49 files changed, 646 insertions, 499 deletions
diff --git a/Documentation/DocBook/kgdb.tmpl b/Documentation/DocBook/kgdb.tmpl
index 028a8444d95e..e8acd1f03456 100644
--- a/Documentation/DocBook/kgdb.tmpl
+++ b/Documentation/DocBook/kgdb.tmpl
@@ -84,10 +84,9 @@
    runs an instance of gdb against the vmlinux file which contains
    the symbols (not boot image such as bzImage, zImage, uImage...).
    In gdb the developer specifies the connection parameters and
-    connects to kgdb.  Depending on which kgdb I/O modules exist in
+    connects to kgdb.  The type of connection a developer makes with
-    the kernel for a given architecture, it may be possible to debug
+    gdb depends on the availability of kgdb I/O modules compiled as
-    the test machine's kernel with the development machine using a
+    builtin's or kernel modules in the test machine's kernel.
-    rs232 or ethernet connection.
    </para>
  </chapter>
  <chapter id="CompilingAKernel">
@@ -223,7 +222,7 @@
  </para>
  <para>
  IMPORTANT NOTE: Using this option with kgdb over the console
-  (kgdboc) or kgdb over ethernet (kgdboe) is not supported.
+  (kgdboc) is not supported.
  </para>
  </sect1>
  </chapter>
@@ -249,18 +248,11 @@
    (gdb) target remote /dev/ttyS0
    </programlisting>
    <para>
-    Example (kgdb to a terminal server):
+    Example (kgdb to a terminal server on tcp port 2012):
    </para>
    <programlisting>
    % gdb ./vmlinux
-    (gdb) target remote udp:192.168.2.2:6443
+    (gdb) target remote 192.168.2.2:2012
-    </programlisting>
-    <para>
-    Example (kgdb over ethernet):
-    </para>
-    <programlisting>
-    % gdb ./vmlinux
-    (gdb) target remote udp:192.168.2.2:6443
    </programlisting>
    <para>
    Once connected, you can debug a kernel the way you would debug an
diff --git a/Makefile b/Makefile
index ab73e496c8ee..ff2c681fb5dd 100644
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,7 @@
 VERSION = 2
 PATCHLEVEL = 6
 SUBLEVEL = 26
-EXTRAVERSION = -rc7
+EXTRAVERSION = -rc8
 NAME = Rotary Wombat
 # *DOCUMENTATION*
diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c
index 082c31dcfd99..39752cdef6ff 100644
--- a/arch/ia64/kernel/iosapic.c
+++ b/arch/ia64/kernel/iosapic.c
@@ -558,8 +558,6 @@ static struct iosapic_rte_info * __init_refok iosapic_alloc_rte (void)
        if (!iosapic_kmalloc_ok && list_empty(&free_rte_list)) {
                rte = alloc_bootmem(sizeof(struct iosapic_rte_info) *
                                    NR_PREALLOCATE_RTE_ENTRIES);
-                if (!rte)
-                        return NULL;
                for (i = 0; i < NR_PREALLOCATE_RTE_ENTRIES; i++, rte++)
                        list_add(&rte->rte_list, &free_rte_list);
        }
diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index f48a809c686d..4ae15c8c2488 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -578,8 +578,6 @@ setup_arch (char **cmdline_p)
        cpu_init();     /* initialize the bootstrap CPU */
        mmu_context_init();     /* initialize context_id bitmap */
-        check_sal_cache_flush();
 #ifdef CONFIG_ACPI
        acpi_boot_init();
 #endif
@@ -607,6 +605,7 @@ setup_arch (char **cmdline_p)
                ia64_mca_init();
        platform_setup(cmdline_p);
+        check_sal_cache_flush();
        paging_init();
 }
diff --git a/arch/ia64/sn/kernel/sn2/sn2_smp.c b/arch/ia64/sn/kernel/sn2/sn2_smp.c
index 6dd886c5d860..e585f9a2afb9 100644
--- a/arch/ia64/sn/kernel/sn2/sn2_smp.c
+++ b/arch/ia64/sn/kernel/sn2/sn2_smp.c
@@ -512,7 +512,7 @@ static ssize_t sn2_ptc_proc_write(struct file *file, const char __user *user, si
        int cpu;
        char optstr[64];
-        if (count > sizeof(optstr))
+        if (count == 0 || count > sizeof(optstr))
                return -EINVAL;
        if (copy_from_user(optstr, user, count))
                return -EFAULT;
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index b0937c03af3c..400135148555 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -385,6 +385,7 @@ config VMI
 config KVM_CLOCK
        bool "KVM paravirtualized clock"
        select PARAVIRT
+        select PARAVIRT_CLOCK
        depends on !(X86_VISWS || X86_VOYAGER)
        help
          Turning on this option will allow you to run a paravirtualized clock
@@ -412,6 +413,10 @@ config PARAVIRT
          over full virtualization.  However, when run without a hypervisor
          the kernel is theoretically slower and slightly larger.
+config PARAVIRT_CLOCK
+        bool
+        default n
 endif
 config MEMTEST_BOOTPARAM
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 739d49acd2f1..5ff67208d4ae 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -90,6 +90,7 @@ obj-$(CONFIG_VMI)		+= vmi_32.o vmiclock_32.o
 obj-$(CONFIG_KVM_GUEST)         += kvm.o
 obj-$(CONFIG_KVM_CLOCK)         += kvmclock.o
 obj-$(CONFIG_PARAVIRT)          += paravirt.o paravirt_patch_$(BITS).o
+obj-$(CONFIG_PARAVIRT_CLOCK)    += pvclock.o
 obj-$(CONFIG_PCSPKR_PLATFORM)   += pcspeaker.o
diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c
index 08a30986d472..87edf1ceb1df 100644
--- a/arch/x86/kernel/kvmclock.c
+++ b/arch/x86/kernel/kvmclock.c
@@ -18,6 +18,7 @@
 #include <linux/clocksource.h>
 #include <linux/kvm_para.h>
+#include <asm/pvclock.h>
 #include <asm/arch_hooks.h>
 #include <asm/msr.h>
 #include <asm/apic.h>
@@ -36,18 +37,9 @@ static int parse_no_kvmclock(char *arg)
 early_param("no-kvmclock", parse_no_kvmclock);
 /* The hypervisor will put information about time periodically here */
-static DEFINE_PER_CPU_SHARED_ALIGNED(struct kvm_vcpu_time_info, hv_clock);
+static DEFINE_PER_CPU_SHARED_ALIGNED(struct pvclock_vcpu_time_info, hv_clock);
-#define get_clock(cpu, field) per_cpu(hv_clock, cpu).field
+static struct pvclock_wall_clock wall_clock;
-static inline u64 kvm_get_delta(u64 last_tsc)
-{
-        int cpu = smp_processor_id();
-        u64 delta = native_read_tsc() - last_tsc;
-        return (delta * get_clock(cpu, tsc_to_system_mul)) >> KVM_SCALE;
-}
-static struct kvm_wall_clock wall_clock;
-static cycle_t kvm_clock_read(void);
 /*
 * The wallclock is the time of day when we booted. Since then, some time may
 * have elapsed since the hypervisor wrote the data. So we try to account for
@@ -55,64 +47,37 @@ static cycle_t kvm_clock_read(void);
 */
 static unsigned long kvm_get_wallclock(void)
 {
-        u32 wc_sec, wc_nsec;
+        struct pvclock_vcpu_time_info *vcpu_time;
-        u64 delta;
        struct timespec ts;
-        int version, nsec;
        int low, high;
        low = (int)__pa(&wall_clock);
        high = ((u64)__pa(&wall_clock) >> 32);
+        native_write_msr(MSR_KVM_WALL_CLOCK, low, high);
-        delta = kvm_clock_read();
+        vcpu_time = &get_cpu_var(hv_clock);
+        pvclock_read_wallclock(&wall_clock, vcpu_time, &ts);
+        put_cpu_var(hv_clock);
-        native_write_msr(MSR_KVM_WALL_CLOCK, low, high);
+        return ts.tv_sec;
-        do {
-                version = wall_clock.wc_version;
-                rmb();
-                wc_sec = wall_clock.wc_sec;
-                wc_nsec = wall_clock.wc_nsec;
-                rmb();
-        } while ((wall_clock.wc_version != version) || (version & 1));
-        delta = kvm_clock_read() - delta;
-        delta += wc_nsec;
-        nsec = do_div(delta, NSEC_PER_SEC);
-        set_normalized_timespec(&ts, wc_sec + delta, nsec);
-        /*
-         * Of all mechanisms of time adjustment I've tested, this one
-         * was the champion!
-         */
-        return ts.tv_sec + 1;
 }
 static int kvm_set_wallclock(unsigned long now)
 {
-        return 0;
+        return -1;
 }
-/*
- * This is our read_clock function. The host puts an tsc timestamp each time
- * it updates a new time. Without the tsc adjustment, we can have a situation
- * in which a vcpu starts to run earlier (smaller system_time), but probes
- * time later (compared to another vcpu), leading to backwards time
- */
 static cycle_t kvm_clock_read(void)
 {
-        u64 last_tsc, now;
+        struct pvclock_vcpu_time_info *src;
-        int cpu;
+        cycle_t ret;
-        preempt_disable();
+        src = &get_cpu_var(hv_clock);
-        cpu = smp_processor_id();
+        ret = pvclock_clocksource_read(src);
+        put_cpu_var(hv_clock);
-        last_tsc = get_clock(cpu, tsc_timestamp);
+        return ret;
-        now = get_clock(cpu, system_time);
-        now += kvm_get_delta(last_tsc);
-        preempt_enable();
-        return now;
 }
 static struct clocksource kvm_clock = {
        .name = "kvm-clock",
        .read = kvm_clock_read,
@@ -123,13 +88,14 @@ static struct clocksource kvm_clock = {
        .flags = CLOCK_SOURCE_IS_CONTINUOUS,
 };
-static int kvm_register_clock(void)
+static int kvm_register_clock(char *txt)
 {
        int cpu = smp_processor_id();
        int low, high;
        low = (int)__pa(&per_cpu(hv_clock, cpu)) | 1;
        high = ((u64)__pa(&per_cpu(hv_clock, cpu)) >> 32);
+        printk(KERN_INFO "kvm-clock: cpu %d, msr %x:%x, %s\n",
+               cpu, high, low, txt);
        return native_write_msr_safe(MSR_KVM_SYSTEM_TIME, low, high);
 }
@@ -140,12 +106,20 @@ static void kvm_setup_secondary_clock(void)
         * Now that the first cpu already had this clocksource initialized,
         * we shouldn't fail.
