Re: [PATCH] Code clean up for percpu_xxx() functions

From: Alex,Shi
Date: Tue Oct 11 2011 - 05:15:53 EST

On Tue, 2011-10-11 at 01:08 +0800, Christoph Lameter wrote:
> On Mon, 10 Oct 2011, Alex,Shi wrote:
>
> > index 3470c9d..d1de56c 100644
> > --- a/arch/x86/include/asm/percpu.h
> > +++ b/arch/x86/include/asm/percpu.h
> > @@ -46,7 +46,7 @@
> >
> > #ifdef CONFIG_SMP
> > #define __percpu_prefix "%%"__stringify(__percpu_seg)":"
> > -#define __my_cpu_offset percpu_read(this_cpu_off)
> > +#define __my_cpu_offset this_cpu_read(this_cpu_off)
>
> Should this not be __this_cpu_read() instead?

OK.
>
> > * per-thread variables implemented as per-cpu variables and thus
> > * stable for the duration of the respective task.
> > */
> > -#define percpu_read(var) percpu_from_op("mov", var, "m" (var))
> > -#define percpu_read_stable(var) percpu_from_op("mov", var, "p" (&(var)))
> > -#define percpu_write(var, val) percpu_to_op("mov", var, val)
> > -#define percpu_add(var, val) percpu_add_op(var, val)
> > -#define percpu_sub(var, val) percpu_add_op(var, -(val))
> > -#define percpu_and(var, val) percpu_to_op("and", var, val)
> > -#define percpu_or(var, val) percpu_to_op("or", var, val)
> > -#define percpu_xor(var, val) percpu_to_op("xor", var, val)
> > -#define percpu_inc(var) percpu_unary_op("inc", var)
> > +#define this_cpu_read(var) percpu_from_op("mov", var, "m" (var))
> > +#define this_cpu_read_stable(var) percpu_from_op("mov", var, "p" (&(var)))
> > +#define this_cpu_write(var, val) percpu_to_op("mov", var, val)
> > +#define this_cpu_add(var, val) percpu_add_op(var, val)
> > +#define this_cpu_sub(var, val) percpu_add_op(var, -(val))
> > +#define this_cpu_and(var, val) percpu_to_op("and", var, val)
> > +#define this_cpu_or(var, val) percpu_to_op("or", var, val)
> > +#define this_cpu_xor(var, val) percpu_to_op("xor", var, val)
> > +#define this_cpu_inc(var) percpu_unary_op("inc", var)
>
> Why define these operations? They are already defined in the form of
> this_cpu_add_1/2/4/8 and include/linux/percpu.h branches out to them.

percpu_read() used in asm/percpu.h yet. Anyway, I removed definitions
here and do a work around for the function as following:

@@ -551,7 +544,11 @@ static __always_inline int x86_this_cpu_constant_test_bit(unsigned int nr,
{
unsigned long __percpu *a = (unsigned long *)addr + nr / BITS_PER_LONG;

- return ((1UL << (nr % BITS_PER_LONG)) & percpu_read(*a)) != 0;
+#ifdef CONFIG_X86_64
+ return ((1UL << (nr % BITS_PER_LONG)) & this_cpu_read_8(*a)) != 0;
+#else
+ return ((1UL << (nr % BITS_PER_LONG)) & this_cpu_read_4(*a)) != 0;
+#endif

>
> > @@ -522,7 +522,7 @@ void sock_release(struct socket *sock)
> > if (rcu_dereference_protected(sock->wq, 1)->fasync_list)
> > printk(KERN_ERR "sock_release: fasync list not empty!\n");
> >
> > - percpu_sub(sockets_in_use, 1);
> > + this_cpu_sub(sockets_in_use, 1);
> > if (!sock->file) {
> > iput(SOCK_INODE(sock));
> > return;
>
> In general __this_cpu_xx operations can be used instead of this_cpu_xx
> operations if the context is already preempt safe. There is no change of
> the code generated for x86 in those cases. But the fallback code on other
> platforms can generate more efficient code.
>


Good reminder!
Currently, only x86 has specific percpu_xxx() definition. Other
platforms use generic version. And for this function, it is not preempt
safe. So, it should be fine here.

