x86: Use this_cpu_ops to optimize code

Go through x86 code and replace __get_cpu_var and get_cpu_var
instances that refer to a scalar and are not used for address
determinations.

Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>
Acked-by: Tejun Heo <tj@kernel.org>
Acked-by: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Tejun Heo <tj@kernel.org>

16 files changed, 57 insertions, 62 deletions

diff --git a/arch/x86/include/asm/debugreg.h b/arch/x86/include/asm/debugreg.h
index b81002f23614..078ad0caefc6 100644
--- a/arch/x86/include/asm/debugreg.h
+++ b/arch/x86/include/asm/debugreg.h
@@ -94,7 +94,7 @@ static inline void hw_breakpoint_disable(void)
 
 static inline int hw_breakpoint_active(void)
 {
-        return __get_cpu_var(cpu_dr7) & DR_GLOBAL_ENABLE_MASK;
+        return __this_cpu_read(cpu_dr7) & DR_GLOBAL_ENABLE_MASK;
 }
 
 extern void aout_dump_debugregs(struct user *dump);
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 7cc0a721f628..8d50922687af 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -2302,7 +2302,7 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void)
                 unsigned int irr;
                 struct irq_desc *desc;
                 struct irq_cfg *cfg;
-                irq = __get_cpu_var(vector_irq)[vector];
+                irq = __this_cpu_read(vector_irq[vector]);
 
                 if (irq == -1)
                         continue;
@@ -2336,7 +2336,7 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void)
                         apic->send_IPI_self(IRQ_MOVE_CLEANUP_VECTOR);
                         goto unlock;
                 }
-                __get_cpu_var(vector_irq)[vector] = -1;
+                __this_cpu_write(vector_irq[vector], -1);
 unlock:
                 raw_spin_unlock(&desc->lock);
         }
diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c
index c90041ccb742..b387dce0b409 100644
--- a/arch/x86/kernel/apic/nmi.c
+++ b/arch/x86/kernel/apic/nmi.c
@@ -306,12 +306,12 @@ void acpi_nmi_disable(void)
  */
 void cpu_nmi_set_wd_enabled(void)
 {
-        __get_cpu_var(wd_enabled) = 1;
+        __this_cpu_write(wd_enabled, 1);
 }
 
 void setup_apic_nmi_watchdog(void *unused)
 {
-        if (__get_cpu_var(wd_enabled))
+        if (__this_cpu_read(wd_enabled))
                 return;
 
         /* cheap hack to support suspend/resume */
@@ -322,12 +322,12 @@ void setup_apic_nmi_watchdog(void *unused)
         switch (nmi_watchdog) {
         case NMI_LOCAL_APIC:
                 if (lapic_watchdog_init(nmi_hz) < 0) {
-                        __get_cpu_var(wd_enabled) = 0;
+                        __this_cpu_write(wd_enabled, 0);
                         return;
                 }
                 /* FALL THROUGH */
         case NMI_IO_APIC:
-                __get_cpu_var(wd_enabled) = 1;
+                __this_cpu_write(wd_enabled, 1);
                 atomic_inc(&nmi_active);
         }
 }
@@ -337,13 +337,13 @@ void stop_apic_nmi_watchdog(void *unused)
         /* only support LOCAL and IO APICs for now */
         if (!nmi_watchdog_active())
                 return;
-        if (__get_cpu_var(wd_enabled) == 0)
+        if (__this_cpu_read(wd_enabled) == 0)
                 return;
         if (nmi_watchdog == NMI_LOCAL_APIC)
                 lapic_watchdog_stop();
         else
                 __acpi_nmi_disable(NULL);
-        __get_cpu_var(wd_enabled) = 0;
+        __this_cpu_write(wd_enabled, 0);
         atomic_dec(&nmi_active);
 }
 
@@ -403,8 +403,8 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason)
 
         sum = get_timer_irqs(cpu);
 
-        if (__get_cpu_var(nmi_touch)) {
-                __get_cpu_var(nmi_touch) = 0;
+        if (__this_cpu_read(nmi_touch)) {
+                __this_cpu_write(nmi_touch, 0);
                 touched = 1;
         }
 
@@ -427,7 +427,7 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason)
                 touched = 1;
 