         */
-        WARN_ON(kvm_register_clock());
+        WARN_ON(kvm_register_clock("secondary cpu clock"));
        /* ok, done with our trickery, call native */
        setup_secondary_APIC_clock();
 }
 #endif
+#ifdef CONFIG_SMP
+void __init kvm_smp_prepare_boot_cpu(void)
+{
+        WARN_ON(kvm_register_clock("primary cpu clock"));
+        native_smp_prepare_boot_cpu();
+}
+#endif
 /*
 * After the clock is registered, the host will keep writing to the
 * registered memory location. If the guest happens to shutdown, this memory
@@ -174,7 +148,7 @@ void __init kvmclock_init(void)
                return;
        if (kvmclock && kvm_para_has_feature(KVM_FEATURE_CLOCKSOURCE)) {
-                if (kvm_register_clock())
+                if (kvm_register_clock("boot clock"))
                        return;
                pv_time_ops.get_wallclock = kvm_get_wallclock;
                pv_time_ops.set_wallclock = kvm_set_wallclock;
@@ -182,6 +156,9 @@ void __init kvmclock_init(void)
 #ifdef CONFIG_X86_LOCAL_APIC
                pv_apic_ops.setup_secondary_clock = kvm_setup_secondary_clock;
 #endif
+#ifdef CONFIG_SMP
+                smp_ops.smp_prepare_boot_cpu = kvm_smp_prepare_boot_cpu;
+#endif
                machine_ops.shutdown  = kvm_shutdown;
 #ifdef CONFIG_KEXEC
                machine_ops.crash_shutdown  = kvm_crash_shutdown;
diff --git a/arch/x86/kernel/pvclock.c b/arch/x86/kernel/pvclock.c
new file mode 100644
index 000000000000..05fbe9a0325a
--- /dev/null
+++ b/arch/x86/kernel/pvclock.c
@@ -0,0 +1,141 @@
+/*  paravirtual clock -- common code used by kvm/xen
+    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
+    (at your option) any later version.
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+    You should have received a copy of the GNU General Public License
+    along with this program; if not, write to the Free Software
+    Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+*/
+#include <linux/kernel.h>
+#include <linux/percpu.h>
+#include <asm/pvclock.h>
+/*
+ * These are perodically updated
+ *    xen: magic shared_info page
+ *    kvm: gpa registered via msr
+ * and then copied here.
+ */
+struct pvclock_shadow_time {
+        u64 tsc_timestamp;     /* TSC at last update of time vals.  */
+        u64 system_timestamp;  /* Time, in nanosecs, since boot.    */
+        u32 tsc_to_nsec_mul;
+        int tsc_shift;
+        u32 version;
+};
+/*
+ * Scale a 64-bit delta by scaling and multiplying by a 32-bit fraction,
+ * yielding a 64-bit result.
+ */
+static inline u64 scale_delta(u64 delta, u32 mul_frac, int shift)
+{
+        u64 product;
+#ifdef __i386__
+        u32 tmp1, tmp2;
+#endif
+        if (shift < 0)
+                delta >>= -shift;
+        else
+                delta <<= shift;
+#ifdef __i386__
+        __asm__ (
+                "mul  %5       ; "
+                "mov  %4,%%eax ; "
+                "mov  %%edx,%4 ; "
+                "mul  %5       ; "
+                "xor  %5,%5    ; "
+                "add  %4,%%eax ; "
+                "adc  %5,%%edx ; "
+                : "=A" (product), "=r" (tmp1), "=r" (tmp2)
+                : "a" ((u32)delta), "1" ((u32)(delta >> 32)), "2" (mul_frac) );
+#elif __x86_64__
+        __asm__ (
+                "mul %%rdx ; shrd $32,%%rdx,%%rax"
+                : "=a" (product) : "0" (delta), "d" ((u64)mul_frac) );
+#else
+#error implement me!
+#endif
+        return product;
+}
+static u64 pvclock_get_nsec_offset(struct pvclock_shadow_time *shadow)
+{
+        u64 delta = native_read_tsc() - shadow->tsc_timestamp;
+        return scale_delta(delta, shadow->tsc_to_nsec_mul, shadow->tsc_shift);
+}
+/*
+ * Reads a consistent set of time-base values from hypervisor,
+ * into a shadow data area.
+ */
+static unsigned pvclock_get_time_values(struct pvclock_shadow_time *dst,
+                                        struct pvclock_vcpu_time_info *src)
+{
+        do {
+                dst->version = src->version;
+                rmb();          /* fetch version before data */
+                dst->tsc_timestamp     = src->tsc_timestamp;
+                dst->system_timestamp  = src->system_time;
+                dst->tsc_to_nsec_mul   = src->tsc_to_system_mul;
+                dst->tsc_shift         = src->tsc_shift;
+                rmb();          /* test version after fetching data */
+        } while ((src->version & 1) || (dst->version != src->version));
+        return dst->version;
+}
+cycle_t pvclock_clocksource_read(struct pvclock_vcpu_time_info *src)
+{
+        struct pvclock_shadow_time shadow;
+        unsigned version;
+        cycle_t ret, offset;
+        do {
+                version = pvclock_get_time_values(&shadow, src);
+                barrier();
+                offset = pvclock_get_nsec_offset(&shadow);
+                ret = shadow.system_timestamp + offset;
+                barrier();
+        } while (version != src->version);
+        return ret;
+}
+void pvclock_read_wallclock(struct pvclock_wall_clock *wall_clock,
+                            struct pvclock_vcpu_time_info *vcpu_time,
+                            struct timespec *ts)
+{
+        u32 version;
+        u64 delta;
+        struct timespec now;
+        /* get wallclock at system boot */
+        do {
+                version = wall_clock->version;
+                rmb();          /* fetch version before time */
+                now.tv_sec  = wall_clock->sec;
+                now.tv_nsec = wall_clock->nsec;
+                rmb();          /* fetch time before checking version */
+        } while ((wall_clock->version & 1) || (version != wall_clock->version));
+        delta = pvclock_clocksource_read(vcpu_time);    /* time since system boot */
+        delta += now.tv_sec * (u64)NSEC_PER_SEC + now.tv_nsec;
+        now.tv_nsec = do_div(delta, NSEC_PER_SEC);
+        now.tv_sec = delta;
+        set_normalized_timespec(ts, now.tv_sec, now.tv_nsec);
+}
diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c
index f2f5d260874e..3829aa7b663f 100644
--- a/arch/x86/kvm/i8254.c
+++ b/arch/x86/kvm/i8254.c
@@ -200,9 +200,12 @@ int __pit_timer_fn(struct kvm_kpit_state *ps)
        atomic_inc(&pt->pending);
        smp_mb__after_atomic_inc();
-        if (vcpu0 && waitqueue_active(&vcpu0->wq)) {
+        if (vcpu0) {
-                vcpu0->arch.mp_state = KVM_MP_STATE_RUNNABLE;
+                set_bit(KVM_REQ_PENDING_TIMER, &vcpu0->requests);
-                wake_up_interruptible(&vcpu0->wq);
+                if (waitqueue_active(&vcpu0->wq)) {
+                        vcpu0->arch.mp_state = KVM_MP_STATE_RUNNABLE;
+                        wake_up_interruptible(&vcpu0->wq);
+                }
        }
        pt->timer.expires = ktime_add_ns(pt->timer.expires, pt->period);
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index c297c50eba63..ebc03f5ae162 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -940,6 +940,7 @@ static int __apic_timer_fn(struct kvm_lapic *apic)
        wait_queue_head_t *q = &apic->vcpu->wq;
        atomic_inc(&apic->timer.pending);
+        set_bit(KVM_REQ_PENDING_TIMER, &apic->vcpu->requests);
        if (waitqueue_active(q)) {
                apic->vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE;
                wake_up_interruptible(q);
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index ee3f53098f0c..7e7c3969f7a2 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -640,6 +640,7 @@ static void rmap_write_protect(struct kvm *kvm, u64 gfn)
                        rmap_remove(kvm, spte);
                        --kvm->stat.lpages;
                        set_shadow_pte(spte, shadow_trap_nonpresent_pte);
+                        spte = NULL;
                        write_protected = 1;
                }
                spte = rmap_next(kvm, rmapp, spte);
@@ -1082,10 +1083,6 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte,
                struct kvm_mmu_page *shadow;
                spte |= PT_WRITABLE_MASK;
-                if (user_fault) {
-                        mmu_unshadow(vcpu->kvm, gfn);
-                        goto unshadowed;
-                }
                shadow = kvm_mmu_lookup_page(vcpu->kvm, gfn);
                if (shadow ||
@@ -1102,8 +1099,6 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte,
                }
        }
-unshadowed:
        if (pte_access & ACC_WRITE_MASK)
                mark_page_dirty(vcpu->kvm, gfn);
@@ -1580,11 +1575,13 @@ static void mmu_pte_write_new_pte(struct kvm_vcpu *vcpu,
                                  u64 *spte,
                                  const void *new)
 {
-        if ((sp->role.level != PT_PAGE_TABLE_LEVEL)
+        if (sp->role.level != PT_PAGE_TABLE_LEVEL) {
-            && !vcpu->arch.update_pte.largepage) {
+                if (!vcpu->arch.update_pte.largepage ||
-                ++vcpu->kvm->stat.mmu_pde_zapped;
+                    sp->role.glevels == PT32_ROOT_LEVEL) {
-                return;
+                        ++vcpu->kvm->stat.mmu_pde_zapped;
-        }
+                        return;
+                }
+        }
        ++vcpu->kvm->stat.mmu_pte_updated;
        if (sp->role.glevels == PT32_ROOT_LEVEL)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 02efbe75f317..540e95179074 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -566,7 +566,7 @@ static void vmx_save_host_state(struct kvm_vcpu *vcpu)
        load_transition_efer(vmx);
 }
-static void vmx_load_host_state(struct vcpu_vmx *vmx)
+static void __vmx_load_host_state(struct vcpu_vmx *vmx)
 {
        unsigned long flags;
@@ -596,6 +596,13 @@ static void vmx_load_host_state(struct vcpu_vmx *vmx)
        reload_host_efer(vmx);
 }
+static void vmx_load_host_state(struct vcpu_vmx *vmx)
+{
+        preempt_disable();
+        __vmx_load_host_state(vmx);
+        preempt_enable();
+}
 /*
 * Switches to specified vcpu, until a matching vcpu_put(), but assumes
 * vcpu mutex is already taken.
@@ -654,7 +661,7 @@ static void vmx_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 static void vmx_vcpu_put(struct kvm_vcpu *vcpu)
 {
-        vmx_load_host_state(to_vmx(vcpu));
+        __vmx_load_host_state(to_vmx(vcpu));
 }
 static void vmx_fpu_activate(struct kvm_vcpu *vcpu)
@@ -884,11 +891,8 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data)
        switch (msr_index) {
 #ifdef CONFIG_X86_64
        case MSR_EFER:
+                vmx_load_host_state(vmx);
                ret = kvm_set_msr_common(vcpu, msr_index, data);
-                if (vmx->host_state.loaded) {
-                        reload_host_efer(vmx);
-                        load_transition_efer(vmx);
-                }
                break;
        case MSR_FS_BASE:
                vmcs_writel(GUEST_FS_BASE, data);
@@ -910,11 +914,10 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data)
                guest_write_tsc(data);
                break;
        default:
+                vmx_load_host_state(vmx);
                msr = find_msr_entry(vmx, msr_index);
                if (msr) {
                        msr->data = data;
-                        if (vmx->host_state.loaded)
-                                load_msrs(vmx->guest_msrs, vmx->save_nmsrs);
                        break;
                }
                ret = kvm_set_msr_common(vcpu, msr_index, data);
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 00acf1301a15..63a77caa59f1 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -492,8 +492,8 @@ static int do_set_msr(struct kvm_vcpu *vcpu, unsigned index, u64 *data)
 static void kvm_write_wall_clock(struct kvm *kvm, gpa_t wall_clock)
 {
        static int version;
-        struct kvm_wall_clock wc;
+        struct pvclock_wall_clock wc;
-        struct timespec wc_ts;
+        struct timespec now, sys, boot;
        if (!wall_clock)
                return;
@@ -502,10 +502,19 @@ static void kvm_write_wall_clock(struct kvm *kvm, gpa_t wall_clock)
        kvm_write_guest(kvm, wall_clock, &version, sizeof(version));
-        wc_ts = current_kernel_time();
+        /*
-        wc.wc_sec = wc_ts.tv_sec;
+         * The guest calculates current wall clock time by adding
-        wc.wc_nsec = wc_ts.tv_nsec;
+         * system time (updated by kvm_write_guest_time below) to the
-        wc.wc_version = version;
+         * wall clock specified here.  guest system time equals host
+         * system time for us, thus we must fill in host boot time here.
