Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6

13 files changed, 397 insertions, 201 deletions

diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile
index 82ec6075c057..82db7f45e2de 100644
--- a/arch/x86/kernel/cpu/Makefile
+++ b/arch/x86/kernel/cpu/Makefile
@@ -2,8 +2,14 @@
 # Makefile for x86-compatible CPU details and quirks
 #
 
+# Don't trace early stages of a secondary CPU boot
+ifdef CONFIG_FUNCTION_TRACER
+CFLAGS_REMOVE_common.o = -pg
+endif
+
 obj-y                   := intel_cacheinfo.o addon_cpuid_features.o
 obj-y                   += proc.o capflags.o powerflags.o common.o
+obj-y                   += vmware.o hypervisor.o
 
 obj-$(CONFIG_X86_32)    += bugs.o cmpxchg.o
 obj-$(CONFIG_X86_64)    += bugs_64.o
diff --git a/arch/x86/kernel/cpu/addon_cpuid_features.c b/arch/x86/kernel/cpu/addon_cpuid_features.c
index ef8f831af823..2cf23634b6d9 100644
--- a/arch/x86/kernel/cpu/addon_cpuid_features.c
+++ b/arch/x86/kernel/cpu/addon_cpuid_features.c
@@ -120,9 +120,17 @@ void __cpuinit detect_extended_topology(struct cpuinfo_x86 *c)
         c->cpu_core_id = phys_pkg_id(c->initial_apicid, ht_mask_width)
                                                  & core_select_mask;
         c->phys_proc_id = phys_pkg_id(c->initial_apicid, core_plus_mask_width);
+        /*
+         * Reinit the apicid, now that we have extended initial_apicid.
+         */
+        c->apicid = phys_pkg_id(c->initial_apicid, 0);
 #else
         c->cpu_core_id = phys_pkg_id(ht_mask_width) & core_select_mask;
         c->phys_proc_id = phys_pkg_id(core_plus_mask_width);
+        /*
+         * Reinit the apicid, now that we have extended initial_apicid.
+         */
+        c->apicid = phys_pkg_id(0);
 #endif
         c->x86_max_cores = (core_level_siblings / smp_num_siblings);
 
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 8f1e31db2ad5..7c878f6aa919 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -283,9 +283,14 @@ static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
 {
         early_init_amd_mc(c);
 
-        /* c->x86_power is 8000_0007 edx. Bit 8 is constant TSC */
-        if (c->x86_power & (1<<8))
+        /*
+         * c->x86_power is 8000_0007 edx. Bit 8 is TSC runs at constant rate
+         * with P/T states and does not stop in deep C-states
+         */
+        if (c->x86_power & (1 << 8)) {
                 set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
+                set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC);
+        }
 
 #ifdef CONFIG_X86_64
         set_cpu_cap(c, X86_FEATURE_SYSCALL32);
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index b9c9ea0217a9..42e0853030cb 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -36,6 +36,7 @@
 #include <asm/proto.h>
 #include <asm/sections.h>
 #include <asm/setup.h>
+#include <asm/hypervisor.h>
 
 #include "cpu.h"
 
@@ -703,6 +704,7 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
         detect_ht(c);
 #endif
 
+        init_hypervisor(c);
         /*
          * On SMP, boot_cpu_data holds the common feature set between
          * all CPUs; so make sure that we indicate which features are
@@ -862,7 +864,7 @@ EXPORT_SYMBOL(_cpu_pda);
 
 struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table };
 
-char boot_cpu_stack[IRQSTACKSIZE] __page_aligned_bss;
+static char boot_cpu_stack[IRQSTACKSIZE] __page_aligned_bss;
 
 void __cpuinit pda_init(int cpu)
 {
@@ -903,8 +905,8 @@ void __cpuinit pda_init(int cpu)
         }
 }
 
-char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ +
-                           DEBUG_STKSZ] __page_aligned_bss;
+static char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ +
+                                  DEBUG_STKSZ] __page_aligned_bss;
 
 extern asmlinkage void ignore_sysret(void);
 
diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
index 8e48c5d4467d..88ea02dcb622 100644
--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -33,6 +33,7 @@
 #include <linux/cpufreq.h>
 #include <linux/compiler.h>
 #include <linux/dmi.h>
+#include <linux/ftrace.h>
 
