8 files changed, 146 insertions, 40 deletions

diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index cae9cabc3031..18514ed26104 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -31,6 +31,11 @@ static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
                 if (c->x86_power & (1<<8))
                         set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
         }
+
+        /*  Set MTRR capability flag if appropriate */
+        if (c->x86_model == 13 || c->x86_model == 9 ||
+           (c->x86_model == 8 && c->x86_mask >= 8))
+                set_cpu_cap(c, X86_FEATURE_K6_MTRR);
 }
 
 static void __cpuinit init_amd(struct cpuinfo_x86 *c)
@@ -166,10 +171,6 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
                                                 mbytes);
                                 }
 
-                                /*  Set MTRR capability flag if appropriate */
-                                if (c->x86_model == 13 || c->x86_model == 9 ||
-                                   (c->x86_model == 8 && c->x86_mask >= 8))
-                                        set_cpu_cap(c, X86_FEATURE_K6_MTRR);
                                 break;
                         }
 
diff --git a/arch/x86/kernel/cpu/centaur.c b/arch/x86/kernel/cpu/centaur.c
index e0f45edd6a55..a0534c04d38a 100644
--- a/arch/x86/kernel/cpu/centaur.c
+++ b/arch/x86/kernel/cpu/centaur.c
@@ -314,6 +314,16 @@ enum {
                 EAMD3D          = 1<<20,
 };
 
+static void __cpuinit early_init_centaur(struct cpuinfo_x86 *c)
+{
+        switch (c->x86) {
+        case 5:
+                /* Emulate MTRRs using Centaur's MCR. */
+                set_cpu_cap(c, X86_FEATURE_CENTAUR_MCR);
+                break;
+        }
+}
+
 static void __cpuinit init_centaur(struct cpuinfo_x86 *c)
 {
 
@@ -462,6 +472,7 @@ centaur_size_cache(struct cpuinfo_x86 *c, unsigned int size)
 static struct cpu_dev centaur_cpu_dev __cpuinitdata = {
         .c_vendor       = "Centaur",
         .c_ident        = { "CentaurHauls" },
+        .c_early_init   = early_init_centaur,
         .c_init         = init_centaur,
         .c_size_cache   = centaur_size_cache,
 };
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 80ab20d4fa39..8aab8517642e 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -13,6 +13,7 @@
 #include <asm/mtrr.h>
 #include <asm/mce.h>
 #include <asm/pat.h>
+#include <asm/asm.h>
 #ifdef CONFIG_X86_LOCAL_APIC
 #include <asm/mpspec.h>
 #include <asm/apic.h>
@@ -334,11 +335,40 @@ static void __init early_cpu_detect(void)
 
         get_cpu_vendor(c, 1);
 
+        early_get_cap(c);
+
         if (c->x86_vendor != X86_VENDOR_UNKNOWN &&
             cpu_devs[c->x86_vendor]->c_early_init)
                 cpu_devs[c->x86_vendor]->c_early_init(c);
+}
 
-        early_get_cap(c);
+/*
+ * The NOPL instruction is supposed to exist on all CPUs with
+ * family >= 6, unfortunately, that's not true in practice because
+ * of early VIA chips and (more importantly) broken virtualizers that
+ * are not easy to detect.  Hence, probe for it based on first
+ * principles.
+ */
+static void __cpuinit detect_nopl(struct cpuinfo_x86 *c)
+{
+        const u32 nopl_signature = 0x888c53b1; /* Random number */
+        u32 has_nopl = nopl_signature;
+
+        clear_cpu_cap(c, X86_FEATURE_NOPL);
+        if (c->x86 >= 6) {
+                asm volatile("\n"
+                             "1:      .byte 0x0f,0x1f,0xc0\n" /* nopl %eax */
+                             "2:\n"
+                             "        .section .fixup,\"ax\"\n"
+                             "3:      xor %0,%0\n"
+                             "        jmp 2b\n"
+                             "        .previous\n"
+                             _ASM_EXTABLE(1b,3b)
+                             : "+a" (has_nopl));
+
+                if (has_nopl == nopl_signature)
+                        set_cpu_cap(c, X86_FEATURE_NOPL);
+        }
 }
 