+         */
+        now = current_kernel_time();
+        ktime_get_ts(&sys);
+        boot = ns_to_timespec(timespec_to_ns(&now) - timespec_to_ns(&sys));
+        wc.sec = boot.tv_sec;
+        wc.nsec = boot.tv_nsec;
+        wc.version = version;
        kvm_write_guest(kvm, wall_clock, &wc, sizeof(wc));
@@ -513,6 +522,45 @@ static void kvm_write_wall_clock(struct kvm *kvm, gpa_t wall_clock)
        kvm_write_guest(kvm, wall_clock, &version, sizeof(version));
 }
+static uint32_t div_frac(uint32_t dividend, uint32_t divisor)
+{
+        uint32_t quotient, remainder;
+        /* Don't try to replace with do_div(), this one calculates
+         * "(dividend << 32) / divisor" */
+        __asm__ ( "divl %4"
+                  : "=a" (quotient), "=d" (remainder)
+                  : "0" (0), "1" (dividend), "r" (divisor) );
+        return quotient;
+}
+static void kvm_set_time_scale(uint32_t tsc_khz, struct pvclock_vcpu_time_info *hv_clock)
+{
+        uint64_t nsecs = 1000000000LL;
+        int32_t  shift = 0;
+        uint64_t tps64;
+        uint32_t tps32;
+        tps64 = tsc_khz * 1000LL;
+        while (tps64 > nsecs*2) {
+                tps64 >>= 1;
+                shift--;
+        }
+        tps32 = (uint32_t)tps64;
+        while (tps32 <= (uint32_t)nsecs) {
+                tps32 <<= 1;
+                shift++;
+        }
+        hv_clock->tsc_shift = shift;
+        hv_clock->tsc_to_system_mul = div_frac(nsecs, tps32);
+        pr_debug("%s: tsc_khz %u, tsc_shift %d, tsc_mul %u\n",
+                 __FUNCTION__, tsc_khz, hv_clock->tsc_shift,
+                 hv_clock->tsc_to_system_mul);
+}
 static void kvm_write_guest_time(struct kvm_vcpu *v)
 {
        struct timespec ts;
@@ -523,6 +571,11 @@ static void kvm_write_guest_time(struct kvm_vcpu *v)
        if ((!vcpu->time_page))
                return;
+        if (unlikely(vcpu->hv_clock_tsc_khz != tsc_khz)) {
+                kvm_set_time_scale(tsc_khz, &vcpu->hv_clock);
+                vcpu->hv_clock_tsc_khz = tsc_khz;
+        }
        /* Keep irq disabled to prevent changes to the clock */
        local_irq_save(flags);
        kvm_get_msr(v, MSR_IA32_TIME_STAMP_COUNTER,
@@ -537,14 +590,14 @@ static void kvm_write_guest_time(struct kvm_vcpu *v)
        /*
         * The interface expects us to write an even number signaling that the
         * update is finished. Since the guest won't see the intermediate
-         * state, we just write "2" at the end
+         * state, we just increase by 2 at the end.
         */
-        vcpu->hv_clock.version = 2;
+        vcpu->hv_clock.version += 2;
        shared_kaddr = kmap_atomic(vcpu->time_page, KM_USER0);
        memcpy(shared_kaddr + vcpu->time_offset, &vcpu->hv_clock,
-                sizeof(vcpu->hv_clock));
+               sizeof(vcpu->hv_clock));
        kunmap_atomic(shared_kaddr, KM_USER0);
@@ -599,10 +652,6 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
                /* ...but clean it before doing the actual write */
                vcpu->arch.time_offset = data & ~(PAGE_MASK | 1);
-                vcpu->arch.hv_clock.tsc_to_system_mul =
-                                        clocksource_khz2mult(tsc_khz, 22);
-                vcpu->arch.hv_clock.tsc_shift = 22;
                down_read(&current->mm->mmap_sem);
                vcpu->arch.time_page =
                                gfn_to_page(vcpu->kvm, data >> PAGE_SHIFT);
@@ -2759,6 +2808,8 @@ again:
        if (vcpu->requests) {
                if (test_and_clear_bit(KVM_REQ_MIGRATE_TIMER, &vcpu->requests))
                        __kvm_migrate_timers(vcpu);
+                if (test_and_clear_bit(KVM_REQ_TLB_FLUSH, &vcpu->requests))
+                        kvm_x86_ops->tlb_flush(vcpu);
                if (test_and_clear_bit(KVM_REQ_REPORT_TPR_ACCESS,
                                       &vcpu->requests)) {
                        kvm_run->exit_reason = KVM_EXIT_TPR_ACCESS;
@@ -2772,6 +2823,7 @@ again:
                }
        }
+        clear_bit(KVM_REQ_PENDING_TIMER, &vcpu->requests);
        kvm_inject_pending_timer_irqs(vcpu);
        preempt_disable();
@@ -2781,21 +2833,13 @@ again:
        local_irq_disable();
-        if (need_resched()) {
+        if (vcpu->requests || need_resched()) {
                local_irq_enable();
                preempt_enable();
                r = 1;
                goto out;
        }
-        if (vcpu->requests)
-                if (test_bit(KVM_REQ_MMU_RELOAD, &vcpu->requests)) {
-                        local_irq_enable();
-                        preempt_enable();
-                        r = 1;
-                        goto out;
-                }
        if (signal_pending(current)) {
                local_irq_enable();
                preempt_enable();
@@ -2825,9 +2869,6 @@ again:
        kvm_guest_enter();
-        if (vcpu->requests)
-                if (test_and_clear_bit(KVM_REQ_TLB_FLUSH, &vcpu->requests))
-                        kvm_x86_ops->tlb_flush(vcpu);
        KVMTRACE_0D(VMENTRY, vcpu, entryexit);
        kvm_x86_ops->run(vcpu, kvm_run);
diff --git a/arch/x86/xen/Kconfig b/arch/x86/xen/Kconfig
index 2e641be2737e..6c388e593bc8 100644
--- a/arch/x86/xen/Kconfig
+++ b/arch/x86/xen/Kconfig
@@ -5,8 +5,9 @@
 config XEN
        bool "Xen guest support"
        select PARAVIRT
+        select PARAVIRT_CLOCK
        depends on X86_32
-        depends on X86_CMPXCHG && X86_TSC && !(X86_VISWS || X86_VOYAGER)
+        depends on X86_CMPXCHG && X86_TSC && X86_PAE && !(X86_VISWS || X86_VOYAGER)
        help
          This is the Linux Xen port.  Enabling this will allow the
          kernel to boot in a paravirtualized environment under the
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index c8a56e457d61..f09c1c69c37a 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -785,38 +785,35 @@ static __init void xen_set_pte_init(pte_t *ptep, pte_t pte)
 static __init void xen_pagetable_setup_start(pgd_t *base)
 {
        pgd_t *xen_pgd = (pgd_t *)xen_start_info->pt_base;
+        int i;
        /* special set_pte for pagetable initialization */
        pv_mmu_ops.set_pte = xen_set_pte_init;
        init_mm.pgd = base;
        /*
-         * copy top-level of Xen-supplied pagetable into place.  For
+         * copy top-level of Xen-supplied pagetable into place.  This
-         * !PAE we can use this as-is, but for PAE it is a stand-in
+         * is a stand-in while we copy the pmd pages.
-         * while we copy the pmd pages.
         */
        memcpy(base, xen_pgd, PTRS_PER_PGD * sizeof(pgd_t));
-        if (PTRS_PER_PMD > 1) {
+        /*
-                int i;
+         * For PAE, need to allocate new pmds, rather than
-                /*
+         * share Xen's, since Xen doesn't like pmd's being
-                 * For PAE, need to allocate new pmds, rather than
+         * shared between address spaces.
-                 * share Xen's, since Xen doesn't like pmd's being
+         */
-                 * shared between address spaces.