=========

Since percpu_xxx() serial functions are duplicate with this_cpu_xxx().
Removing percpu_xxx() definition and replacing them by this_cpu_xxx() in
using. This patch is just do code clean up, no function change.


Signed-off-by: Alex Shi <alex.shi@xxxxxxxxx>
---
arch/x86/include/asm/current.h | 2 +-
arch/x86/include/asm/hardirq.h | 8 ++--
arch/x86/include/asm/irq_regs.h | 4 +-
arch/x86/include/asm/mmu_context.h | 12 ++++----
arch/x86/include/asm/percpu.h | 25 +++++++---------
arch/x86/include/asm/smp.h | 4 +-
arch/x86/include/asm/stackprotector.h | 4 +-
arch/x86/include/asm/thread_info.h | 2 +-
arch/x86/include/asm/tlbflush.h | 4 +-
arch/x86/kernel/cpu/common.c | 2 +-
arch/x86/kernel/cpu/mcheck/mce.c | 4 +-
arch/x86/kernel/cpu/perf_event.c | 4 +-
arch/x86/kernel/paravirt.c | 12 ++++----
arch/x86/kernel/process_32.c | 2 +-
arch/x86/kernel/process_64.c | 12 ++++----
arch/x86/kvm/x86.c | 14 ++++----
arch/x86/mm/tlb.c | 10 +++---
arch/x86/xen/enlighten.c | 6 ++--
arch/x86/xen/irq.c | 8 ++--
arch/x86/xen/mmu.c | 20 ++++++------
arch/x86/xen/multicalls.h | 2 +-
arch/x86/xen/smp.c | 2 +-
include/linux/percpu.h | 53 ---------------------------------
include/linux/topology.h | 4 +-
net/netfilter/xt_TEE.c | 12 ++++----
net/socket.c | 4 +-
26 files changed, 90 insertions(+), 146 deletions(-)

diff --git a/arch/x86/include/asm/current.h b/arch/x86/include/asm/current.h
index 4d447b7..9476c04 100644
--- a/arch/x86/include/asm/current.h
+++ b/arch/x86/include/asm/current.h
@@ -11,7 +11,7 @@ DECLARE_PER_CPU(struct task_struct *, current_task);

static __always_inline struct task_struct *get_current(void)
{
- return percpu_read_stable(current_task);
+ return this_cpu_read_stable(current_task);
}

#define current get_current()
diff --git a/arch/x86/include/asm/hardirq.h b/arch/x86/include/asm/hardirq.h
index 55e4de6..fc594da 100644
--- a/arch/x86/include/asm/hardirq.h
+++ b/arch/x86/include/asm/hardirq.h
@@ -35,14 +35,14 @@ DECLARE_PER_CPU_SHARED_ALIGNED(irq_cpustat_t, irq_stat);

#define __ARCH_IRQ_STAT

-#define inc_irq_stat(member) percpu_inc(irq_stat.member)
+#define inc_irq_stat(member) this_cpu_inc(irq_stat.member)

-#define local_softirq_pending() percpu_read(irq_stat.__softirq_pending)
+#define local_softirq_pending() this_cpu_read(irq_stat.__softirq_pending)

#define __ARCH_SET_SOFTIRQ_PENDING

-#define set_softirq_pending(x) percpu_write(irq_stat.__softirq_pending, (x))
-#define or_softirq_pending(x) percpu_or(irq_stat.__softirq_pending, (x))
+#define set_softirq_pending(x) this_cpu_write(irq_stat.__softirq_pending, (x))
+#define or_softirq_pending(x) this_cpu_or(irq_stat.__softirq_pending, (x))

extern void ack_bad_irq(unsigned int irq);

