9 files changed, 98 insertions, 24 deletions

diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index 62776c07cff1..222b5b46d2b2 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -76,6 +76,12 @@ static inline int bad_addr(unsigned long *addrp, unsigned long size)
                 *addrp = __pa_symbol(&_end);
                 return 1;
         }
+
+        if (last >= ebda_addr && addr < ebda_addr + ebda_size) {
+                *addrp = ebda_addr + ebda_size;
+                return 1;
+        }
+
         /* XXX ramdisk image here? */ 
         return 0;
 } 
diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index 77b4c608cca0..0de3ea938830 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -1777,6 +1777,8 @@ static inline void unlock_ExtINT_logic(void)
         spin_unlock_irqrestore(&ioapic_lock, flags);
 }
 
+int timer_uses_ioapic_pin_0;
+
 /*
  * This code may look a bit paranoid, but it's supposed to cooperate with
  * a wide range of boards and BIOS bugs.  Fortunately only the timer IRQ
@@ -1814,6 +1816,9 @@ static inline void check_timer(void)
         pin2  = ioapic_i8259.pin;
         apic2 = ioapic_i8259.apic;
 
+        if (pin1 == 0)
+                timer_uses_ioapic_pin_0 = 1;
+
         apic_printk(APIC_VERBOSE,KERN_INFO "..TIMER: vector=0x%02X apic1=%d pin1=%d apic2=%d pin2=%d\n",
                 vector, apic1, pin1, apic2, pin2);
 
diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c
index 1eaa5dae6174..fa1d19ca700a 100644
--- a/arch/x86_64/kernel/kprobes.c
+++ b/arch/x86_64/kernel/kprobes.c
@@ -514,13 +514,13 @@ static void __kprobes resume_execution(struct kprobe *p,
                 *tos = orig_rip + (*tos - copy_rip);
                 break;
         case 0xff:
-                if ((*insn & 0x30) == 0x10) {
+                if ((insn[1] & 0x30) == 0x10) {
                         /* call absolute, indirect */
                         /* Fix return addr; rip is correct. */
                         next_rip = regs->rip;
                         *tos = orig_rip + (*tos - copy_rip);
-                } else if (((*insn & 0x31) == 0x20) ||  /* jmp near, absolute indirect */
-                           ((*insn & 0x31) == 0x21)) {  /* jmp far, absolute indirect */
+                } else if (((insn[1] & 0x31) == 0x20) ||        /* jmp near, absolute indirect */
+                           ((insn[1] & 0x31) == 0x21)) {        /* jmp far, absolute indirect */
                         /* rip is correct. */
                         next_rip = regs->rip;
                 }
diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index b17cf3eba359..083da7e606b1 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -968,7 +968,17 @@ int mp_register_gsi(u32 gsi, int triggering, int polarity)
                  */
                 int irq = gsi;
                 if (gsi < MAX_GSI_NUM) {
-                        if (gsi > 15)
+                        /*
+                         * Retain the VIA chipset work-around (gsi > 15), but
+                         * avoid a problem where the 8254 timer (IRQ0) is setup
+                         * via an override (so it's not on pin 0 of the ioapic),
+                         * and at the same time, the pin 0 interrupt is a PCI
+                         * type.  The gsi > 15 test could cause these two pins
+                         * to be shared as IRQ0, and they are not shareable.
+                         * So test for this condition, and if necessary, avoid
+                         * the pin collision.
+                         */
+                        if (gsi > 15 || (gsi == 0 && !timer_uses_ioapic_pin_0))
                                 gsi = pci_irq++;
                         /*
                          * Don't assign IRQ used by ACPI SCI
diff --git a/arch/x86_64/kernel/pci-gart.c b/arch/x86_64/kernel/pci-gart.c
index 9d3d76c85ae7..2480d3f08a47 100644
--- a/arch/x86_64/kernel/pci-gart.c
+++ b/arch/x86_64/kernel/pci-gart.c
@@ -639,6 +639,14 @@ static int __init pci_iommu_init(void)
                 return -1;
         }
 