 #include <linux/acpi.h>
 #include <acpi/processor.h>
@@ -391,6 +392,7 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy,
         unsigned int next_perf_state = 0; /* Index into perf table */
         unsigned int i;
         int result = 0;
+        struct power_trace it;
 
         dprintk("acpi_cpufreq_target %d (%d)\n", target_freq, policy->cpu);
 
@@ -427,6 +429,8 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy,
                 }
         }
 
+        trace_power_mark(&it, POWER_PSTATE, next_perf_state);
+
         switch (data->cpu_feature) {
         case SYSTEM_INTEL_MSR_CAPABLE:
                 cmd.type = SYSTEM_INTEL_MSR_CAPABLE;
diff --git a/arch/x86/kernel/cpu/hypervisor.c b/arch/x86/kernel/cpu/hypervisor.c
new file mode 100644
index 000000000000..fb5b86af0b01
--- /dev/null
+++ b/arch/x86/kernel/cpu/hypervisor.c
@@ -0,0 +1,58 @@
+/*
+ * Common hypervisor code
+ *
+ * Copyright (C) 2008, VMware, Inc.
+ * Author : Alok N Kataria <akataria@vmware.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (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, GOOD TITLE or
+ * NON INFRINGEMENT.  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 <asm/processor.h>
+#include <asm/vmware.h>
+#include <asm/hypervisor.h>
+
+static inline void __cpuinit
+detect_hypervisor_vendor(struct cpuinfo_x86 *c)
+{
+        if (vmware_platform()) {
+                c->x86_hyper_vendor = X86_HYPER_VENDOR_VMWARE;
+        } else {
+                c->x86_hyper_vendor = X86_HYPER_VENDOR_NONE;
+        }
+}
+
+unsigned long get_hypervisor_tsc_freq(void)
+{
+        if (boot_cpu_data.x86_hyper_vendor == X86_HYPER_VENDOR_VMWARE)
+                return vmware_get_tsc_khz();
+        return 0;
+}
+
+static inline void __cpuinit
+hypervisor_set_feature_bits(struct cpuinfo_x86 *c)
+{
+        if (boot_cpu_data.x86_hyper_vendor == X86_HYPER_VENDOR_VMWARE) {
+                vmware_set_feature_bits(c);
+                return;
+        }
+}
+
+void __cpuinit init_hypervisor(struct cpuinfo_x86 *c)
+{
+        detect_hypervisor_vendor(c);
+        hypervisor_set_feature_bits(c);
+}
diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
index cce0b6118d55..8ea6929e974c 100644
--- a/arch/x86/kernel/cpu/intel.c
+++ b/arch/x86/kernel/cpu/intel.c
@@ -11,7 +11,6 @@
 #include <asm/pgtable.h>
 #include <asm/msr.h>
 #include <asm/uaccess.h>
-#include <asm/ptrace.h>
 #include <asm/ds.h>
 #include <asm/bugs.h>
 
@@ -41,6 +40,16 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c)
         if (c->x86 == 15 && c->x86_cache_alignment == 64)
                 c->x86_cache_alignment = 128;
 #endif
+
+        /*
+         * c->x86_power is 8000_0007 edx. Bit 8 is TSC runs at constant rate
+         * with P/T states and does not stop in deep C-states
+         */
+        if (c->x86_power & (1 << 8)) {
+                set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
+                set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC);
+        }
+
 }
 
 #ifdef CONFIG_X86_32
@@ -242,6 +251,13 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 
         intel_workarounds(c);
 
+        /*
+         * Detect the extended topology information if available. This
+         * will reinitialise the initial_apicid which will be used
+         * in init_intel_cacheinfo()
+         */
+        detect_extended_topology(c);
+
         l2 = init_intel_cacheinfo(c);
         if (c->cpuid_level > 9) {
                 unsigned eax = cpuid_eax(10);
@@ -307,13 +323,8 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
                 set_cpu_cap(c, X86_FEATURE_P4);
         if (c->x86 == 6)
                 set_cpu_cap(c, X86_FEATURE_P3);
-
-        if (cpu_has_bts)
-                ptrace_bts_init_intel(c);
-
 #endif
 