diff --git a/arch/x86/include/asm/irq_regs.h b/arch/x86/include/asm/irq_regs.h
index 7784322..d82250b 100644
--- a/arch/x86/include/asm/irq_regs.h
+++ b/arch/x86/include/asm/irq_regs.h
@@ -15,7 +15,7 @@ DECLARE_PER_CPU(struct pt_regs *, irq_regs);

static inline struct pt_regs *get_irq_regs(void)
{
- return percpu_read(irq_regs);
+ return this_cpu_read(irq_regs);
}

static inline struct pt_regs *set_irq_regs(struct pt_regs *new_regs)
@@ -23,7 +23,7 @@ static inline struct pt_regs *set_irq_regs(struct pt_regs *new_regs)
struct pt_regs *old_regs;

old_regs = get_irq_regs();
- percpu_write(irq_regs, new_regs);
+ this_cpu_write(irq_regs, new_regs);

return old_regs;
}
diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h
index 6902152..cdbf367 100644
--- a/arch/x86/include/asm/mmu_context.h
+++ b/arch/x86/include/asm/mmu_context.h
@@ -25,8 +25,8 @@ void destroy_context(struct mm_struct *mm);
static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
{
#ifdef CONFIG_SMP
- if (percpu_read(cpu_tlbstate.state) == TLBSTATE_OK)
- percpu_write(cpu_tlbstate.state, TLBSTATE_LAZY);
+ if (this_cpu_read(cpu_tlbstate.state) == TLBSTATE_OK)
+ this_cpu_write(cpu_tlbstate.state, TLBSTATE_LAZY);
#endif
}

@@ -37,8 +37,8 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,

if (likely(prev != next)) {
#ifdef CONFIG_SMP
- percpu_write(cpu_tlbstate.state, TLBSTATE_OK);
- percpu_write(cpu_tlbstate.active_mm, next);
+ this_cpu_write(cpu_tlbstate.state, TLBSTATE_OK);
+ this_cpu_write(cpu_tlbstate.active_mm, next);
#endif
cpumask_set_cpu(cpu, mm_cpumask(next));

@@ -56,8 +56,8 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
}
#ifdef CONFIG_SMP
else {
- percpu_write(cpu_tlbstate.state, TLBSTATE_OK);
- BUG_ON(percpu_read(cpu_tlbstate.active_mm) != next);
+ this_cpu_write(cpu_tlbstate.state, TLBSTATE_OK);
+ BUG_ON(this_cpu_read(cpu_tlbstate.active_mm) != next);

if (!cpumask_test_and_set_cpu(cpu, mm_cpumask(next))) {
/* We were in lazy tlb mode and leave_mm disabled
diff --git a/arch/x86/include/asm/percpu.h b/arch/x86/include/asm/percpu.h
index 3470c9d..66b50cb 100644
--- a/arch/x86/include/asm/percpu.h
+++ b/arch/x86/include/asm/percpu.h
@@ -46,7 +46,7 @@

#ifdef CONFIG_SMP
#define __percpu_prefix "%%"__stringify(__percpu_seg)":"
-#define __my_cpu_offset percpu_read(this_cpu_off)
+#define __my_cpu_offset __this_cpu_read(this_cpu_off)

/*
* Compared to the generic __my_cpu_offset version, the following
@@ -351,23 +351,16 @@ do { \
})

/*
- * percpu_read() makes gcc load the percpu variable every time it is
- * accessed while percpu_read_stable() allows the value to be cached.
- * percpu_read_stable() is more efficient and can be used if its value
+ * this_cpu_read() makes gcc load the percpu variable every time it is
+ * accessed while this_cpu_read_stable() allows the value to be cached.
+ * this_cpu_read_stable() is more efficient and can be used if its value
* is guaranteed to be valid across cpus. The current users include
* get_current() and get_thread_info() both of which are actually
* per-thread variables implemented as per-cpu variables and thus
* stable for the duration of the respective task.
*/
-#define percpu_read(var) percpu_from_op("mov", var, "m" (var))
-#define percpu_read_stable(var) percpu_from_op("mov", var, "p" (&(var)))
-#define percpu_write(var, val) percpu_to_op("mov", var, val)
-#define percpu_add(var, val) percpu_add_op(var, val)
-#define percpu_sub(var, val) percpu_add_op(var, -(val))
-#define percpu_and(var, val) percpu_to_op("and", var, val)
-#define percpu_or(var, val) percpu_to_op("or", var, val)
-#define percpu_xor(var, val) percpu_to_op("xor", var, val)
-#define percpu_inc(var) percpu_unary_op("inc", var)
+#define this_cpu_read_stable(var) percpu_from_op("mov", var, "p" (&(var)))
+#define this_cpu_inc(var) percpu_unary_op("inc", var)