         /* if the none of the timers isn't firing, this cpu isn't doing much */
-        if (!touched && __get_cpu_var(last_irq_sum) == sum) {
+        if (!touched && __this_cpu_read(last_irq_sum) == sum) {
                 /*
                  * Ayiee, looks like this CPU is stuck ...
                  * wait a few IRQs (5 seconds) before doing the oops ...
@@ -440,12 +440,12 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason)
                         die_nmi("BUG: NMI Watchdog detected LOCKUP",
                                 regs, panic_on_timeout);
         } else {
-                __get_cpu_var(last_irq_sum) = sum;
+                __this_cpu_write(last_irq_sum, sum);
                 __this_cpu_write(alert_counter, 0);
         }
 
         /* see if the nmi watchdog went off */
-        if (!__get_cpu_var(wd_enabled))
+        if (!__this_cpu_read(wd_enabled))
                 return rc;
         switch (nmi_watchdog) {
         case NMI_LOCAL_APIC:
@@ -467,7 +467,7 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason)
 
 static void enable_ioapic_nmi_watchdog_single(void *unused)
 {
-        __get_cpu_var(wd_enabled) = 1;
+        __this_cpu_write(wd_enabled, 1);
         atomic_inc(&nmi_active);
         __acpi_nmi_enable(NULL);
 }
diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c
index c1c52c341f40..26ec9a7c3518 100644
--- a/arch/x86/kernel/apic/x2apic_uv_x.c
+++ b/arch/x86/kernel/apic/x2apic_uv_x.c
@@ -118,8 +118,8 @@ static int __init uv_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
                 else if (!strcmp(oem_table_id, "UVX"))
                         uv_system_type = UV_X2APIC;
                 else if (!strcmp(oem_table_id, "UVH")) {
-                        __get_cpu_var(x2apic_extra_bits) =
-                                nodeid << (uvh_apicid.s.pnode_shift - 1);
+                        __this_cpu_write(x2apic_extra_bits,
+                                nodeid << (uvh_apicid.s.pnode_shift - 1));
                         uv_system_type = UV_NON_UNIQUE_APIC;
                         uv_set_apicid_hibit();
                         return 1;
@@ -284,7 +284,7 @@ static unsigned int x2apic_get_apic_id(unsigned long x)
         unsigned int id;
 
         WARN_ON(preemptible() && num_online_cpus() > 1);
-        id = x | __get_cpu_var(x2apic_extra_bits);
+        id = x | __this_cpu_read(x2apic_extra_bits);
 
         return id;
 }
@@ -376,7 +376,7 @@ struct apic __refdata apic_x2apic_uv_x = {
 
 static __cpuinit void set_x2apic_extra_bits(int pnode)
 {
-        __get_cpu_var(x2apic_extra_bits) = (pnode << 6);
+        __this_cpu_write(x2apic_extra_bits, (pnode << 6));
 }
 
 /*
diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
index 491977baf6c0..42a36046823e 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
@@ -1377,7 +1377,7 @@ static int __devexit powernowk8_cpu_exit(struct cpufreq_policy *pol)
 static void query_values_on_cpu(void *_err)
 {
         int *err = _err;
-        struct powernow_k8_data *data = __get_cpu_var(powernow_data);
+        struct powernow_k8_data *data = __this_cpu_read(powernow_data);
 
         *err = query_current_values_with_pending_wait(data);
 }
diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index 7a35b72d7c03..0c746af6c5eb 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -326,7 +326,7 @@ static void mce_panic(char *msg, struct mce *final, char *exp)
 
 static int msr_to_offset(u32 msr)
 {
-        unsigned bank = __get_cpu_var(injectm.bank);
+        unsigned bank = __this_cpu_read(injectm.bank);
 
         if (msr == rip_msr)
                 return offsetof(struct mce, ip);
@@ -346,7 +346,7 @@ static u64 mce_rdmsrl(u32 msr)
 {
         u64 v;
 