-        detect_extended_topology(c);
         if (!cpu_has(c, X86_FEATURE_XTOPOLOGY)) {
                 /*
                  * let's use the legacy cpuid vector 0x1 and 0x4 for topology
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index 43ea612d3e9d..15cf14e9bf26 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -644,20 +644,17 @@ static inline ssize_t show_shared_cpu_list(struct _cpuid4_info *leaf, char *buf)
         return show_shared_cpu_map_func(leaf, 1, buf);
 }
 
-static ssize_t show_type(struct _cpuid4_info *this_leaf, char *buf) {
-        switch(this_leaf->eax.split.type) {
-            case CACHE_TYPE_DATA:
+static ssize_t show_type(struct _cpuid4_info *this_leaf, char *buf)
+{
+        switch (this_leaf->eax.split.type) {
+        case CACHE_TYPE_DATA:
                 return sprintf(buf, "Data\n");
-                break;
-            case CACHE_TYPE_INST:
+        case CACHE_TYPE_INST:
                 return sprintf(buf, "Instruction\n");
-                break;
-            case CACHE_TYPE_UNIFIED:
+        case CACHE_TYPE_UNIFIED:
                 return sprintf(buf, "Unified\n");
-                break;
-            default:
+        default:
                 return sprintf(buf, "Unknown\n");
-                break;
         }
 }
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c
index 4b031a4ac856..1c838032fd37 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_64.c
@@ -510,12 +510,9 @@ static void __cpuinit mce_cpu_features(struct cpuinfo_x86 *c)
  */
 void __cpuinit mcheck_init(struct cpuinfo_x86 *c)
 {
-        static cpumask_t mce_cpus = CPU_MASK_NONE;
-
         mce_cpu_quirks(c);
 
         if (mce_dont_init ||
-            cpu_test_and_set(smp_processor_id(), mce_cpus) ||
             !mce_available(c))
                 return;
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
index 5eb390a4b2e9..748c8f9e7a05 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
@@ -237,7 +237,7 @@ asmlinkage void mce_threshold_interrupt(void)
                 }
         }
 out:
-        add_pda(irq_threshold_count, 1);
+        inc_irq_stat(irq_threshold_count);
         irq_exit();
 }
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c
index c17eaf5dd6dd..4b48f251fd39 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c
@@ -26,7 +26,7 @@ asmlinkage void smp_thermal_interrupt(void)
         if (therm_throt_process(msr_val & 1))
                 mce_log_therm_throt_event(smp_processor_id(), msr_val);
 
-        add_pda(irq_thermal_count, 1);
+        inc_irq_stat(irq_thermal_count);
         irq_exit();
 }
 
diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index c78c04821ea1..1159e269e596 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -803,6 +803,7 @@ x86_get_mtrr_mem_range(struct res_range *range, int nr_range,
 }
 
 static struct res_range __initdata range[RANGE_NUM];
+static int __initdata nr_range;
 
 #ifdef CONFIG_MTRR_SANITIZER
 
@@ -1206,39 +1207,43 @@ struct mtrr_cleanup_result {
 #define PSHIFT          (PAGE_SHIFT - 10)
 
 static struct mtrr_cleanup_result __initdata result[NUM_RESULT];
-static struct res_range __initdata range_new[RANGE_NUM];
 static unsigned long __initdata min_loss_pfn[RANGE_NUM];
 
-static int __init mtrr_cleanup(unsigned address_bits)
+static void __init print_out_mtrr_range_state(void)
 {
-        unsigned long extra_remove_base, extra_remove_size;
-        unsigned long base, size, def, dummy;
-        mtrr_type type;
-        int nr_range, nr_range_new;
-        u64 chunk_size, gran_size;
-        unsigned long range_sums, range_sums_new;
-        int index_good;
-        int num_reg_good;
         int i;
+        char start_factor = 'K', size_factor = 'K';
+        unsigned long start_base, size_base;
+        mtrr_type type;
 
