, IBM Corporation
- * Copyright (C) 2006 Muli Ben-Yehuda <muli@il.ibm.com>, IBM Corporation
+ * Copyright (C) IBM Corporation, 2006
+ *
+ * Author: Jon Mason <jdmason@us.ibm.com>
+ * Author: Muli Ben-Yehuda <muli@il.ibm.com>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -94,7 +96,6 @@ static inline unsigned int table_size_to_number_of_entries(unsigned char size)
 static int tce_table_setparms(struct pci_dev *dev, struct iommu_table *tbl)
 {
         unsigned int bitmapsz;
-        unsigned int tce_table_index;
         unsigned long bmppages;
         int ret;
 
@@ -103,8 +104,7 @@ static int tce_table_setparms(struct pci_dev *dev, struct iommu_table *tbl)
         /* set the tce table size - measured in entries */
         tbl->it_size = table_size_to_number_of_entries(specified_table_size);
 
-        tce_table_index = bus_to_phb(tbl->it_busno);
-        tbl->it_base = (unsigned long)tce_table_kva[tce_table_index];
+        tbl->it_base = (unsigned long)tce_table_kva[dev->bus->number];
         if (!tbl->it_base) {
                 printk(KERN_ERR "Calgary: iommu_table_setparms: "
                        "no table allocated?!\n");
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index b9ff75992c16..7a9b18224182 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -28,6 +28,7 @@
 #include <linux/acpi.h>
 #ifdef CONFIG_ACPI
 #include <acpi/achware.h>       /* for PM timer frequency */
+#include <acpi/acpi_bus.h>
 #endif
 #include <asm/8253pit.h>
 #include <asm/pgtable.h>
@@ -193,7 +194,7 @@ unsigned long profile_pc(struct pt_regs *regs)
            is just accounted to the spinlock function.
            Better would be to write these functions in assembler again
            and check exactly. */
-        if (in_lock_functions(pc)) {
+        if (!user_mode(regs) && in_lock_functions(pc)) {
                 char *v = *(char **)regs->rsp;
                 if ((v >= _stext && v <= _etext) ||
                         (v >= _sinittext && v <= _einittext) ||
@@ -953,11 +954,18 @@ __cpuinit int unsynchronized_tsc(void)
 #ifdef CONFIG_SMP
         if (apic_is_clustered_box())
                 return 1;
-        /* Intel systems are normally all synchronized. Exceptions
-           are handled in the check above. */
-        if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL)
-                return 0;
 #endif
+        /* Most intel systems have synchronized TSCs except for
+           multi node systems */
+        if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) {
+#ifdef CONFIG_ACPI
+                /* But TSC doesn't tick in C3 so don't use it there */
+                if (acpi_fadt.length > 0 && acpi_fadt.plvl3_lat < 100)
+                        return 1;
+#endif
+                return 0;
+        }
+
         /* Assume multi socket systems are not synchronized */
         return num_present_cpus() > 1;
 }
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 79d05c482072..4e9938dee060 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -76,13 +76,13 @@ int register_die_notifier(struct notifier_block *nb)
         vmalloc_sync_all();
         return atomic_notifier_chain_register(&die_chain, nb);
 }
-EXPORT_SYMBOL(register_die_notifier);
+EXPORT_SYMBOL(register_die_notifier); /* used modular by kdb */
 
 int unregister_die_notifier(struct notifier_block *nb)
 {
         return atomic_notifier_chain_unregister(&die_chain, nb);
 }
-EXPORT_SYMBOL(unregister_die_notifier);
+EXPORT_SYMBOL(unregister_die_notifier); /* used modular by kdb */
 
 static inline void conditional_sti(struct pt_regs *regs)
 {
@@ -254,7 +254,6 @@ void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s
 {
         const unsigned cpu = safe_smp_processor_id();
         unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr;
-        int i = 11;
         unsigned used = 0;
 
         printk("\nCall Trace:\n");
@@ -275,11 +274,20 @@ void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s
                         if (unwind_init_blocked(&info, tsk) == 0)
                                 unw_ret = show_trace_unwind(&info, NULL);
                 }
-                if (unw_ret > 0) {
-                        if (call_trace > 0)
+                if (unw_ret > 0 && !arch_unw_user_mode(&info)) {
+#ifdef CONFIG_STACK_UNWIND
+                        unsigned long rip = info.regs.rip;
+                        print_symbol("DWARF2 unwinder stuck at %s\n", rip);
+                        if (call_trace == 1) {
+                                printk("Leftover inexact backtrace:\n");
+                                stack = (unsigned long *)info.regs.rsp;
+                        } else if (call_trace > 1)
                                 return;
-                        printk("Legacy call trace:");
-                        i = 18;
+                        else
+                                printk("Full inexact backtrace again:\n");
+#else
+                        printk("Inexact backtrace:\n");
+#endif
                 }
         }
 
@@ -521,7 +529,7 @@ void __kprobes oops_end(unsigned long flags)
                 /* Nest count reaches zero, release the lock. */
                 spin_unlock_irqrestore(&die_lock, flags);
         if (panic_on_oops)
-                panic("Oops");
+                panic("Fatal exception: panic_on_oops");
 }
 
 void __kprobes __die(const char * str, struct pt_regs * regs, long err)
@@ -1118,8 +1126,10 @@ static int __init call_trace_setup(char *s)
                 call_trace = -1;
         else if (strcmp(s, "both") == 0)
                 call_trace = 0;
-        else if (strcmp(s, "new") == 0)
+        else if (strcmp(s, "newfallback") == 0)
                 call_trace = 1;
+        else if (strcmp(s, "new") == 0)
+                call_trace = 2;
         return 1;
 }
 __setup("call_trace=", call_trace_setup);