-        if (__get_cpu_var(injectm).finished) {
+        if (__this_cpu_read(injectm.finished)) {
                 int offset = msr_to_offset(msr);
 
                 if (offset < 0)
@@ -369,7 +369,7 @@ static u64 mce_rdmsrl(u32 msr)
 
 static void mce_wrmsrl(u32 msr, u64 v)
 {
-        if (__get_cpu_var(injectm).finished) {
+        if (__this_cpu_read(injectm.finished)) {
                 int offset = msr_to_offset(msr);
 
                 if (offset >= 0)
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 6d75b9145b13..ba85814f2590 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -968,8 +968,7 @@ x86_perf_event_set_period(struct perf_event *event)
 
 static void x86_pmu_enable_event(struct perf_event *event)
 {
-        struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-        if (cpuc->enabled)
+        if (__this_cpu_read(cpu_hw_events.enabled))
                 __x86_pmu_enable_event(&event->hw,
                                        ARCH_PERFMON_EVENTSEL_ENABLE);
 }
@@ -1243,7 +1242,7 @@ perf_event_nmi_handler(struct notifier_block *self,
                 break;
         case DIE_NMIUNKNOWN:
                 this_nmi = percpu_read(irq_stat.__nmi_count);
-                if (this_nmi != __get_cpu_var(pmu_nmi).marked)
+                if (this_nmi != __this_cpu_read(pmu_nmi.marked))
                         /* let the kernel handle the unknown nmi */
                         return NOTIFY_DONE;
                 /*
@@ -1267,8 +1266,8 @@ perf_event_nmi_handler(struct notifier_block *self,
         this_nmi = percpu_read(irq_stat.__nmi_count);
         if ((handled > 1) ||
                 /* the next nmi could be a back-to-back nmi */
-            ((__get_cpu_var(pmu_nmi).marked == this_nmi) &&
-             (__get_cpu_var(pmu_nmi).handled > 1))) {
+            ((__this_cpu_read(pmu_nmi.marked) == this_nmi) &&
+             (__this_cpu_read(pmu_nmi.handled) > 1))) {
                 /*
                  * We could have two subsequent back-to-back nmis: The
                  * first handles more than one counter, the 2nd
@@ -1279,8 +1278,8 @@ perf_event_nmi_handler(struct notifier_block *self,
                  * handling more than one counter. We will mark the
                  * next (3rd) and then drop it if unhandled.
                  */
-                __get_cpu_var(pmu_nmi).marked   = this_nmi + 1;
-                __get_cpu_var(pmu_nmi).handled  = handled;
+                __this_cpu_write(pmu_nmi.marked, this_nmi + 1);
+                __this_cpu_write(pmu_nmi.handled, handled);
         }
 
         return NOTIFY_STOP;
@@ -1454,11 +1453,9 @@ static inline void x86_pmu_read(struct perf_event *event)
  */
 static void x86_pmu_start_txn(struct pmu *pmu)
 {
-        struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-
         perf_pmu_disable(pmu);
-        cpuc->group_flag |= PERF_EVENT_TXN;
-        cpuc->n_txn = 0;
+        __this_cpu_or(cpu_hw_events.group_flag, PERF_EVENT_TXN);
+        __this_cpu_write(cpu_hw_events.n_txn, 0);
 }
 
 /*
@@ -1468,14 +1465,12 @@ static void x86_pmu_start_txn(struct pmu *pmu)
  */
 static void x86_pmu_cancel_txn(struct pmu *pmu)
 {
-        struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-
-        cpuc->group_flag &= ~PERF_EVENT_TXN;
+        __this_cpu_and(cpu_hw_events.group_flag, ~PERF_EVENT_TXN);
         /*
          * Truncate the collected events.
          */
-        cpuc->n_added -= cpuc->n_txn;
-        cpuc->n_events -= cpuc->n_txn;
+        __this_cpu_sub(cpu_hw_events.n_added, __this_cpu_read(cpu_hw_events.n_txn));
+        __this_cpu_sub(cpu_hw_events.n_events, __this_cpu_read(cpu_hw_events.n_txn));
         perf_pmu_enable(pmu);
 }
 
diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index c8f5c088cad1..4ee59bcbdad3 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -649,7 +649,7 @@ static void intel_pmu_enable_event(struct perf_event *event)
         struct hw_perf_event *hwc = &event->hw;
 
         if (unlikely(hwc->idx == X86_PMC_IDX_FIXED_BTS)) {
-                if (!__get_cpu_var(cpu_hw_events).enabled)
+                if (!__this_cpu_read(cpu_hw_events.enabled))
                         return;
 