-        /* extra one for all 0 */
-        int num[MTRR_NUM_TYPES + 1];
+        for (i = 0; i < num_var_ranges; i++) {
 
-        if (!is_cpu(INTEL) || enable_mtrr_cleanup < 1)
-                return 0;
-        rdmsr(MTRRdefType_MSR, def, dummy);
-        def &= 0xff;
-        if (def != MTRR_TYPE_UNCACHABLE)
-                return 0;
+                size_base = range_state[i].size_pfn << (PAGE_SHIFT - 10);
+                if (!size_base)
+                        continue;
 
-        /* get it and store it aside */
-        memset(range_state, 0, sizeof(range_state));
-        for (i = 0; i < num_var_ranges; i++) {
-                mtrr_if->get(i, &base, &size, &type);
-                range_state[i].base_pfn = base;
-                range_state[i].size_pfn = size;
-                range_state[i].type = type;
+                size_base = to_size_factor(size_base, &size_factor),
+                start_base = range_state[i].base_pfn << (PAGE_SHIFT - 10);
+                start_base = to_size_factor(start_base, &start_factor),
+                type = range_state[i].type;
+
+                printk(KERN_DEBUG "reg %d, base: %ld%cB, range: %ld%cB, type %s\n",
+                        i, start_base, start_factor,
+                        size_base, size_factor,
+                        (type == MTRR_TYPE_UNCACHABLE) ? "UC" :
+                            ((type == MTRR_TYPE_WRPROT) ? "WP" :
+                             ((type == MTRR_TYPE_WRBACK) ? "WB" : "Other"))
+                        );
         }
+}
+
+static int __init mtrr_need_cleanup(void)
+{
+        int i;
+        mtrr_type type;
+        unsigned long size;
+        /* extra one for all 0 */
+        int num[MTRR_NUM_TYPES + 1];
 
         /* check entries number */
         memset(num, 0, sizeof(num));
@@ -1263,29 +1268,133 @@ static int __init mtrr_cleanup(unsigned address_bits)
                 num_var_ranges - num[MTRR_NUM_TYPES])
                 return 0;
 
-        /* print original var MTRRs at first, for debugging: */
-        printk(KERN_DEBUG "original variable MTRRs\n");
-        for (i = 0; i < num_var_ranges; i++) {
-                char start_factor = 'K', size_factor = 'K';
-                unsigned long start_base, size_base;
+        return 1;
+}
 
-                size_base = range_state[i].size_pfn << (PAGE_SHIFT - 10);
-                if (!size_base)
-                        continue;
+static unsigned long __initdata range_sums;
+static void __init mtrr_calc_range_state(u64 chunk_size, u64 gran_size,
+                                         unsigned long extra_remove_base,
+                                         unsigned long extra_remove_size,
+                                         int i)
+{
+        int num_reg;
+        static struct res_range range_new[RANGE_NUM];
+        static int nr_range_new;
+        unsigned long range_sums_new;
+
+        /* convert ranges to var ranges state */
+        num_reg = x86_setup_var_mtrrs(range, nr_range,
+                                                chunk_size, gran_size);
+
+        /* we got new setting in range_state, check it */
+        memset(range_new, 0, sizeof(range_new));
+        nr_range_new = x86_get_mtrr_mem_range(range_new, 0,
+                                extra_remove_base, extra_remove_size);
+        range_sums_new = sum_ranges(range_new, nr_range_new);
+
+        result[i].chunk_sizek = chunk_size >> 10;
+        result[i].gran_sizek = gran_size >> 10;
+        result[i].num_reg = num_reg;
+        if (range_sums < range_sums_new) {
+                result[i].lose_cover_sizek =
+                        (range_sums_new - range_sums) << PSHIFT;
+                result[i].bad = 1;
+        } else
+                result[i].lose_cover_sizek =
+                        (range_sums - range_sums_new) << PSHIFT;
 
-                size_base = to_size_factor(size_base, &size_factor),
-                start_base = range_state[i].base_pfn << (PAGE_SHIFT - 10);
-                start_base = to_size_factor(start_base, &start_factor),
-                type = range_state[i].type;
+        /* double check it */
+        if (!result[i].bad && !result[i].lose_cover_sizek) {
+                if (nr_range_new != nr_range ||
+                        memcmp(range, range_new, sizeof(range)))
+                                result[i].bad = 1;
+        }
 