+        i = 0;
+        for_all_nb(dev)
+                i++;
+        if (i > MAX_NB) {
+                printk(KERN_ERR "PCI-GART: Too many northbridges (%ld). Disabled\n", i);
+                return -1;
+        }
+
         printk(KERN_INFO "PCI-DMA: using GART IOMMU.\n");
         aper_size = info.aper_size * 1024 * 1024;       
         iommu_size = check_iommu_size(info.aper_base, aper_size); 
diff --git a/arch/x86_64/kernel/pci-nommu.c b/arch/x86_64/kernel/pci-nommu.c
index 44adcc2d5e5b..1f6ecc62061d 100644
--- a/arch/x86_64/kernel/pci-nommu.c
+++ b/arch/x86_64/kernel/pci-nommu.c
@@ -12,9 +12,10 @@ static int
 check_addr(char *name, struct device *hwdev, dma_addr_t bus, size_t size)
 {
         if (hwdev && bus + size > *hwdev->dma_mask) {
-                printk(KERN_ERR
-                    "nommu_%s: overflow %Lx+%lu of device mask %Lx\n",
-               name, (long long)bus, size, (long long)*hwdev->dma_mask);
+                if (*hwdev->dma_mask >= 0xffffffffULL)
+                        printk(KERN_ERR
+                            "nommu_%s: overflow %Lx+%lu of device mask %Lx\n",
+                                name, (long long)bus, size, (long long)*hwdev->dma_mask);
                 return 0;
         }
         return 1;
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index ebc3c33b1c6c..f0870bef24d1 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -571,17 +571,28 @@ static inline void copy_edd(void)
 #endif
 
 #define EBDA_ADDR_POINTER 0x40E
-static void __init reserve_ebda_region(void)
+
+unsigned __initdata ebda_addr;
+unsigned __initdata ebda_size;
+
+static void discover_ebda(void)
 {
-        unsigned int addr;
-        /** 
+        /*
          * there is a real-mode segmented pointer pointing to the 
          * 4K EBDA area at 0x40E
          */
-        addr = *(unsigned short *)phys_to_virt(EBDA_ADDR_POINTER);
-        addr <<= 4;
-        if (addr)
-                reserve_bootmem_generic(addr, PAGE_SIZE);
+        ebda_addr = *(unsigned short *)EBDA_ADDR_POINTER;
+        ebda_addr <<= 4;
+
+        ebda_size = *(unsigned short *)(unsigned long)ebda_addr;
+
+        /* Round EBDA up to pages */
+        if (ebda_size == 0)
+                ebda_size = 1;
+        ebda_size <<= 10;
+        ebda_size = round_up(ebda_size + (ebda_addr & ~PAGE_MASK), PAGE_SIZE);
+        if (ebda_size > 64*1024)
+                ebda_size = 64*1024;
 }
 
 void __init setup_arch(char **cmdline_p)
@@ -627,6 +638,8 @@ void __init setup_arch(char **cmdline_p)
 
         check_efer();
 
+        discover_ebda();
+
         init_memory_mapping(0, (end_pfn_map << PAGE_SHIFT));
 
         dmi_scan_machine();
@@ -669,7 +682,8 @@ void __init setup_arch(char **cmdline_p)
         reserve_bootmem_generic(0, PAGE_SIZE);
 
         /* reserve ebda region */
-        reserve_ebda_region();
+        if (ebda_addr)
+                reserve_bootmem_generic(ebda_addr, ebda_size);
 
 #ifdef CONFIG_SMP
         /*
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 2700b1375c1f..cea335e8746c 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -102,6 +102,8 @@ static inline void preempt_conditional_cli(struct pt_regs *regs)
 {
         if (regs->eflags & X86_EFLAGS_IF)
                 local_irq_disable();
+        /* Make sure to not schedule here because we could be running
+           on an exception stack. */
         preempt_enable_no_resched();
 }
 
@@ -385,6 +387,7 @@ void out_of_line_bug(void)
 
 static DEFINE_SPINLOCK(die_lock);
 static int die_owner = -1;
+static unsigned int die_nest_count;
 
 unsigned __kprobes long oops_begin(void)
 {
@@ -399,6 +402,7 @@ unsigned __kprobes long oops_begin(void)
                 else
                         spin_lock(&die_lock);
         }
+        die_nest_count++;
         die_owner = cpu;
         console_verbose();
         bust_spinlocks(1);
@@ -409,7 +413,13 @@ void __kprobes oops_end(unsigned long flags)
 { 
         die_owner = -1;
         bust_spinlocks(0);
-        spin_unlock_irqrestore(&die_lock, flags);
+        die_nest_count--;
+        if (die_nest_count)
+                /* We still own the lock */
+                local_irq_restore(flags);
+        else
+                /* Nest count reaches zero, release the lock. */
+                spin_unlock_irqrestore(&die_lock, flags);
         if (panic_on_oops)
                 panic("Oops");
 }
@@ -464,6 +474,8 @@ void __kprobes die_nmi(char *str, struct pt_regs *regs)
                 panic("nmi watchdog");
         printk("console shuts up ...\n");
         oops_end(flags);
+        nmi_exit();
+        local_irq_enable();
         do_exit(SIGSEGV);
 }
 
@@ -473,8 +485,6 @@ static void __kprobes do_trap(int trapnr, int signr, char *str,
 {
         struct task_struct *tsk = current;
 
-        conditional_sti(regs);
-
         tsk->thread.error_code = error_code;
         tsk->thread.trap_no = trapnr;
 