 static void __cpuinit generic_identify(struct cpuinfo_x86 *c)
@@ -395,8 +425,8 @@ static void __cpuinit generic_identify(struct cpuinfo_x86 *c)
                 }
 
                 init_scattered_cpuid_features(c);
+                detect_nopl(c);
         }
-
 }
 
 static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
diff --git a/arch/x86/kernel/cpu/common_64.c b/arch/x86/kernel/cpu/common_64.c
index dd6e3f15017e..a11f5d4477cd 100644
--- a/arch/x86/kernel/cpu/common_64.c
+++ b/arch/x86/kernel/cpu/common_64.c
@@ -18,6 +18,7 @@
 #include <asm/mtrr.h>
 #include <asm/mce.h>
 #include <asm/pat.h>
+#include <asm/asm.h>
 #include <asm/numa.h>
 #ifdef CONFIG_X86_LOCAL_APIC
 #include <asm/mpspec.h>
@@ -215,6 +216,39 @@ static void __init early_cpu_support_print(void)
         }
 }
 
+/*
+ * The NOPL instruction is supposed to exist on all CPUs with
+ * family >= 6, unfortunately, that's not true in practice because
+ * of early VIA chips and (more importantly) broken virtualizers that
+ * are not easy to detect.  Hence, probe for it based on first
+ * principles.
+ *
+ * Note: no 64-bit chip is known to lack these, but put the code here
+ * for consistency with 32 bits, and to make it utterly trivial to
+ * diagnose the problem should it ever surface.
+ */
+static void __cpuinit detect_nopl(struct cpuinfo_x86 *c)
+{
+        const u32 nopl_signature = 0x888c53b1; /* Random number */
+        u32 has_nopl = nopl_signature;
+
+        clear_cpu_cap(c, X86_FEATURE_NOPL);
+        if (c->x86 >= 6) {
+                asm volatile("\n"
+                             "1:      .byte 0x0f,0x1f,0xc0\n" /* nopl %eax */
+                             "2:\n"
+                             "        .section .fixup,\"ax\"\n"
+                             "3:      xor %0,%0\n"
+                             "        jmp 2b\n"
+                             "        .previous\n"
+                             _ASM_EXTABLE(1b,3b)
+                             : "+a" (has_nopl));
+
+                if (has_nopl == nopl_signature)
+                        set_cpu_cap(c, X86_FEATURE_NOPL);
+        }
+}
+
 static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c);
 
 void __init early_cpu_init(void)
@@ -313,6 +347,8 @@ static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
                 c->x86_phys_bits = eax & 0xff;
         }
 
+        detect_nopl(c);
+
         if (c->x86_vendor != X86_VENDOR_UNKNOWN &&
             cpu_devs[c->x86_vendor]->c_early_init)
                 cpu_devs[c->x86_vendor]->c_early_init(c);
@@ -493,17 +529,20 @@ void pda_init(int cpu)
                 /* others are initialized in smpboot.c */
                 pda->pcurrent = &init_task;
                 pda->irqstackptr = boot_cpu_stack;
+                pda->irqstackptr += IRQSTACKSIZE - 64;
         } else {
-                pda->irqstackptr = (char *)
-                        __get_free_pages(GFP_ATOMIC, IRQSTACK_ORDER);
-                if (!pda->irqstackptr)
-                        panic("cannot allocate irqstack for cpu %d", cpu);
+                if (!pda->irqstackptr) {
+                        pda->irqstackptr = (char *)
+                                __get_free_pages(GFP_ATOMIC, IRQSTACK_ORDER);
+                        if (!pda->irqstackptr)
+                                panic("cannot allocate irqstack for cpu %d",
+                                      cpu);
+                        pda->irqstackptr += IRQSTACKSIZE - 64;
+                }
 
                 if (pda->nodenumber == 0 && cpu_to_node(cpu) != NUMA_NO_NODE)
                         pda->nodenumber = cpu_to_node(cpu);
         }
-
-        pda->irqstackptr += IRQSTACKSIZE-64;
 }
 