-                printk(KERN_DEBUG "reg %d, base: %ld%cB, range: %ld%cB, type %s\n",
-                        i, start_base, start_factor,
-                        size_base, size_factor,
-                        (type == MTRR_TYPE_UNCACHABLE) ? "UC" :
-                            ((type == MTRR_TYPE_WRPROT) ? "WP" :
-                             ((type == MTRR_TYPE_WRBACK) ? "WB" : "Other"))
-                        );
+        if (!result[i].bad && (range_sums - range_sums_new <
+                                min_loss_pfn[num_reg])) {
+                min_loss_pfn[num_reg] =
+                        range_sums - range_sums_new;
         }
+}
+
+static void __init mtrr_print_out_one_result(int i)
+{
+        char gran_factor, chunk_factor, lose_factor;
+        unsigned long gran_base, chunk_base, lose_base;
+
+        gran_base = to_size_factor(result[i].gran_sizek, &gran_factor),
+        chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor),
+        lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor),
+        printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t",
+                        result[i].bad ? "*BAD*" : " ",
+                        gran_base, gran_factor, chunk_base, chunk_factor);
+        printk(KERN_CONT "num_reg: %d  \tlose cover RAM: %s%ld%c\n",
+                        result[i].num_reg, result[i].bad ? "-" : "",
+                        lose_base, lose_factor);
+}
+
+static int __init mtrr_search_optimal_index(void)
+{
+        int i;
+        int num_reg_good;
+        int index_good;
+
+        if (nr_mtrr_spare_reg >= num_var_ranges)
+                nr_mtrr_spare_reg = num_var_ranges - 1;
+        num_reg_good = -1;
+        for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) {
+                if (!min_loss_pfn[i])
+                        num_reg_good = i;
+        }
+
+        index_good = -1;
+        if (num_reg_good != -1) {
+                for (i = 0; i < NUM_RESULT; i++) {
+                        if (!result[i].bad &&
+                            result[i].num_reg == num_reg_good &&
+                            !result[i].lose_cover_sizek) {
+                                index_good = i;
+                                break;
+                        }
+                }
+        }
+
+        return index_good;
+}
+
+
+static int __init mtrr_cleanup(unsigned address_bits)
+{
+        unsigned long extra_remove_base, extra_remove_size;
+        unsigned long base, size, def, dummy;
+        mtrr_type type;
+        u64 chunk_size, gran_size;
+        int index_good;
+        int i;
+
+        if (!is_cpu(INTEL) || enable_mtrr_cleanup < 1)
+                return 0;
+        rdmsr(MTRRdefType_MSR, def, dummy);
+        def &= 0xff;
+        if (def != MTRR_TYPE_UNCACHABLE)
+                return 0;
+
+        /* get it and store it aside */
+        memset(range_state, 0, sizeof(range_state));
+        for (i = 0; i < num_var_ranges; i++) {
+                mtrr_if->get(i, &base, &size, &type);
+                range_state[i].base_pfn = base;
+                range_state[i].size_pfn = size;
+                range_state[i].type = type;
+        }
+
+        /* check if we need handle it and can handle it */
+        if (!mtrr_need_cleanup())
+                return 0;
+
+        /* print original var MTRRs at first, for debugging: */
+        printk(KERN_DEBUG "original variable MTRRs\n");
+        print_out_mtrr_range_state();
 
         memset(range, 0, sizeof(range));
         extra_remove_size = 0;
@@ -1309,176 +1418,64 @@ static int __init mtrr_cleanup(unsigned address_bits)
                range_sums >> (20 - PAGE_SHIFT));
 
         if (mtrr_chunk_size && mtrr_gran_size) {
-                int num_reg;
-                char gran_factor, chunk_factor, lose_factor;
-                unsigned long gran_base, chunk_base, lose_base;
-
-                debug_print++;
-                /* convert ranges to var ranges state */
-                num_reg = x86_setup_var_mtrrs(range, nr_range, mtrr_chunk_size,
-                                              mtrr_gran_size);
+                i = 0;
+                mtrr_calc_range_state(mtrr_chunk_size, mtrr_gran_size,
+                                      extra_remove_base, extra_remove_size, i);
 