                 intel_pmu_enable_bts(hwc->config);
@@ -679,7 +679,7 @@ static int intel_pmu_save_and_restart(struct perf_event *event)
 
 static void intel_pmu_reset(void)
 {
-        struct debug_store *ds = __get_cpu_var(cpu_hw_events).ds;
+        struct debug_store *ds = __this_cpu_read(cpu_hw_events.ds);
         unsigned long flags;
         int idx;
 
diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c
index 3afb33f14d2d..b45246f9a640 100644
--- a/arch/x86/kernel/ftrace.c
+++ b/arch/x86/kernel/ftrace.c
@@ -167,9 +167,9 @@ static void ftrace_mod_code(void)
 
 void ftrace_nmi_enter(void)
 {
-        __get_cpu_var(save_modifying_code) = modifying_code;
+        __this_cpu_write(save_modifying_code, modifying_code);
 
-        if (!__get_cpu_var(save_modifying_code))
+        if (!__this_cpu_read(save_modifying_code))
                 return;
 
         if (atomic_inc_return(&nmi_running) & MOD_CODE_WRITE_FLAG) {
@@ -183,7 +183,7 @@ void ftrace_nmi_enter(void)
 
 void ftrace_nmi_exit(void)
 {
-        if (!__get_cpu_var(save_modifying_code))
+        if (!__this_cpu_read(save_modifying_code))
                 return;
 
         /* Finish all executions before clearing nmi_running */
diff --git a/arch/x86/kernel/hw_breakpoint.c b/arch/x86/kernel/hw_breakpoint.c
index 42c594254507..02f07634d265 100644
--- a/arch/x86/kernel/hw_breakpoint.c
+++ b/arch/x86/kernel/hw_breakpoint.c
@@ -122,7 +122,7 @@ int arch_install_hw_breakpoint(struct perf_event *bp)
                 return -EBUSY;
 
         set_debugreg(info->address, i);
-        __get_cpu_var(cpu_debugreg[i]) = info->address;
+        __this_cpu_write(cpu_debugreg[i], info->address);
 
         dr7 = &__get_cpu_var(cpu_dr7);
         *dr7 |= encode_dr7(i, info->len, info->type);
@@ -397,12 +397,12 @@ void flush_ptrace_hw_breakpoint(struct task_struct *tsk)
 
 void hw_breakpoint_restore(void)
 {
-        set_debugreg(__get_cpu_var(cpu_debugreg[0]), 0);
-        set_debugreg(__get_cpu_var(cpu_debugreg[1]), 1);
-        set_debugreg(__get_cpu_var(cpu_debugreg[2]), 2);
-        set_debugreg(__get_cpu_var(cpu_debugreg[3]), 3);
+        set_debugreg(__this_cpu_read(cpu_debugreg[0]), 0);
+        set_debugreg(__this_cpu_read(cpu_debugreg[1]), 1);
+        set_debugreg(__this_cpu_read(cpu_debugreg[2]), 2);
+        set_debugreg(__this_cpu_read(cpu_debugreg[3]), 3);
         set_debugreg(current->thread.debugreg6, 6);
-        set_debugreg(__get_cpu_var(cpu_dr7), 7);
+        set_debugreg(__this_cpu_read(cpu_dr7), 7);
 }
 EXPORT_SYMBOL_GPL(hw_breakpoint_restore);
 
diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c
index 83ec0175f986..3a43caa3beb7 100644
--- a/arch/x86/kernel/irq.c
+++ b/arch/x86/kernel/irq.c
@@ -234,7 +234,7 @@ unsigned int __irq_entry do_IRQ(struct pt_regs *regs)
         exit_idle();
         irq_enter();
 
-        irq = __get_cpu_var(vector_irq)[vector];
+        irq = __this_cpu_read(vector_irq[vector]);
 
         if (!handle_irq(irq, regs)) {
                 ack_APIC_irq();
@@ -350,12 +350,12 @@ void fixup_irqs(void)
         for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS; vector++) {
                 unsigned int irr;
 
-                if (__get_cpu_var(vector_irq)[vector] < 0)
+                if (__this_cpu_read(vector_irq[vector]) < 0)
                         continue;
 
                 irr = apic_read(APIC_IRR + (vector / 32 * 0x10));
                 if (irr  & (1 << (vector % 32))) {
-                        irq = __get_cpu_var(vector_irq)[vector];
+                        irq = __this_cpu_read(vector_irq[vector]);
 