@@ -511,6 +521,7 @@ asmlinkage void do_##name(struct pt_regs * regs, long error_code) \
         if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
                                                         == NOTIFY_STOP) \
                 return; \
+        conditional_sti(regs);                                          \
         do_trap(trapnr, signr, str, regs, error_code, NULL); \
 }
 
@@ -525,6 +536,7 @@ asmlinkage void do_##name(struct pt_regs * regs, long error_code) \
         if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
                                                         == NOTIFY_STOP) \
                 return; \
+        conditional_sti(regs);                                          \
         do_trap(trapnr, signr, str, regs, error_code, &info); \
 }
 
@@ -538,7 +550,17 @@ DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS)
 DO_ERROR(11, SIGBUS,  "segment not present", segment_not_present)
 DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0)
 DO_ERROR(18, SIGSEGV, "reserved", reserved)
-DO_ERROR(12, SIGBUS,  "stack segment", stack_segment)
+
+/* Runs on IST stack */
+asmlinkage void do_stack_segment(struct pt_regs *regs, long error_code)
+{
+        if (notify_die(DIE_TRAP, "stack segment", regs, error_code,
+                        12, SIGBUS) == NOTIFY_STOP)
+                return;
+        preempt_conditional_sti(regs);
+        do_trap(12, SIGBUS, "stack segment", regs, error_code, NULL);
+        preempt_conditional_cli(regs);
+}
 
 asmlinkage void do_double_fault(struct pt_regs * regs, long error_code)
 {
@@ -672,8 +694,9 @@ asmlinkage void __kprobes do_int3(struct pt_regs * regs, long error_code)
         if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP) == NOTIFY_STOP) {
                 return;
         }
+        preempt_conditional_sti(regs);
         do_trap(3, SIGTRAP, "int3", regs, error_code, NULL);
-        return;
+        preempt_conditional_cli(regs);
 }
 
 /* Help handler running on IST stack to switch back to user stack
diff --git a/arch/x86_64/mm/srat.c b/arch/x86_64/mm/srat.c
index 15ae9fcd65a7..e1513532df29 100644
--- a/arch/x86_64/mm/srat.c
+++ b/arch/x86_64/mm/srat.c
@@ -34,7 +34,10 @@ static nodemask_t nodes_found __initdata;
 static struct bootnode nodes[MAX_NUMNODES] __initdata;
 static struct bootnode nodes_add[MAX_NUMNODES] __initdata;
 static int found_add_area __initdata;
-int hotadd_percent __initdata = 10;
+int hotadd_percent __initdata = 0;
+#ifndef RESERVE_HOTADD
+#define hotadd_percent 0        /* Ignore all settings */
+#endif
 static u8 pxm2node[256] = { [0 ... 255] = 0xff };
 
 /* Too small nodes confuse the VM badly. Usually they result
@@ -103,6 +106,7 @@ static __init void bad_srat(void)
         int i;
         printk(KERN_ERR "SRAT: SRAT not used.\n");
         acpi_numa = -1;
+        found_add_area = 0;
         for (i = 0; i < MAX_LOCAL_APIC; i++)
                 apicid_to_node[i] = NUMA_NO_NODE;
         for (i = 0; i < MAX_NUMNODES; i++)
@@ -154,7 +158,8 @@ acpi_numa_processor_affinity_init(struct acpi_table_processor_affinity *pa)
         int pxm, node;
         if (srat_disabled())
                 return;
-        if (pa->header.length != sizeof(struct acpi_table_processor_affinity)) {                bad_srat();
+        if (pa->header.length != sizeof(struct acpi_table_processor_affinity)) {
+                bad_srat();
                 return;
         }
         if (pa->flags.enabled == 0)
@@ -191,15 +196,17 @@ static int hotadd_enough_memory(struct bootnode *nd)
         allowed = (end_pfn - e820_hole_size(0, end_pfn)) * PAGE_SIZE;
         allowed = (allowed / 100) * hotadd_percent;
         if (allocated + mem > allowed) {
+                unsigned long range;
                 /* Give them at least part of their hotadd memory upto hotadd_percent
                    It would be better to spread the limit out
                    over multiple hotplug areas, but that is too complicated
                    right now */
                 if (allocated >= allowed)
                         return 0;
-                pages = (allowed - allocated + mem) / sizeof(struct page);
+                range = allowed - allocated;
+                pages = (range / PAGE_SIZE);
                 mem = pages * sizeof(struct page);
-                nd->end = nd->start + pages*PAGE_SIZE;
+                nd->end = nd->start + range;
         }
         /* Not completely fool proof, but a good sanity check */
         addr = find_e820_area(last_area_end, end_pfn<<PAGE_SHIFT, mem);