-                /* we got new setting in range_state, check it */
-                memset(range_new, 0, sizeof(range_new));
-                nr_range_new = x86_get_mtrr_mem_range(range_new, 0,
-                                                      extra_remove_base,
-                                                      extra_remove_size);
-                range_sums_new = sum_ranges(range_new, nr_range_new);
+                mtrr_print_out_one_result(i);
 
-                i = 0;
-                result[i].chunk_sizek = mtrr_chunk_size >> 10;
-                result[i].gran_sizek = mtrr_gran_size >> 10;
-                result[i].num_reg = num_reg;
-                if (range_sums < range_sums_new) {
-                        result[i].lose_cover_sizek =
-                                (range_sums_new - range_sums) << PSHIFT;
-                        result[i].bad = 1;
-                } else
-                        result[i].lose_cover_sizek =
-                                (range_sums - range_sums_new) << PSHIFT;
-
-                gran_base = to_size_factor(result[i].gran_sizek, &gran_factor),
-                chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor),
-                lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor),
-                printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t",
-                         result[i].bad?"*BAD*":" ",
-                         gran_base, gran_factor, chunk_base, chunk_factor);
-                printk(KERN_CONT "num_reg: %d  \tlose cover RAM: %s%ld%c\n",
-                         result[i].num_reg, result[i].bad?"-":"",
-                         lose_base, lose_factor);
                 if (!result[i].bad) {
                         set_var_mtrr_all(address_bits);
                         return 1;
                 }
                 printk(KERN_INFO "invalid mtrr_gran_size or mtrr_chunk_size, "
                        "will find optimal one\n");
-                debug_print--;
-                memset(result, 0, sizeof(result[0]));
         }
 
         i = 0;
         memset(min_loss_pfn, 0xff, sizeof(min_loss_pfn));
         memset(result, 0, sizeof(result));
         for (gran_size = (1ULL<<16); gran_size < (1ULL<<32); gran_size <<= 1) {
-                char gran_factor;
-                unsigned long gran_base;
-
-                if (debug_print)
-                        gran_base = to_size_factor(gran_size >> 10, &gran_factor);
 
                 for (chunk_size = gran_size; chunk_size < (1ULL<<32);
                      chunk_size <<= 1) {
-                        int num_reg;
 
-                        if (debug_print) {
-                                char chunk_factor;
-                                unsigned long chunk_base;
-
-                                chunk_base = to_size_factor(chunk_size>>10, &chunk_factor),
-                                printk(KERN_INFO "\n");
-                                printk(KERN_INFO "gran_size: %ld%c   chunk_size: %ld%c \n",
-                                       gran_base, gran_factor, chunk_base, chunk_factor);
-                        }
                         if (i >= NUM_RESULT)
                                 continue;
 
-                        /* convert ranges to var ranges state */
-                        num_reg = x86_setup_var_mtrrs(range, nr_range,
-                                                         chunk_size, gran_size);
-
-                        /* we got new setting in range_state, check it */
-                        memset(range_new, 0, sizeof(range_new));
-                        nr_range_new = x86_get_mtrr_mem_range(range_new, 0,
-                                         extra_remove_base, extra_remove_size);
-                        range_sums_new = sum_ranges(range_new, nr_range_new);
-
-                        result[i].chunk_sizek = chunk_size >> 10;
-                        result[i].gran_sizek = gran_size >> 10;
-                        result[i].num_reg = num_reg;
-                        if (range_sums < range_sums_new) {
-                                result[i].lose_cover_sizek =
-                                        (range_sums_new - range_sums) << PSHIFT;
-                                result[i].bad = 1;
-                        } else
-                                result[i].lose_cover_sizek =
-                                        (range_sums - range_sums_new) << PSHIFT;
-
-                        /* double check it */
-                        if (!result[i].bad && !result[i].lose_cover_sizek) {
-                                if (nr_range_new != nr_range ||
-                                        memcmp(range, range_new, sizeof(range)))
-                                                result[i].bad = 1;
+                        mtrr_calc_range_state(chunk_size, gran_size,
+                                      extra_remove_base, extra_remove_size, i);
+                        if (debug_print) {
+                                mtrr_print_out_one_result(i);
+                                printk(KERN_INFO "\n");
                         }
 