                         data = irq_get_irq_data(irq);
                         raw_spin_lock(&desc->lock);
diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c
index 96656f207751..48ff6dcffa02 100644
--- a/arch/x86/kernel/irq_32.c
+++ b/arch/x86/kernel/irq_32.c
@@ -79,7 +79,7 @@ execute_on_irq_stack(int overflow, struct irq_desc *desc, int irq)
         u32 *isp, arg1, arg2;
 
         curctx = (union irq_ctx *) current_thread_info();
-        irqctx = __get_cpu_var(hardirq_ctx);
+        irqctx = __this_cpu_read(hardirq_ctx);
 
         /*
          * this is where we switch to the IRQ stack. However, if we are
@@ -166,7 +166,7 @@ asmlinkage void do_softirq(void)
 
         if (local_softirq_pending()) {
                 curctx = current_thread_info();
-                irqctx = __get_cpu_var(softirq_ctx);
+                irqctx = __this_cpu_read(softirq_ctx);
                 irqctx->tinfo.task = curctx->task;
                 irqctx->tinfo.previous_esp = current_stack_pointer;
 
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 083e99d1b7df..ff4e5a113a5b 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -1377,7 +1377,7 @@ void play_dead_common(void)
 
         mb();
         /* Ack it */
-        __get_cpu_var(cpu_state) = CPU_DEAD;
+        __this_cpu_write(cpu_state, CPU_DEAD);
 
         /*
          * With physical CPU hotplug, we should halt the cpu
diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index 0c40d8b72416..acb08dd7bb57 100644
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -659,7 +659,7 @@ void restore_sched_clock_state(void)
 
         local_irq_save(flags);
 
-        __get_cpu_var(cyc2ns_offset) = 0;
+        __this_cpu_write(cyc2ns_offset, 0);
         offset = cyc2ns_suspend - sched_clock();
 
         for_each_possible_cpu(cpu)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index cdac9e592aa5..79d9606c202c 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -981,7 +981,7 @@ static inline u64 nsec_to_cycles(u64 nsec)
         if (kvm_tsc_changes_freq())
                 printk_once(KERN_WARNING
                  "kvm: unreliable cycle conversion on adjustable rate TSC\n");
-        ret = nsec * __get_cpu_var(cpu_tsc_khz);
+        ret = nsec * __this_cpu_read(cpu_tsc_khz);
         do_div(ret, USEC_PER_SEC);
         return ret;
 }
@@ -1066,7 +1066,7 @@ static int kvm_guest_time_update(struct kvm_vcpu *v)
         local_irq_save(flags);
         kvm_get_msr(v, MSR_IA32_TSC, &tsc_timestamp);
         kernel_ns = get_kernel_ns();
-        this_tsc_khz = __get_cpu_var(cpu_tsc_khz);
+        this_tsc_khz = __this_cpu_read(cpu_tsc_khz);
 
         if (unlikely(this_tsc_khz == 0)) {
                 local_irq_restore(flags);
@@ -4432,7 +4432,7 @@ EXPORT_SYMBOL_GPL(kvm_fast_pio_out);
 
 static void tsc_bad(void *info)
 {
-        __get_cpu_var(cpu_tsc_khz) = 0;
+        __this_cpu_write(cpu_tsc_khz, 0);
 }
 
 static void tsc_khz_changed(void *data)
@@ -4446,7 +4446,7 @@ static void tsc_khz_changed(void *data)
                 khz = cpufreq_quick_get(raw_smp_processor_id());
         if (!khz)
                 khz = tsc_khz;
-        __get_cpu_var(cpu_tsc_khz) = khz;
+        __this_cpu_write(cpu_tsc_khz, khz);
 }
 
 static int kvmclock_cpufreq_notifier(struct notifier_block *nb, unsigned long val,
diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 4e8baad36d37..a0cae67a657a 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -143,7 +143,7 @@ static inline int has_mux(void)
 
 inline int op_x86_phys_to_virt(int phys)
 {
-        return __get_cpu_var(switch_index) + phys;
+        return __this_cpu_read(switch_index) + phys;
 }
 
 inline int op_x86_virt_to_phys(int virt)