+        for (i = 0; i < PTRS_PER_PGD; i++) {
-                 */
+                if (pgd_val_ma(xen_pgd[i]) & _PAGE_PRESENT) {
-                for (i = 0; i < PTRS_PER_PGD; i++) {
+                        pmd_t *pmd = (pmd_t *)alloc_bootmem_low_pages(PAGE_SIZE);
-                        if (pgd_val_ma(xen_pgd[i]) & _PAGE_PRESENT) {
-                                pmd_t *pmd = (pmd_t *)alloc_bootmem_low_pages(PAGE_SIZE);
-                                memcpy(pmd, (void *)pgd_page_vaddr(xen_pgd[i]),
+                        memcpy(pmd, (void *)pgd_page_vaddr(xen_pgd[i]),
-                                       PAGE_SIZE);
+                               PAGE_SIZE);
-                                make_lowmem_page_readonly(pmd);
+                        make_lowmem_page_readonly(pmd);
-                                set_pgd(&base[i], __pgd(1 + __pa(pmd)));
+                        set_pgd(&base[i], __pgd(1 + __pa(pmd)));
-                        } else
+                } else
-                                pgd_clear(&base[i]);
+                        pgd_clear(&base[i]);
-                }
        }
        /* make sure zero_page is mapped RO so we can use it in pagetables */
@@ -873,17 +870,7 @@ static __init void xen_pagetable_setup_done(pgd_t *base)
        /* Actually pin the pagetable down, but we can't set PG_pinned
           yet because the page structures don't exist yet. */
-        {
+        pin_pagetable_pfn(MMUEXT_PIN_L3_TABLE, PFN_DOWN(__pa(base)));
-                unsigned level;
-#ifdef CONFIG_X86_PAE
-                level = MMUEXT_PIN_L3_TABLE;
-#else
-                level = MMUEXT_PIN_L2_TABLE;
-#endif
-                pin_pagetable_pfn(level, PFN_DOWN(__pa(base)));
-        }
 }
 /* This is called once we have the cpu_possible_map */
@@ -1093,7 +1080,6 @@ static const struct pv_mmu_ops xen_mmu_ops __initdata = {
        .make_pte = xen_make_pte,
        .make_pgd = xen_make_pgd,
-#ifdef CONFIG_X86_PAE
        .set_pte_atomic = xen_set_pte_atomic,
        .set_pte_present = xen_set_pte_at,
        .set_pud = xen_set_pud,
@@ -1102,7 +1088,6 @@ static const struct pv_mmu_ops xen_mmu_ops __initdata = {
        .make_pmd = xen_make_pmd,
        .pmd_val = xen_pmd_val,
-#endif  /* PAE */
        .activate_mm = xen_activate_mm,
        .dup_mmap = xen_dup_mmap,
@@ -1228,6 +1213,11 @@ asmlinkage void __init xen_start_kernel(void)
        if (xen_feature(XENFEAT_supervisor_mode_kernel))
                pv_info.kernel_rpl = 0;
+        /* Prevent unwanted bits from being set in PTEs. */
+        __supported_pte_mask &= ~_PAGE_GLOBAL;
+        if (!is_initial_xendomain())
+                __supported_pte_mask &= ~(_PAGE_PWT | _PAGE_PCD);
        /* set the limit of our address space */
        xen_reserve_top();
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index 3525ef523a74..df40bf74ea75 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -179,50 +179,56 @@ out:
                preempt_enable();
 }
-pteval_t xen_pte_val(pte_t pte)
+/* Assume pteval_t is equivalent to all the other *val_t types. */
+static pteval_t pte_mfn_to_pfn(pteval_t val)
+{
+        if (val & _PAGE_PRESENT) {
+                unsigned long mfn = (val & PTE_MASK) >> PAGE_SHIFT;
+                pteval_t flags = val & ~PTE_MASK;
+                val = (mfn_to_pfn(mfn) << PAGE_SHIFT) | flags;
+        }
+        return val;
+}
+static pteval_t pte_pfn_to_mfn(pteval_t val)
 {
-        pteval_t ret = pte.pte;
+        if (val & _PAGE_PRESENT) {
+                unsigned long pfn = (val & PTE_MASK) >> PAGE_SHIFT;
+                pteval_t flags = val & ~PTE_MASK;
+                val = (pfn_to_mfn(pfn) << PAGE_SHIFT) | flags;
+        }
-        if (ret & _PAGE_PRESENT)
+        return val;
-                ret = machine_to_phys(XMADDR(ret)).paddr | _PAGE_PRESENT;
+}
-        return ret;
+pteval_t xen_pte_val(pte_t pte)
+{
+        return pte_mfn_to_pfn(pte.pte);
 }
 pgdval_t xen_pgd_val(pgd_t pgd)
 {
-        pgdval_t ret = pgd.pgd;
+        return pte_mfn_to_pfn(pgd.pgd);
-        if (ret & _PAGE_PRESENT)
-                ret = machine_to_phys(XMADDR(ret)).paddr | _PAGE_PRESENT;
-        return ret;
 }
 pte_t xen_make_pte(pteval_t pte)
 {
-        if (pte & _PAGE_PRESENT) {
+        pte = pte_pfn_to_mfn(pte);
-                pte = phys_to_machine(XPADDR(pte)).maddr;
+        return native_make_pte(pte);
-                pte &= ~(_PAGE_PCD | _PAGE_PWT);
-        }
-        return (pte_t){ .pte = pte };
 }
 pgd_t xen_make_pgd(pgdval_t pgd)
 {
-        if (pgd & _PAGE_PRESENT)
+        pgd = pte_pfn_to_mfn(pgd);
-                pgd = phys_to_machine(XPADDR(pgd)).maddr;
+        return native_make_pgd(pgd);
-        return (pgd_t){ pgd };
 }
 pmdval_t xen_pmd_val(pmd_t pmd)
 {
-        pmdval_t ret = native_pmd_val(pmd);
+        return pte_mfn_to_pfn(pmd.pmd);
-        if (ret & _PAGE_PRESENT)
-                ret = machine_to_phys(XMADDR(ret)).paddr | _PAGE_PRESENT;
-        return ret;
 }
-#ifdef CONFIG_X86_PAE
 void xen_set_pud(pud_t *ptr, pud_t val)
 {
        struct multicall_space mcs;
@@ -267,17 +273,9 @@ void xen_pmd_clear(pmd_t *pmdp)
 pmd_t xen_make_pmd(pmdval_t pmd)
 {
-        if (pmd & _PAGE_PRESENT)
+        pmd = pte_pfn_to_mfn(pmd);
-                pmd = phys_to_machine(XPADDR(pmd)).maddr;
        return native_make_pmd(pmd);
 }
-#else  /* !PAE */
-void xen_set_pte(pte_t *ptep, pte_t pte)
-{
-        *ptep = pte;
-}
-#endif  /* CONFIG_X86_PAE */
 /*
  (Yet another) pagetable walker.  This one is intended for pinning a
@@ -430,8 +428,6 @@ static int pin_page(struct page *page, enum pt_level level)
   read-only, and can be pinned. */
 void xen_pgd_pin(pgd_t *pgd)
 {
-        unsigned level;
        xen_mc_batch();
        if (pgd_walk(pgd, pin_page, TASK_SIZE)) {
@@ -441,14 +437,7 @@ void xen_pgd_pin(pgd_t *pgd)
                xen_mc_batch();
        }
-#ifdef CONFIG_X86_PAE
+        xen_do_pin(MMUEXT_PIN_L3_TABLE, PFN_DOWN(__pa(pgd)));
-        level = MMUEXT_PIN_L3_TABLE;
-#else
-        level = MMUEXT_PIN_L2_TABLE;
-#endif
-        xen_do_pin(level, PFN_DOWN(__pa(pgd)));
        xen_mc_issue(0);
 }
diff --git a/arch/x86/xen/mmu.h b/arch/x86/xen/mmu.h
index b5e189b1519d..5fe961caffd4 100644
--- a/arch/x86/xen/mmu.h
+++ b/arch/x86/xen/mmu.h
@@ -37,14 +37,13 @@ void xen_exit_mmap(struct mm_struct *mm);
 void xen_pgd_pin(pgd_t *pgd);
 //void xen_pgd_unpin(pgd_t *pgd);
-#ifdef CONFIG_X86_PAE
+pteval_t xen_pte_val(pte_t);
-unsigned long long xen_pte_val(pte_t);
+pmdval_t xen_pmd_val(pmd_t);
-unsigned long long xen_pmd_val(pmd_t);
+pgdval_t xen_pgd_val(pgd_t);
-unsigned long long xen_pgd_val(pgd_t);
-pte_t xen_make_pte(unsigned long long);
+pte_t xen_make_pte(pteval_t);
-pmd_t xen_make_pmd(unsigned long long);
+pmd_t xen_make_pmd(pmdval_t);
-pgd_t xen_make_pgd(unsigned long long);
+pgd_t xen_make_pgd(pgdval_t);
 void xen_set_pte_at(struct mm_struct *mm, unsigned long addr,
                    pte_t *ptep, pte_t pteval);
@@ -53,15 +52,4 @@ void xen_set_pud(pud_t *ptr, pud_t val);
 void xen_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep);
 void xen_pmd_clear(pmd_t *pmdp);
-#else
-unsigned long xen_pte_val(pte_t);
-unsigned long xen_pmd_val(pmd_t);
-unsigned long xen_pgd_val(pgd_t);
-pte_t xen_make_pte(unsigned long);
-pmd_t xen_make_pmd(unsigned long);
-pgd_t xen_make_pgd(unsigned long);
-#endif
 #endif  /* _XEN_MMU_H */
diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c
index 52b2e3856980..41e217503c96 100644
--- a/arch/x86/xen/time.c
+++ b/arch/x86/xen/time.c
@@ -14,6 +14,7 @@
 #include <linux/kernel_stat.h>
 #include <linux/math64.h>
+#include <asm/pvclock.h>
 #include <asm/xen/hypervisor.h>
 #include <asm/xen/hypercall.h>
@@ -31,17 +32,6 @@
 static cycle_t xen_clocksource_read(void);
-/* These are perodically updated in shared_info, and then copied here. */
-struct shadow_time_info {
-        u64 tsc_timestamp;     /* TSC at last update of time vals.  */
-        u64 system_timestamp;  /* Time, in nanosecs, since boot.    */
-        u32 tsc_to_nsec_mul;
-        int tsc_shift;
-        u32 version;
-};
-static DEFINE_PER_CPU(struct shadow_time_info, shadow_time);
 /* runstate info updated by Xen */
 static DEFINE_PER_CPU(struct vcpu_runstate_info, runstate);
@@ -211,7 +201,7 @@ unsigned long long xen_sched_clock(void)
 unsigned long xen_cpu_khz(void)
 {
        u64 xen_khz = 1000000ULL << 32;
-        const struct vcpu_time_info *info =
+        const struct pvclock_vcpu_time_info *info =
                &HYPERVISOR_shared_info->vcpu_info[0].time;
        do_div(xen_khz, info->tsc_to_system_mul);
@@ -223,121 +213,26 @@ unsigned long xen_cpu_khz(void)
        return xen_khz;
 }
-/*
- * Reads a consistent set of time-base values from Xen, into a shadow data
- * area.
- */
-static unsigned get_time_values_from_xen(void)
-{
-        struct vcpu_time_info   *src;
-        struct shadow_time_info *dst;
-        /* src is shared memory with the hypervisor, so we need to
-           make sure we get a consistent snapshot, even in the face of
-           being preempted. */
-        src = &__get_cpu_var(xen_vcpu)->time;
-        dst = &__get_cpu_var(shadow_time);
-        do {
-                dst->version = src->version;
-                rmb();          /* fetch version before data */
-                dst->tsc_timestamp     = src->tsc_timestamp;
-                dst->system_timestamp  = src->system_time;
-                dst->tsc_to_nsec_mul   = src->tsc_to_system_mul;
-                dst->tsc_shift         = src->tsc_shift;
-                rmb();          /* test version after fetching data */
-        } while ((src->version & 1) | (dst->version ^ src->version));
-        return dst->version;
-}
-/*
- * Scale a 64-bit delta by scaling and multiplying by a 32-bit fraction,
- * yielding a 64-bit result.
- */
-static inline u64 scale_delta(u64 delta, u32 mul_frac, int shift)
-{
-        u64 product;
-#ifdef __i386__
-        u32 tmp1, tmp2;
-#endif
-        if (shift < 0)
-                delta >>= -shift;
-        else
-                delta <<= shift;
-#ifdef __i386__
-        __asm__ (
-                "mul  %5       ; "
-                "mov  %4,%%eax ; "
-                "mov  %%edx,%4 ; "
-                "mul  %5       ; "
-                "xor  %5,%5    ; "
-                "add  %4,%%eax ; "
-                "adc  %5,%%edx ; "
-                : "=A" (product), "=r" (tmp1), "=r" (tmp2)
-                : "a" ((u32)delta), "1" ((u32)(delta >> 32)), "2" (mul_frac) );
-#elif __x86_64__
-        __asm__ (
-                "mul %%rdx ; shrd $32,%%rdx,%%rax"
-                : "=a" (product) : "0" (delta), "d" ((u64)mul_frac) );
-#else
-#error implement me!