-                        if (!result[i].bad && (range_sums - range_sums_new <
-                                               min_loss_pfn[num_reg])) {
-                                min_loss_pfn[num_reg] =
-                                        range_sums - range_sums_new;
-                        }
                         i++;
                 }
         }
 
-        /* print out all */
-        for (i = 0; i < NUM_RESULT; i++) {
-                char gran_factor, chunk_factor, lose_factor;
-                unsigned long gran_base, chunk_base, lose_base;
-
-                gran_base = to_size_factor(result[i].gran_sizek, &gran_factor),
-                chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor),
-                lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor),
-                printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t",
-                         result[i].bad?"*BAD*":" ",
-                         gran_base, gran_factor, chunk_base, chunk_factor);
-                printk(KERN_CONT "num_reg: %d  \tlose cover RAM: %s%ld%c\n",
-                         result[i].num_reg, result[i].bad?"-":"",
-                         lose_base, lose_factor);
-        }
-
         /* try to find the optimal index */
-        if (nr_mtrr_spare_reg >= num_var_ranges)
-                nr_mtrr_spare_reg = num_var_ranges - 1;
-        num_reg_good = -1;
-        for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) {
-                if (!min_loss_pfn[i])
-                        num_reg_good = i;
-        }
-
-        index_good = -1;
-        if (num_reg_good != -1) {
-                for (i = 0; i < NUM_RESULT; i++) {
-                        if (!result[i].bad &&
-                            result[i].num_reg == num_reg_good &&
-                            !result[i].lose_cover_sizek) {
-                                index_good = i;
-                                break;
-                        }
-                }
-        }
+        index_good = mtrr_search_optimal_index();
 
         if (index_good != -1) {
-                char gran_factor, chunk_factor, lose_factor;
-                unsigned long gran_base, chunk_base, lose_base;
-
                 printk(KERN_INFO "Found optimal setting for mtrr clean up\n");
                 i = index_good;
-                gran_base = to_size_factor(result[i].gran_sizek, &gran_factor),
-                chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor),
-                lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor),
-                printk(KERN_INFO "gran_size: %ld%c \tchunk_size: %ld%c \t",
-                         gran_base, gran_factor, chunk_base, chunk_factor);
-                printk(KERN_CONT "num_reg: %d  \tlose RAM: %ld%c\n",
-                         result[i].num_reg, lose_base, lose_factor);
+                mtrr_print_out_one_result(i);
+
                 /* convert ranges to var ranges state */
                 chunk_size = result[i].chunk_sizek;
                 chunk_size <<= 10;
                 gran_size = result[i].gran_sizek;
                 gran_size <<= 10;
-                debug_print++;
                 x86_setup_var_mtrrs(range, nr_range, chunk_size, gran_size);
-                debug_print--;
                 set_var_mtrr_all(address_bits);
+                printk(KERN_DEBUG "New variable MTRRs\n");
+                print_out_mtrr_range_state();
                 return 1;
+        } else {
+                /* print out all */
+                for (i = 0; i < NUM_RESULT; i++)
+                        mtrr_print_out_one_result(i);
         }
 
         printk(KERN_INFO "mtrr_cleanup: can not find optimal value\n");
@@ -1562,7 +1559,6 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn)
 {
         unsigned long i, base, size, highest_pfn = 0, def, dummy;
         mtrr_type type;
-        int nr_range;
         u64 total_trim_size;
 