 char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ +
@@ -601,19 +640,22 @@ void __cpuinit cpu_init(void)
         /*
          * set up and load the per-CPU TSS
          */
-        for (v = 0; v < N_EXCEPTION_STACKS; v++) {
+        if (!orig_ist->ist[0]) {
                 static const unsigned int order[N_EXCEPTION_STACKS] = {
-                        [0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER,
-                        [DEBUG_STACK - 1] = DEBUG_STACK_ORDER
+                  [0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER,
+                  [DEBUG_STACK - 1] = DEBUG_STACK_ORDER
                 };
-                if (cpu) {
-                        estacks = (char *)__get_free_pages(GFP_ATOMIC, order[v]);
-                        if (!estacks)
-                                panic("Cannot allocate exception stack %ld %d\n",
-                                      v, cpu);
+                for (v = 0; v < N_EXCEPTION_STACKS; v++) {
+                        if (cpu) {
+                                estacks = (char *)__get_free_pages(GFP_ATOMIC, order[v]);
+                                if (!estacks)
+                                        panic("Cannot allocate exception "
+                                              "stack %ld %d\n", v, cpu);
+                        }
+                        estacks += PAGE_SIZE << order[v];
+                        orig_ist->ist[v] = t->x86_tss.ist[v] =
+                                        (unsigned long)estacks;
                 }
-                estacks += PAGE_SIZE << order[v];
-                orig_ist->ist[v] = t->x86_tss.ist[v] = (unsigned long)estacks;
         }
 
         t->x86_tss.io_bitmap_base = offsetof(struct tss_struct, io_bitmap);
diff --git a/arch/x86/kernel/cpu/cyrix.c b/arch/x86/kernel/cpu/cyrix.c
index e710a21bb6e8..898a5a2002ed 100644
--- a/arch/x86/kernel/cpu/cyrix.c
+++ b/arch/x86/kernel/cpu/cyrix.c
@@ -15,13 +15,11 @@
 /*
  * Read NSC/Cyrix DEVID registers (DIR) to get more detailed info. about the CPU
  */
-static void __cpuinit do_cyrix_devid(unsigned char *dir0, unsigned char *dir1)
+static void __cpuinit __do_cyrix_devid(unsigned char *dir0, unsigned char *dir1)
 {
         unsigned char ccr2, ccr3;
-        unsigned long flags;
 
         /* we test for DEVID by checking whether CCR3 is writable */
-        local_irq_save(flags);
         ccr3 = getCx86(CX86_CCR3);
         setCx86(CX86_CCR3, ccr3 ^ 0x80);
         getCx86(0xc0);   /* dummy to change bus */
@@ -44,9 +42,16 @@ static void __cpuinit do_cyrix_devid(unsigned char *dir0, unsigned char *dir1)
                 *dir0 = getCx86(CX86_DIR0);
                 *dir1 = getCx86(CX86_DIR1);
         }
-        local_irq_restore(flags);
 }
 
+static void __cpuinit do_cyrix_devid(unsigned char *dir0, unsigned char *dir1)
+{
+        unsigned long flags;
+
+        local_irq_save(flags);
+        __do_cyrix_devid(dir0, dir1);
+        local_irq_restore(flags);
+}
 /*
  * Cx86_dir0_msb is a HACK needed by check_cx686_cpuid/slop in bugs.h in
  * order to identify the Cyrix CPU model after we're out of setup.c
@@ -161,6 +166,24 @@ static void __cpuinit geode_configure(void)
         local_irq_restore(flags);
 }
 
+static void __cpuinit early_init_cyrix(struct cpuinfo_x86 *c)
+{
+        unsigned char dir0, dir0_msn, dir1 = 0;
+
+        __do_cyrix_devid(&dir0, &dir1);
+        dir0_msn = dir0 >> 4; /* identifies CPU "family"   */
+
+        switch (dir0_msn) {
+        case 3: /* 6x86/6x86L */
+                /* Emulate MTRRs using Cyrix's ARRs. */
+                set_cpu_cap(c, X86_FEATURE_CYRIX_ARR);
+                break;
+        case 5: /* 6x86MX/M II */
+                /* Emulate MTRRs using Cyrix's ARRs. */
+                set_cpu_cap(c, X86_FEATURE_CYRIX_ARR);
+                break;
+        }
+}
 