-#endif
-        return product;
-}
-static u64 get_nsec_offset(struct shadow_time_info *shadow)
-{
-        u64 now, delta;
-        now = native_read_tsc();
-        delta = now - shadow->tsc_timestamp;
-        return scale_delta(delta, shadow->tsc_to_nsec_mul, shadow->tsc_shift);
-}
 static cycle_t xen_clocksource_read(void)
 {
-        struct shadow_time_info *shadow = &get_cpu_var(shadow_time);
+        struct pvclock_vcpu_time_info *src;
        cycle_t ret;
-        unsigned version;
-        do {
-                version = get_time_values_from_xen();
-                barrier();
-                ret = shadow->system_timestamp + get_nsec_offset(shadow);
-                barrier();
-        } while (version != __get_cpu_var(xen_vcpu)->time.version);
-        put_cpu_var(shadow_time);
+        src = &get_cpu_var(xen_vcpu)->time;
+        ret = pvclock_clocksource_read(src);
+        put_cpu_var(xen_vcpu);
        return ret;
 }
 static void xen_read_wallclock(struct timespec *ts)
 {
-        const struct shared_info *s = HYPERVISOR_shared_info;
+        struct shared_info *s = HYPERVISOR_shared_info;
-        u32 version;
+        struct pvclock_wall_clock *wall_clock = &(s->wc);
-        u64 delta;
+        struct pvclock_vcpu_time_info *vcpu_time;
-        struct timespec now;
-        /* get wallclock at system boot */
-        do {
-                version = s->wc_version;
-                rmb();          /* fetch version before time */
-                now.tv_sec  = s->wc_sec;
-                now.tv_nsec = s->wc_nsec;
-                rmb();          /* fetch time before checking version */
-        } while ((s->wc_version & 1) | (version ^ s->wc_version));
-        delta = xen_clocksource_read(); /* time since system boot */
+        vcpu_time = &get_cpu_var(xen_vcpu)->time;
-        delta += now.tv_sec * (u64)NSEC_PER_SEC + now.tv_nsec;
+        pvclock_read_wallclock(wall_clock, vcpu_time, ts);
+        put_cpu_var(xen_vcpu);
-        now.tv_nsec = do_div(delta, NSEC_PER_SEC);
-        now.tv_sec = delta;
-        set_normalized_timespec(ts, now.tv_sec, now.tv_nsec);
 }
 unsigned long xen_get_wallclock(void)
@@ -345,7 +240,6 @@ unsigned long xen_get_wallclock(void)
        struct timespec ts;
        xen_read_wallclock(&ts);
        return ts.tv_sec;
 }
@@ -569,8 +463,6 @@ __init void xen_time_init(void)
 {
        int cpu = smp_processor_id();
-        get_time_values_from_xen();
        clocksource_register(&xen_clocksource);
        if (HYPERVISOR_vcpu_op(VCPUOP_stop_periodic_timer, cpu, NULL) == 0) {
diff --git a/arch/x86/xen/xen-head.S b/arch/x86/xen/xen-head.S
index 288d587ce73c..6ec3b4f7719b 100644
--- a/arch/x86/xen/xen-head.S
+++ b/arch/x86/xen/xen-head.S
@@ -17,7 +17,7 @@ ENTRY(startup_xen)
        __FINIT
-.pushsection .bss.page_aligned
+.pushsection .text
        .align PAGE_SIZE_asm
 ENTRY(hypercall_page)
        .skip 0x1000
@@ -30,11 +30,7 @@ ENTRY(hypercall_page)
        ELFNOTE(Xen, XEN_ELFNOTE_ENTRY,          .long  startup_xen)
        ELFNOTE(Xen, XEN_ELFNOTE_HYPERCALL_PAGE, .long  hypercall_page)
        ELFNOTE(Xen, XEN_ELFNOTE_FEATURES,       .asciz "!writable_page_tables|pae_pgdir_above_4gb")
-#ifdef CONFIG_X86_PAE
        ELFNOTE(Xen, XEN_ELFNOTE_PAE_MODE,       .asciz "yes")
-#else
-        ELFNOTE(Xen, XEN_ELFNOTE_PAE_MODE,       .asciz "no")
-#endif
        ELFNOTE(Xen, XEN_ELFNOTE_LOADER,         .asciz "generic")
 #endif /*CONFIG_XEN */
diff --git a/drivers/char/drm/i915_drv.c b/drivers/char/drm/i915_drv.c
index e8f3d682e3b1..93aed1c38bd2 100644
--- a/drivers/char/drm/i915_drv.c
+++ b/drivers/char/drm/i915_drv.c
@@ -389,6 +389,7 @@ static int i915_resume(struct drm_device *dev)
        pci_restore_state(dev->pdev);
        if (pci_enable_device(dev->pdev))
                return -1;
+        pci_set_master(dev->pdev);
        pci_write_config_byte(dev->pdev, LBB, dev_priv->saveLBB);
diff --git a/drivers/char/tty_ioctl.c b/drivers/char/tty_ioctl.c
index b1a757a5ee27..8f81139d6194 100644
--- a/drivers/char/tty_ioctl.c
+++ b/drivers/char/tty_ioctl.c
@@ -981,16 +981,9 @@ EXPORT_SYMBOL_GPL(tty_perform_flush);
 int n_tty_ioctl(struct tty_struct *tty, struct file *file,
                       unsigned int cmd, unsigned long arg)
 {
-        struct tty_struct *real_tty;
        unsigned long flags;
        int retval;
-        if (tty->driver->type == TTY_DRIVER_TYPE_PTY &&
-            tty->driver->subtype == PTY_TYPE_MASTER)
-                real_tty = tty->link;
-        else
-                real_tty = tty;
        switch (cmd) {
        case TCXONC:
                retval = tty_check_change(tty);
diff --git a/drivers/infiniband/hw/mthca/mthca_memfree.c b/drivers/infiniband/hw/mthca/mthca_memfree.c
index b224079d4e1f..d5862e5d99a0 100644
--- a/drivers/infiniband/hw/mthca/mthca_memfree.c
+++ b/drivers/infiniband/hw/mthca/mthca_memfree.c
@@ -109,7 +109,11 @@ static int mthca_alloc_icm_pages(struct scatterlist *mem, int order, gfp_t gfp_m
 {
        struct page *page;
-        page = alloc_pages(gfp_mask, order);
+        /*
+         * Use __GFP_ZERO because buggy firmware assumes ICM pages are
+         * cleared, and subtle failures are seen if they aren't.
+         */
+        page = alloc_pages(gfp_mask | __GFP_ZERO, order);
        if (!page)
                return -ENOMEM;
diff --git a/drivers/lguest/x86/core.c b/drivers/lguest/x86/core.c
index 5126d5d9ea0e..2e554a4ab337 100644
--- a/drivers/lguest/x86/core.c
+++ b/drivers/lguest/x86/core.c
@@ -176,7 +176,7 @@ void lguest_arch_run_guest(struct lg_cpu *cpu)
         * we set it now, so we can trap and pass that trap to the Guest if it
         * uses the FPU. */
        if (cpu->ts)
-                lguest_set_ts();
+                unlazy_fpu(current);
        /* SYSENTER is an optimized way of doing system calls.  We can't allow
         * it because it always jumps to privilege level 0.  A normal Guest
@@ -196,6 +196,10 @@ void lguest_arch_run_guest(struct lg_cpu *cpu)
         * trap made the switcher code come back, and an error code which some
         * traps set.  */
+         /* Restore SYSENTER if it's supposed to be on. */
+         if (boot_cpu_has(X86_FEATURE_SEP))
+                wrmsr(MSR_IA32_SYSENTER_CS, __KERNEL_CS, 0);
        /* If the Guest page faulted, then the cr2 register will tell us the
         * bad virtual address.  We have to grab this now, because once we
         * re-enable interrupts an interrupt could fault and thus overwrite
@@ -203,13 +207,12 @@ void lguest_arch_run_guest(struct lg_cpu *cpu)
        if (cpu->regs->trapnum == 14)
                cpu->arch.last_pagefault = read_cr2();
        /* Similarly, if we took a trap because the Guest used the FPU,
-         * we have to restore the FPU it expects to see. */
+         * we have to restore the FPU it expects to see.
+         * math_state_restore() may sleep and we may even move off to
+         * a different CPU. So all the critical stuff should be done
+         * before this.  */
        else if (cpu->regs->trapnum == 7)
                math_state_restore();
-        /* Restore SYSENTER if it's supposed to be on. */
-        if (boot_cpu_has(X86_FEATURE_SEP))
-                wrmsr(MSR_IA32_SYSENTER_CS, __KERNEL_CS, 0);
 }
 /*H:130 Now we've examined the hypercall code; our Guest can make requests.
diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
index 8662a6b7a30b..25b352b664d9 100644
--- a/drivers/watchdog/Makefile
+++ b/drivers/watchdog/Makefile
@@ -68,7 +68,6 @@ obj-$(CONFIG_WAFER_WDT) += wafer5823wdt.o
 obj-$(CONFIG_I6300ESB_WDT) += i6300esb.o
 obj-$(CONFIG_ITCO_WDT) += iTCO_wdt.o iTCO_vendor_support.o
 obj-$(CONFIG_IT8712F_WDT) += it8712f_wdt.o
-CFLAGS_hpwdt.o += -O
 obj-$(CONFIG_HP_WATCHDOG) += hpwdt.o
 obj-$(CONFIG_SC1200_WDT) += sc1200wdt.o
 obj-$(CONFIG_SCx200_WDT) += scx200_wdt.o
diff --git a/drivers/xen/events.c b/drivers/xen/events.c
index 4f0f22b020ea..76e5b7386af9 100644
--- a/drivers/xen/events.c
+++ b/drivers/xen/events.c
@@ -529,7 +529,7 @@ void xen_evtchn_do_upcall(struct pt_regs *regs)
 #ifndef CONFIG_X86 /* No need for a barrier -- XCHG is a barrier on x86. */
                /* Clear master flag /before/ clearing selector flag. */
-                rmb();
+                wmb();
 #endif
                pending_words = xchg(&vcpu_info->evtchn_pending_sel, 0);
                while (pending_words != 0) {
diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c
index c19184f2e70e..bec76b1c2bb0 100644
--- a/fs/gfs2/bmap.c
+++ b/fs/gfs2/bmap.c
@@ -246,15 +246,11 @@ static void find_metapath(const struct gfs2_sbd *sdp, u64 block,
 }
-static inline unsigned int zero_metapath_length(const struct metapath *mp,
+static inline unsigned int metapath_branch_start(const struct metapath *mp)
-                                                unsigned height)
 {
-        unsigned int i;
+        if (mp->mp_list[0] == 0)
-        for (i = 0; i < height - 1; i++) {
+                return 2;
-                if (mp->mp_list[i] != 0)
+        return 1;
-                        return i;
-        }
-        return height;
 }
 /**
@@ -436,7 +432,7 @@ static int gfs2_bmap_alloc(struct inode *inode, const sector_t lblock,
        struct gfs2_sbd *sdp = GFS2_SB(inode);
        struct buffer_head *dibh = mp->mp_bh[0];
        u64 bn, dblock = 0;
-        unsigned n, i, blks, alloced = 0, iblks = 0, zmpl = 0;
+        unsigned n, i, blks, alloced = 0, iblks = 0, branch_start = 0;
        unsigned dblks = 0;
        unsigned ptrs_per_blk;
        const unsigned end_of_metadata = height - 1;
@@ -471,9 +467,8 @@ static int gfs2_bmap_alloc(struct inode *inode, const sector_t lblock,
                        /* Building up tree height */
                        state = ALLOC_GROW_HEIGHT;
                        iblks = height - ip->i_height;
-                        zmpl = zero_metapath_length(mp, height);
+                        branch_start = metapath_branch_start(mp);
-                        iblks -= zmpl;
+                        iblks += (height - branch_start);
-                        iblks += height;
                }
        }
@@ -509,13 +504,13 @@ static int gfs2_bmap_alloc(struct inode *inode, const sector_t lblock,
                                        sizeof(struct gfs2_meta_header));
                                *ptr = zero_bn;
                                state = ALLOC_GROW_DEPTH;
-                                for(i = zmpl; i < height; i++) {
+                                for(i = branch_start; i < height; i++) {
                                        if (mp->mp_bh[i] == NULL)
                                                break;
                                        brelse(mp->mp_bh[i]);
                                        mp->mp_bh[i] = NULL;
                                }
-                                i = zmpl;
+                                i = branch_start;
                        }
                        if (n == 0)
                                break;
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index 6387523a3153..3401628d742b 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -195,7 +195,7 @@ ulong_aligned:
           depending on architecture.  I've experimented with several ways
           of writing this section such as using an else before the goto
           but this one seems to be the fastest. */
-        while ((unsigned char *)plong < end - 1) {
+        while ((unsigned char *)plong < end - sizeof(unsigned long)) {
                prefetch(plong + 1);
                if (((*plong) & LBITMASK) != lskipval)
                        break;
diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c
index 49c7cd0502cc..779d2eb649c5 100644
--- a/fs/nfs/mount_clnt.c
+++ b/fs/nfs/mount_clnt.c
@@ -130,10 +130,11 @@ static int xdr_decode_fhstatus3(struct rpc_rqst *req, __be32 *p,
                                struct mnt_fhstatus *res)
 {
        struct nfs_fh *fh = res->fh;
+        unsigned size;
        if ((res->status = ntohl(*p++)) == 0) {
-                int size = ntohl(*p++);
+                size = ntohl(*p++);
-                if (size <= NFS3_FHSIZE) {
+                if (size <= NFS3_FHSIZE && size != 0) {
                        fh->size = size;
                        memcpy(fh->data, p, size);
                } else
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 2a4a024a4e7b..614efeed5437 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -1216,8 +1216,6 @@ static int nfs_validate_mount_data(void *options,
 {
        struct nfs_mount_data *data = (struct nfs_mount_data *)options;
-        memset(args, 0, sizeof(*args));
        if (data == NULL)
                goto out_no_data;
@@ -1251,13 +1249,13 @@ static int nfs_validate_mount_data(void *options,
        case 5:
                memset(data->context, 0, sizeof(data->context));
        case 6:
-                if (data->flags & NFS_MOUNT_VER3)
+                if (data->flags & NFS_MOUNT_VER3) {
+                        if (data->root.size > NFS3_FHSIZE || data->root.size == 0)
+                                goto out_invalid_fh;
                        mntfh->size = data->root.size;
-                else
+                } else
                        mntfh->size = NFS2_FHSIZE;
-                if (mntfh->size > sizeof(mntfh->data))
-                        goto out_invalid_fh;
                memcpy(mntfh->data, data->root.data, mntfh->size);
                if (mntfh->size < sizeof(mntfh->data))
@@ -1585,24 +1583,29 @@ static int nfs_get_sb(struct file_system_type *fs_type,
 {
        struct nfs_server *server = NULL;
        struct super_block *s;
-        struct nfs_fh mntfh;
+        struct nfs_parsed_mount_data *data;
-        struct nfs_parsed_mount_data data;
+        struct nfs_fh *mntfh;
        struct dentry *mntroot;
        int (*compare_super)(struct super_block *, void *) = nfs_compare_super;
        struct nfs_sb_mountdata sb_mntdata = {
                .mntflags = flags,
        };
-        int error;
+        int error = -ENOMEM;
+        data = kzalloc(sizeof(*data), GFP_KERNEL);
+        mntfh = kzalloc(sizeof(*mntfh), GFP_KERNEL);
+        if (data == NULL || mntfh == NULL)
+                goto out_free_fh;
-        security_init_mnt_opts(&data.lsm_opts);
+        security_init_mnt_opts(&data->lsm_opts);
        /* Validate the mount data */
-        error = nfs_validate_mount_data(raw_data, &data, &mntfh, dev_name);
+        error = nfs_validate_mount_data(raw_data, data, mntfh, dev_name);
        if (error < 0)
                goto out;
        /* Get a volume representation */
-        server = nfs_create_server(&data, &mntfh);
+        server = nfs_create_server(data, mntfh);
        if (IS_ERR(server)) {
                error = PTR_ERR(server);
                goto out;
@@ -1630,16 +1633,16 @@ static int nfs_get_sb(struct file_system_type *fs_type,
        if (!s->s_root) {
                /* initial superblock/root creation */
-                nfs_fill_super(s, &data);
+                nfs_fill_super(s, data);
        }
-        mntroot = nfs_get_root(s, &mntfh);
+        mntroot = nfs_get_root(s, mntfh);
        if (IS_ERR(mntroot)) {
                error = PTR_ERR(mntroot);
                goto error_splat_super;
        }
-        error = security_sb_set_mnt_opts(s, &data.lsm_opts);
+        error = security_sb_set_mnt_opts(s, &data->lsm_opts);
        if (error)
                goto error_splat_root;
@@ -1649,9 +1652,12 @@ static int nfs_get_sb(struct file_system_type *fs_type,
        error = 0;
 out:
-        kfree(data.nfs_server.hostname);
+        kfree(data->nfs_server.hostname);
-        kfree(data.mount_server.hostname);
+        kfree(data->mount_server.hostname);
-        security_free_mnt_opts(&data.lsm_opts);
+        security_free_mnt_opts(&data->lsm_opts);
+out_free_fh:
+        kfree(mntfh);
+        kfree(data);
        return error;
 out_err_nosb:
@@ -1800,8 +1806,6 @@ static int nfs4_validate_mount_data(void *options,
        struct nfs4_mount_data *data = (struct nfs4_mount_data *)options;
        char *c;
-        memset(args, 0, sizeof(*args));
        if (data == NULL)
                goto out_no_data;
@@ -1959,26 +1963,31 @@ out_no_client_address:
 static int nfs4_get_sb(struct file_system_type *fs_type,
        int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt)
 {
-        struct nfs_parsed_mount_data data;
+        struct nfs_parsed_mount_data *data;
        struct super_block *s;
        struct nfs_server *server;
-        struct nfs_fh mntfh;
+        struct nfs_fh *mntfh;
        struct dentry *mntroot;
        int (*compare_super)(struct super_block *, void *) = nfs_compare_super;
        struct nfs_sb_mountdata sb_mntdata = {
                .mntflags = flags,
        };
-        int error;
+        int error = -ENOMEM;
-        security_init_mnt_opts(&data.lsm_opts);
+        data = kzalloc(sizeof(*data), GFP_KERNEL);
+        mntfh = kzalloc(sizeof(*mntfh), GFP_KERNEL);
+        if (data == NULL || mntfh == NULL)
+                goto out_free_fh;
+        security_init_mnt_opts(&data->lsm_opts);
        /* Validate the mount data */
-        error = nfs4_validate_mount_data(raw_data, &data, dev_name);
+        error = nfs4_validate_mount_data(raw_data, data, dev_name);
        if (error < 0)
                goto out;
        /* Get a volume representation */
-        server = nfs4_create_server(&data, &mntfh);
+        server = nfs4_create_server(data, mntfh);
        if (IS_ERR(server)) {
                error = PTR_ERR(server);
                goto out;
@@ -2009,13 +2018,13 @@ static int nfs4_get_sb(struct file_system_type *fs_type,
                nfs4_fill_super(s);
        }
-        mntroot = nfs4_get_root(s, &mntfh);
+        mntroot = nfs4_get_root(s, mntfh);
        if (IS_ERR(mntroot)) {
                error = PTR_ERR(mntroot);
                goto error_splat_super;
        }
-        error = security_sb_set_mnt_opts(s, &data.lsm_opts);
+        error = security_sb_set_mnt_opts(s, &data->lsm_opts);
        if (error)
                goto error_splat_root;
@@ -2025,10 +2034,13 @@ static int nfs4_get_sb(struct file_system_type *fs_type,
        error = 0;
 out:
-        kfree(data.client_address);
+        kfree(data->client_address);
-        kfree(data.nfs_server.export_path);
+        kfree(data->nfs_server.export_path);
-        kfree(data.nfs_server.hostname);
+        kfree(data->nfs_server.hostname);
-        security_free_mnt_opts(&data.lsm_opts);
+        security_free_mnt_opts(&data->lsm_opts);
+out_free_fh:
+        kfree(mntfh);
+        kfree(data);
        return error;
 out_free:
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 6d8ace3e3259..f333848fd3be 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -739,12 +739,13 @@ int nfs_updatepage(struct file *file, struct page *page,
        }
        status = nfs_writepage_setup(ctx, page, offset, count);
-        __set_page_dirty_nobuffers(page);
+        if (status < 0)
+                nfs_set_pageerror(page);
+        else
+                __set_page_dirty_nobuffers(page);
        dprintk("NFS:      nfs_updatepage returns %d (isize %Ld)\n",
                        status, (long long)i_size_read(inode));
-        if (status < 0)
-                nfs_set_pageerror(page);
        return status;
 }
diff --git a/fs/select.c b/fs/select.c
index 8dda969614a9..da0e88201c3a 100644
--- a/fs/select.c
+++ b/fs/select.c
@@ -249,7 +249,6 @@ int do_select(int n, fd_set_bits *fds, s64 *timeout)
                                                retval++;
                                        }
                                }
-                                cond_resched();
                        }
                        if (res_in)
                                *rinp = res_in;
@@ -257,6 +256,7 @@ int do_select(int n, fd_set_bits *fds, s64 *timeout)
                                *routp = res_out;
                        if (res_ex)
                                *rexp = res_ex;
+                        cond_resched();
                }
                wait = NULL;
                if (retval || !*timeout || signal_pending(current))
diff --git a/include/asm-alpha/percpu.h b/include/asm-alpha/percpu.h
index 82e8a94b4b2f..3495e8e00d70 100644
--- a/include/asm-alpha/percpu.h
+++ b/include/asm-alpha/percpu.h
@@ -69,6 +69,8 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
 #define __get_cpu_var(var)              per_cpu_var(var)
 #define __raw_get_cpu_var(var)          per_cpu_var(var)
+#define PER_CPU_ATTRIBUTES
 #endif /* SMP */
 #define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu_var(name)
diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
index 1d8cd01fa514..844f2a89afbc 100644
--- a/include/asm-x86/kvm_host.h
+++ b/include/asm-x86/kvm_host.h
@@ -18,6 +18,7 @@
 #include <linux/kvm_para.h>
 #include <linux/kvm_types.h>
+#include <asm/pvclock-abi.h>
 #include <asm/desc.h>
 #define KVM_MAX_VCPUS 16
@@ -282,7 +283,8 @@ struct kvm_vcpu_arch {
        struct x86_emulate_ctxt emulate_ctxt;
        gpa_t time;
-        struct kvm_vcpu_time_info hv_clock;
+        struct pvclock_vcpu_time_info hv_clock;
+        unsigned int hv_clock_tsc_khz;
        unsigned int time_offset;
        struct page *time_page;
 };
diff --git a/include/asm-x86/kvm_para.h b/include/asm-x86/kvm_para.h
index 509845942070..bfd9900742bf 100644
--- a/include/asm-x86/kvm_para.h
+++ b/include/asm-x86/kvm_para.h
@@ -48,24 +48,6 @@ struct kvm_mmu_op_release_pt {
 #ifdef __KERNEL__
 #include <asm/processor.h>
-/* xen binary-compatible interface. See xen headers for details */
-struct kvm_vcpu_time_info {
-        uint32_t version;
-        uint32_t pad0;
-        uint64_t tsc_timestamp;
-        uint64_t system_time;
-        uint32_t tsc_to_system_mul;
-        int8_t   tsc_shift;
-        int8_t   pad[3];
-} __attribute__((__packed__)); /* 32 bytes */
-struct kvm_wall_clock {
-        uint32_t wc_version;
-        uint32_t wc_sec;
-        uint32_t wc_nsec;
-} __attribute__((__packed__));
 extern void kvmclock_init(void);
diff --git a/include/asm-x86/pvclock-abi.h b/include/asm-x86/pvclock-abi.h
new file mode 100644
index 000000000000..6857f840b243
--- /dev/null
+++ b/include/asm-x86/pvclock-abi.h
@@ -0,0 +1,42 @@
+#ifndef _ASM_X86_PVCLOCK_ABI_H_
+#define _ASM_X86_PVCLOCK_ABI_H_
+#ifndef __ASSEMBLY__
+/*
+ * These structs MUST NOT be changed.