         /* extra one for all 0 */
diff --git a/arch/x86/kernel/cpu/vmware.c b/arch/x86/kernel/cpu/vmware.c
new file mode 100644
index 000000000000..284c399e3234
--- /dev/null
+++ b/arch/x86/kernel/cpu/vmware.c
@@ -0,0 +1,112 @@
+/*
+ * VMware Detection code.
+ *
+ * Copyright (C) 2008, VMware, Inc.
+ * Author : Alok N Kataria <akataria@vmware.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (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, GOOD TITLE or
+ * NON INFRINGEMENT.  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/dmi.h>
+#include <asm/div64.h>
+#include <asm/vmware.h>
+
+#define CPUID_VMWARE_INFO_LEAF  0x40000000
+#define VMWARE_HYPERVISOR_MAGIC 0x564D5868
+#define VMWARE_HYPERVISOR_PORT  0x5658
+
+#define VMWARE_PORT_CMD_GETVERSION      10
+#define VMWARE_PORT_CMD_GETHZ           45
+
+#define VMWARE_PORT(cmd, eax, ebx, ecx, edx)                            \
+        __asm__("inl (%%dx)" :                                          \
+                        "=a"(eax), "=c"(ecx), "=d"(edx), "=b"(ebx) :    \
+                        "0"(VMWARE_HYPERVISOR_MAGIC),                   \
+                        "1"(VMWARE_PORT_CMD_##cmd),                     \
+                        "2"(VMWARE_HYPERVISOR_PORT), "3"(UINT_MAX) :    \
+                        "memory");
+
+static inline int __vmware_platform(void)
+{
+        uint32_t eax, ebx, ecx, edx;
+        VMWARE_PORT(GETVERSION, eax, ebx, ecx, edx);
+        return eax != (uint32_t)-1 && ebx == VMWARE_HYPERVISOR_MAGIC;
+}
+
+static unsigned long __vmware_get_tsc_khz(void)
+{
+        uint64_t tsc_hz;
+        uint32_t eax, ebx, ecx, edx;
+
+        VMWARE_PORT(GETHZ, eax, ebx, ecx, edx);
+
+        if (ebx == UINT_MAX)
+                return 0;
+        tsc_hz = eax | (((uint64_t)ebx) << 32);
+        do_div(tsc_hz, 1000);
+        BUG_ON(tsc_hz >> 32);
+        return tsc_hz;
+}
+
+/*
+ * While checking the dmi string infomation, just checking the product
+ * serial key should be enough, as this will always have a VMware
+ * specific string when running under VMware hypervisor.
+ */
+int vmware_platform(void)
+{
+        if (cpu_has_hypervisor) {
+                unsigned int eax, ebx, ecx, edx;
+                char hyper_vendor_id[13];
+
+                cpuid(CPUID_VMWARE_INFO_LEAF, &eax, &ebx, &ecx, &edx);
+                memcpy(hyper_vendor_id + 0, &ebx, 4);
+                memcpy(hyper_vendor_id + 4, &ecx, 4);
+                memcpy(hyper_vendor_id + 8, &edx, 4);
+                hyper_vendor_id[12] = '\0';
+                if (!strcmp(hyper_vendor_id, "VMwareVMware"))
+                        return 1;
+        } else if (dmi_available && dmi_name_in_serial("VMware") &&
+                   __vmware_platform())
+                return 1;
+
+        return 0;
+}
+
+unsigned long vmware_get_tsc_khz(void)
+{
+        BUG_ON(!vmware_platform());
+        return __vmware_get_tsc_khz();
+}
+
+/*
+ * VMware hypervisor takes care of exporting a reliable TSC to the guest.
+ * Still, due to timing difference when running on virtual cpus, the TSC can
+ * be marked as unstable in some cases. For example, the TSC sync check at
+ * bootup can fail due to a marginal offset between vcpus' TSCs (though the
+ * TSCs do not drift from each other).  Also, the ACPI PM timer clocksource
+ * is not suitable as a watchdog when running on a hypervisor because the
+ * kernel may miss a wrap of the counter if the vcpu is descheduled for a
+ * long time. To skip these checks at runtime we set these capability bits,
+ * so that the kernel could just trust the hypervisor with providing a
+ * reliable virtual TSC that is suitable for timekeeping.
+ */
+void __cpuinit vmware_set_feature_bits(struct cpuinfo_x86 *c)
+{
+        set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
+        set_cpu_cap(c, X86_FEATURE_TSC_RELIABLE);
+}