 static void __cpuinit init_cyrix(struct cpuinfo_x86 *c)
 {
@@ -416,6 +439,7 @@ static void __cpuinit cyrix_identify(struct cpuinfo_x86 *c)
 static struct cpu_dev cyrix_cpu_dev __cpuinitdata = {
         .c_vendor       = "Cyrix",
         .c_ident        = { "CyrixInstead" },
+        .c_early_init   = early_init_cyrix,
         .c_init         = init_cyrix,
         .c_identify     = cyrix_identify,
 };
diff --git a/arch/x86/kernel/cpu/feature_names.c b/arch/x86/kernel/cpu/feature_names.c
index e43ad4ad4cba..c9017799497c 100644
--- a/arch/x86/kernel/cpu/feature_names.c
+++ b/arch/x86/kernel/cpu/feature_names.c
@@ -39,7 +39,8 @@ const char * const x86_cap_flags[NCAPINTS*32] = {
         NULL, NULL, NULL, NULL,
         "constant_tsc", "up", NULL, "arch_perfmon",
         "pebs", "bts", NULL, NULL,
-        "rep_good", NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+        "rep_good", NULL, NULL, NULL,
+        "nopl", NULL, NULL, NULL,
         NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
 
         /* Intel-defined (#2) */
diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c
index 65a339678ece..726a5fcdf341 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_64.c
@@ -759,6 +759,7 @@ static struct sysdev_class mce_sysclass = {
 };
 
 DEFINE_PER_CPU(struct sys_device, device_mce);
+void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu) __cpuinitdata;
 
 /* Why are there no generic functions for this? */
 #define ACCESSOR(name, var, start) \
@@ -883,9 +884,13 @@ static int __cpuinit mce_cpu_callback(struct notifier_block *nfb,
         case CPU_ONLINE:
         case CPU_ONLINE_FROZEN:
                 mce_create_device(cpu);
+                if (threshold_cpu_callback)
+                        threshold_cpu_callback(action, cpu);
                 break;
         case CPU_DEAD:
         case CPU_DEAD_FROZEN:
+                if (threshold_cpu_callback)
+                        threshold_cpu_callback(action, cpu);
                 mce_remove_device(cpu);
                 break;
         }
diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
index 88736cadbaa6..5eb390a4b2e9 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
@@ -628,6 +628,7 @@ static void threshold_remove_bank(unsigned int cpu, int bank)
         deallocate_threshold_block(cpu, bank);
 
 free_out:
+        kobject_del(b->kobj);
         kobject_put(b->kobj);
         kfree(b);
         per_cpu(threshold_banks, cpu)[bank] = NULL;
@@ -645,14 +646,11 @@ static void threshold_remove_device(unsigned int cpu)
 }
 
 /* get notified when a cpu comes on/off */
-static int __cpuinit threshold_cpu_callback(struct notifier_block *nfb,
-                                            unsigned long action, void *hcpu)
+static void __cpuinit amd_64_threshold_cpu_callback(unsigned long action,
+                                                     unsigned int cpu)
 {
-        /* cpu was unsigned int to begin with */
-        unsigned int cpu = (unsigned long)hcpu;
-
         if (cpu >= NR_CPUS)
-                goto out;
+                return;
 
         switch (action) {
         case CPU_ONLINE:
@@ -666,14 +664,8 @@ static int __cpuinit threshold_cpu_callback(struct notifier_block *nfb,
         default:
                 break;
         }
-      out:
-        return NOTIFY_OK;
 }
 
-static struct notifier_block threshold_cpu_notifier __cpuinitdata = {
-        .notifier_call = threshold_cpu_callback,
-};
-
 static __init int threshold_init_device(void)
 {
         unsigned lcpu = 0;
@@ -684,7 +676,7 @@ static __init int threshold_init_device(void)
                 if (err)
                         return err;
         }
-        register_hotcpu_notifier(&threshold_cpu_notifier);
+        threshold_cpu_callback = amd_64_threshold_cpu_callback;
         return 0;
 }