+ * They are the ABI between hypervisor and guest OS.
+ * Both Xen and KVM are using this.
+ *
+ * pvclock_vcpu_time_info holds the system time and the tsc timestamp
+ * of the last update. So the guest can use the tsc delta to get a
+ * more precise system time.  There is one per virtual cpu.
+ *
+ * pvclock_wall_clock references the point in time when the system
+ * time was zero (usually boot time), thus the guest calculates the
+ * current wall clock by adding the system time.
+ *
+ * Protocol for the "version" fields is: hypervisor raises it (making
+ * it uneven) before it starts updating the fields and raises it again
+ * (making it even) when it is done.  Thus the guest can make sure the
+ * time values it got are consistent by checking the version before
+ * and after reading them.
+ */
+struct pvclock_vcpu_time_info {
+        u32   version;
+        u32   pad0;
+        u64   tsc_timestamp;
+        u64   system_time;
+        u32   tsc_to_system_mul;
+        s8    tsc_shift;
+        u8    pad[3];
+} __attribute__((__packed__)); /* 32 bytes */
+struct pvclock_wall_clock {
+        u32   version;
+        u32   sec;
+        u32   nsec;
+} __attribute__((__packed__));
+#endif /* __ASSEMBLY__ */
+#endif /* _ASM_X86_PVCLOCK_ABI_H_ */
diff --git a/include/asm-x86/pvclock.h b/include/asm-x86/pvclock.h
new file mode 100644
index 000000000000..85b1bba8e0a3
--- /dev/null
+++ b/include/asm-x86/pvclock.h
@@ -0,0 +1,13 @@
+#ifndef _ASM_X86_PVCLOCK_H_
+#define _ASM_X86_PVCLOCK_H_
+#include <linux/clocksource.h>
+#include <asm/pvclock-abi.h>
+/* some helper functions for xen and kvm pv clock sources */
+cycle_t pvclock_clocksource_read(struct pvclock_vcpu_time_info *src);
+void pvclock_read_wallclock(struct pvclock_wall_clock *wall,
+                            struct pvclock_vcpu_time_info *vcpu,
+                            struct timespec *ts);
+#endif /* _ASM_X86_PVCLOCK_H_ */
diff --git a/include/asm-x86/xen/page.h b/include/asm-x86/xen/page.h
index baf3a4dce28c..e11f24038b1d 100644
--- a/include/asm-x86/xen/page.h
+++ b/include/asm-x86/xen/page.h
@@ -150,13 +150,9 @@ static inline pte_t __pte_ma(pteval_t x)
        return (pte_t) { .pte = x };
 }
-#ifdef CONFIG_X86_PAE
 #define pmd_val_ma(v) ((v).pmd)
 #define pud_val_ma(v) ((v).pgd.pgd)
 #define __pmd_ma(x)     ((pmd_t) { (x) } )
-#else  /* !X86_PAE */
-#define pmd_val_ma(v)   ((v).pud.pgd.pgd)
-#endif  /* CONFIG_X86_PAE */
 #define pgd_val_ma(x)   ((x).pgd)
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 092b1b25291d..de9d1df4bba2 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -33,6 +33,7 @@
 #define KVM_REQ_REPORT_TPR_ACCESS  2
 #define KVM_REQ_MMU_RELOAD         3
 #define KVM_REQ_TRIPLE_FAULT       4
+#define KVM_REQ_PENDING_TIMER      5
 struct kvm_vcpu;
 extern struct kmem_cache *kvm_vcpu_cache;
diff --git a/include/linux/tty_driver.h b/include/linux/tty_driver.h
index 59f1c0bd8f9c..d2a003586761 100644
--- a/include/linux/tty_driver.h
+++ b/include/linux/tty_driver.h
@@ -27,8 +27,7 @@
 *      This routine is called by the kernel to write a series of
 *      characters to the tty device.  The characters may come from
 *      user space or kernel space.  This routine will return the
- *      number of characters actually accepted for writing.  This
+ *      number of characters actually accepted for writing.
- *      routine is mandatory.
 *
 *      Optional: Required for writable devices.
 *
@@ -134,7 +133,7 @@
 *      This routine notifies the tty driver that it should hangup the
 *      tty device.
 *
- *      Required:
+ *      Optional:
 *
 * void (*break_ctl)(struct tty_stuct *tty, int state);
 *
diff --git a/include/xen/interface/xen.h b/include/xen/interface/xen.h
index 9b018da48cf3..819a0331cda9 100644
--- a/include/xen/interface/xen.h
+++ b/include/xen/interface/xen.h
@@ -10,6 +10,7 @@
 #define __XEN_PUBLIC_XEN_H__
 #include <asm/xen/interface.h>
+#include <asm/pvclock-abi.h>
 /*
 * XEN "SYSTEM CALLS" (a.k.a. HYPERCALLS).
@@ -336,7 +337,7 @@ struct vcpu_info {
        uint8_t evtchn_upcall_mask;
        unsigned long evtchn_pending_sel;
        struct arch_vcpu_info arch;
-        struct vcpu_time_info time;
+        struct pvclock_vcpu_time_info time;
 }; /* 64 bytes (x86) */
 /*
@@ -384,9 +385,7 @@ struct shared_info {
         * Wallclock time: updated only by control software. Guests should base
         * their gettimeofday() syscall on this wallclock-base value.
         */
-        uint32_t wc_version;      /* Version counter: see vcpu_time_info_t. */
+        struct pvclock_wall_clock wc;
-        uint32_t wc_sec;          /* Secs  00:00:00 UTC, Jan 1, 1970.  */
-        uint32_t wc_nsec;         /* Nsecs 00:00:00 UTC, Jan 1, 1970.  */
        struct arch_shared_info arch;
diff --git a/kernel/futex.c b/kernel/futex.c
index 449def8074fe..7d1136e97c14 100644
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -1096,21 +1096,64 @@ static void unqueue_me_pi(struct futex_q *q)
 * private futexes.
 */
 static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q,
-                                struct task_struct *newowner)
+                                struct task_struct *newowner,
+                                struct rw_semaphore *fshared)
 {
        u32 newtid = task_pid_vnr(newowner) | FUTEX_WAITERS;
        struct futex_pi_state *pi_state = q->pi_state;
+        struct task_struct *oldowner = pi_state->owner;
        u32 uval, curval, newval;
-        int ret;
+        int ret, attempt = 0;
        /* Owner died? */
+        if (!pi_state->owner)
+                newtid |= FUTEX_OWNER_DIED;
+        /*
+         * We are here either because we stole the rtmutex from the
+         * pending owner or we are the pending owner which failed to
+         * get the rtmutex. We have to replace the pending owner TID
+         * in the user space variable. This must be atomic as we have
+         * to preserve the owner died bit here.
+         *
+         * Note: We write the user space value _before_ changing the
+         * pi_state because we can fault here. Imagine swapped out
+         * pages or a fork, which was running right before we acquired
+         * mmap_sem, that marked all the anonymous memory readonly for
+         * cow.
+         *
+         * Modifying pi_state _before_ the user space value would
+         * leave the pi_state in an inconsistent state when we fault
+         * here, because we need to drop the hash bucket lock to
+         * handle the fault. This might be observed in the PID check
+         * in lookup_pi_state.
+         */
+retry:
+        if (get_futex_value_locked(&uval, uaddr))
+                goto handle_fault;
+        while (1) {
+                newval = (uval & FUTEX_OWNER_DIED) | newtid;
+                curval = cmpxchg_futex_value_locked(uaddr, uval, newval);
+                if (curval == -EFAULT)
+                        goto handle_fault;
+                if (curval == uval)
+                        break;
+                uval = curval;
+        }
+        /*
+         * We fixed up user space. Now we need to fix the pi_state
+         * itself.
+         */
        if (pi_state->owner != NULL) {
                spin_lock_irq(&pi_state->owner->pi_lock);
                WARN_ON(list_empty(&pi_state->list));
                list_del_init(&pi_state->list);
                spin_unlock_irq(&pi_state->owner->pi_lock);
-        } else
+        }
-                newtid |= FUTEX_OWNER_DIED;
        pi_state->owner = newowner;
@@ -1118,26 +1161,35 @@ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q,
        WARN_ON(!list_empty(&pi_state->list));
        list_add(&pi_state->list, &newowner->pi_state_list);
        spin_unlock_irq(&newowner->pi_lock);
+        return 0;
        /*
-         * We own it, so we have to replace the pending owner
+         * To handle the page fault we need to drop the hash bucket
-         * TID. This must be atomic as we have preserve the
+         * lock here. That gives the other task (either the pending
-         * owner died bit here.
+         * owner itself or the task which stole the rtmutex) the
+         * chance to try the fixup of the pi_state. So once we are
+         * back from handling the fault we need to check the pi_state
+         * after reacquiring the hash bucket lock and before trying to
+         * do another fixup. When the fixup has been done already we
+         * simply return.
         */
-        ret = get_futex_value_locked(&uval, uaddr);
+handle_fault:
+        spin_unlock(q->lock_ptr);
-        while (!ret) {
+        ret = futex_handle_fault((unsigned long)uaddr, fshared, attempt++);
-                newval = (uval & FUTEX_OWNER_DIED) | newtid;
-                curval = cmpxchg_futex_value_locked(uaddr, uval, newval);
+        spin_lock(q->lock_ptr);
-                if (curval == -EFAULT)
+        /*
-                        ret = -EFAULT;
+         * Check if someone else fixed it for us:
-                if (curval == uval)
+         */
-                        break;
+        if (pi_state->owner != oldowner)
-                uval = curval;
+                return 0;
-        }
-        return ret;
+        if (ret)
+                return ret;
+        goto retry;
 }
 /*
@@ -1507,7 +1559,7 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared,
                 * that case:
                 */
                if (q.pi_state->owner != curr)
-                        ret = fixup_pi_state_owner(uaddr, &q, curr);
+                        ret = fixup_pi_state_owner(uaddr, &q, curr, fshared);
        } else {
                /*
                 * Catch the rare case, where the lock was released
@@ -1539,7 +1591,8 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared,
                                int res;
                                owner = rt_mutex_owner(&q.pi_state->pi_mutex);
-                                res = fixup_pi_state_owner(uaddr, &q, owner);
+                                res = fixup_pi_state_owner(uaddr, &q, owner,
+                                                           fshared);
                                /* propagate -EFAULT, if the fixup failed */
                                if (res)
diff --git a/kernel/kgdb.c b/kernel/kgdb.c
index 79e3c90113c2..3ec23c3ec97f 100644
--- a/kernel/kgdb.c
+++ b/kernel/kgdb.c
@@ -1499,7 +1499,8 @@ int kgdb_nmicallback(int cpu, void *regs)
        return 1;
 }
-void kgdb_console_write(struct console *co, const char *s, unsigned count)
+static void kgdb_console_write(struct console *co, const char *s,
+   unsigned count)
 {
        unsigned long flags;
diff --git a/kernel/sched.c b/kernel/sched.c
index 27eaa47d438a..1ffa76813a01 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4449,22 +4449,20 @@ do_wait_for_common(struct completion *x, long timeout, int state)
                             signal_pending(current)) ||
                            (state == TASK_KILLABLE &&
                             fatal_signal_pending(current))) {
-                                __remove_wait_queue(&x->wait, &wait);
+                                timeout = -ERESTARTSYS;
-                                return -ERESTARTSYS;
+                                break;
                        }
                        __set_current_state(state);
                        spin_unlock_irq(&x->wait.lock);
                        timeout = schedule_timeout(timeout);
                        spin_lock_irq(&x->wait.lock);
-                        if (!timeout) {
+                } while (!x->done && timeout);
-                                __remove_wait_queue(&x->wait, &wait);
-                                return timeout;
-                        }
-                } while (!x->done);
                __remove_wait_queue(&x->wait, &wait);
+                if (!x->done)
+                        return timeout;
        }
        x->done--;
-        return timeout;
+        return timeout ?: 1;
 }
 static long __sched
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 1dad5bbb59b6..0f3c19197fa4 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -250,7 +250,8 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun)
                        if (rt_rq->rt_time || rt_rq->rt_nr_running)
                                idle = 0;
                        spin_unlock(&rt_rq->rt_runtime_lock);
-                }
+                } else if (rt_rq->rt_nr_running)
+                        idle = 0;
                if (enqueue)
                        sched_rt_rq_enqueue(rt_rq);
diff --git a/mm/memory.c b/mm/memory.c
index 9aefaae46858..d14b251a25a6 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1045,6 +1045,26 @@ no_page_table:
        return page;
 }
+/* Can we do the FOLL_ANON optimization? */
+static inline int use_zero_page(struct vm_area_struct *vma)
+{
+        /*
+         * We don't want to optimize FOLL_ANON for make_pages_present()
+         * when it tries to page in a VM_LOCKED region. As to VM_SHARED,
+         * we want to get the page from the page tables to make sure
+         * that we serialize and update with any other user of that
+         * mapping.
+         */
+        if (vma->vm_flags & (VM_LOCKED | VM_SHARED))
+                return 0;
+        /*
+         * And if we have a fault or a nopfn routine, it's not an
+         * anonymous region.
+         */
+        return !vma->vm_ops ||
+                (!vma->vm_ops->fault && !vma->vm_ops->nopfn);
+}
 int get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
                unsigned long start, int len, int write, int force,
                struct page **pages, struct vm_area_struct **vmas)
@@ -1119,8 +1139,7 @@ int get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
                foll_flags = FOLL_TOUCH;
                if (pages)
                        foll_flags |= FOLL_GET;
-                if (!write && !(vma->vm_flags & VM_LOCKED) &&
+                if (!write && use_zero_page(vma))
-                    (!vma->vm_ops || !vma->vm_ops->fault))
                        foll_flags |= FOLL_ANON;
                do {
@@ -1766,7 +1785,6 @@ gotten:
        page_table = pte_offset_map_lock(mm, pmd, address, &ptl);
        if (likely(pte_same(*page_table, orig_pte))) {
                if (old_page) {
-                        page_remove_rmap(old_page, vma);
                        if (!PageAnon(old_page)) {
                                dec_mm_counter(mm, file_rss);
                                inc_mm_counter(mm, anon_rss);
@@ -1788,6 +1806,32 @@ gotten:
                lru_cache_add_active(new_page);
                page_add_new_anon_rmap(new_page, vma, address);
+                if (old_page) {
+                        /*
+                         * Only after switching the pte to the new page may
+                         * we remove the mapcount here. Otherwise another
+                         * process may come and find the rmap count decremented
+                         * before the pte is switched to the new page, and
+                         * "reuse" the old page writing into it while our pte
+                         * here still points into it and can be read by other
+                         * threads.
+                         *
+                         * The critical issue is to order this
+                         * page_remove_rmap with the ptp_clear_flush above.
+                         * Those stores are ordered by (if nothing else,)
+                         * the barrier present in the atomic_add_negative
+                         * in page_remove_rmap.
+                         *
+                         * Then the TLB flush in ptep_clear_flush ensures that
+                         * no process can access the old page before the
+                         * decremented mapcount is visible. And the old page
+                         * cannot be reused until after the decremented
+                         * mapcount is visible. So transitively, TLBs to
+                         * old page will be flushed before it can be reused.
+                         */
+                        page_remove_rmap(old_page, vma);
+                }
                /* Free the old page.. */
                new_page = old_page;
                ret |= VM_FAULT_WRITE;
diff --git a/sound/isa/sb/sb_mixer.c b/sound/isa/sb/sb_mixer.c
index 91d14224f6b3..73d4572d136b 100644
--- a/sound/isa/sb/sb_mixer.c
+++ b/sound/isa/sb/sb_mixer.c
@@ -925,7 +925,7 @@ static unsigned char als4000_saved_regs[] = {
 static void save_mixer(struct snd_sb *chip, unsigned char *regs, int num_regs)
 {
        unsigned char *val = chip->saved_regs;
-        snd_assert(num_regs > ARRAY_SIZE(chip->saved_regs), return);
+        snd_assert(num_regs <= ARRAY_SIZE(chip->saved_regs), return);
        for (; num_regs; num_regs--)
                *val++ = snd_sbmixer_read(chip, *regs++);
 }
@@ -933,7 +933,7 @@ static void save_mixer(struct snd_sb *chip, unsigned char *regs, int num_regs)
 static void restore_mixer(struct snd_sb *chip, unsigned char *regs, int num_regs)
 {
        unsigned char *val = chip->saved_regs;
-        snd_assert(num_regs > ARRAY_SIZE(chip->saved_regs), return);
+        snd_assert(num_regs <= ARRAY_SIZE(chip->saved_regs), return);
        for (; num_regs; num_regs--)
                snd_sbmixer_write(chip, *regs++, *val++);
 }
diff --git a/sound/pci/aw2/aw2-alsa.c b/sound/pci/aw2/aw2-alsa.c
index 56f87cd33c19..3f00ddf450f8 100644
--- a/sound/pci/aw2/aw2-alsa.c
+++ b/sound/pci/aw2/aw2-alsa.c
@@ -316,6 +316,8 @@ static int __devinit snd_aw2_create(struct snd_card *card,
                return -ENOMEM;
        }
+        /* (2) initialization of the chip hardware */
+        snd_aw2_saa7146_setup(&chip->saa7146, chip->iobase_virt);
        if (request_irq(pci->irq, snd_aw2_saa7146_interrupt,
                        IRQF_SHARED, "Audiowerk2", chip)) {
@@ -329,8 +331,6 @@ static int __devinit snd_aw2_create(struct snd_card *card,
        }
        chip->irq = pci->irq;
-        /* (2) initialization of the chip hardware */
-        snd_aw2_saa7146_setup(&chip->saa7146, chip->iobase_virt);
        err = snd_device_new(card, SNDRV_DEV_LOWLEVEL, chip, &ops);
        if (err < 0) {
                free_irq(chip->irq, (void *)chip);
diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c
index 98778cb69c6e..1dcf9f3d1107 100644
--- a/virt/kvm/ioapic.c
+++ b/virt/kvm/ioapic.c
@@ -269,28 +269,9 @@ void kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level)
        }
 }
-static int get_eoi_gsi(struct kvm_ioapic *ioapic, int vector)
+static void __kvm_ioapic_update_eoi(struct kvm_ioapic *ioapic, int gsi)
 {
-        int i;
-        for (i = 0; i < IOAPIC_NUM_PINS; i++)
-                if (ioapic->redirtbl[i].fields.vector == vector)
-                        return i;
-        return -1;
-}
-void kvm_ioapic_update_eoi(struct kvm *kvm, int vector)
-{
-        struct kvm_ioapic *ioapic = kvm->arch.vioapic;
        union ioapic_redir_entry *ent;
-        int gsi;
-        gsi = get_eoi_gsi(ioapic, vector);
-        if (gsi == -1) {
-                printk(KERN_WARNING "Can't find redir item for %d EOI\n",
-                       vector);
-                return;
-        }
        ent = &ioapic->redirtbl[gsi];
        ASSERT(ent->fields.trig_mode == IOAPIC_LEVEL_TRIG);
@@ -300,6 +281,16 @@ void kvm_ioapic_update_eoi(struct kvm *kvm, int vector)
                ioapic_deliver(ioapic, gsi);
 }
+void kvm_ioapic_update_eoi(struct kvm *kvm, int vector)
+{
+        struct kvm_ioapic *ioapic = kvm->arch.vioapic;
+        int i;
+        for (i = 0; i < IOAPIC_NUM_PINS; i++)
+                if (ioapic->redirtbl[i].fields.vector == vector)
+                        __kvm_ioapic_update_eoi(ioapic, i);
+}
 static int ioapic_in_range(struct kvm_io_device *this, gpa_t addr)
 {
        struct kvm_ioapic *ioapic = (struct kvm_ioapic *)this->private;