#define __this_cpu_read_1(pcp) percpu_from_op("mov", (pcp), "m"(pcp))
#define __this_cpu_read_2(pcp) percpu_from_op("mov", (pcp), "m"(pcp))
@@ -551,7 +544,11 @@ static __always_inline int x86_this_cpu_constant_test_bit(unsigned int nr,
{
unsigned long __percpu *a = (unsigned long *)addr + nr / BITS_PER_LONG;

- return ((1UL << (nr % BITS_PER_LONG)) & percpu_read(*a)) != 0;
+#ifdef CONFIG_X86_64
+ return ((1UL << (nr % BITS_PER_LONG)) & this_cpu_read_8(*a)) != 0;
+#else
+ return ((1UL << (nr % BITS_PER_LONG)) & this_cpu_read_4(*a)) != 0;
+#endif
}

static inline int x86_this_cpu_variable_test_bit(int nr,
diff --git a/arch/x86/include/asm/smp.h b/arch/x86/include/asm/smp.h
index 73b11bc..dd5749b 100644
--- a/arch/x86/include/asm/smp.h
+++ b/arch/x86/include/asm/smp.h
@@ -188,11 +188,11 @@ extern unsigned disabled_cpus __cpuinitdata;
* from the initial startup. We map APIC_BASE very early in page_setup(),
* so this is correct in the x86 case.
*/
-#define raw_smp_processor_id() (percpu_read(cpu_number))
+#define raw_smp_processor_id() (this_cpu_read(cpu_number))
extern int safe_smp_processor_id(void);

#elif defined(CONFIG_X86_64_SMP)
-#define raw_smp_processor_id() (percpu_read(cpu_number))
+#define raw_smp_processor_id() (this_cpu_read(cpu_number))

#define stack_smp_processor_id() \
({ \
diff --git a/arch/x86/include/asm/stackprotector.h b/arch/x86/include/asm/stackprotector.h
index 1575177..3d7a009 100644
--- a/arch/x86/include/asm/stackprotector.h
+++ b/arch/x86/include/asm/stackprotector.h
@@ -76,9 +76,9 @@ static __always_inline void boot_init_stack_canary(void)

current->stack_canary = canary;
#ifdef CONFIG_X86_64
- percpu_write(irq_stack_union.stack_canary, canary);
+ this_cpu_write(irq_stack_union.stack_canary, canary);
#else
- percpu_write(stack_canary.canary, canary);
+ this_cpu_write(stack_canary.canary, canary);
#endif
}

diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h
index a1fe5c1..8b19667 100644
--- a/arch/x86/include/asm/thread_info.h
+++ b/arch/x86/include/asm/thread_info.h
@@ -219,7 +219,7 @@ DECLARE_PER_CPU(unsigned long, kernel_stack);
static inline struct thread_info *current_thread_info(void)
{
struct thread_info *ti;
- ti = (void *)(percpu_read_stable(kernel_stack) +
+ ti = (void *)(this_cpu_read_stable(kernel_stack) +
KERNEL_STACK_OFFSET - THREAD_SIZE);
return ti;
}
diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h
index 169be89..200c09c 100644
--- a/arch/x86/include/asm/tlbflush.h
+++ b/arch/x86/include/asm/tlbflush.h
@@ -156,8 +156,8 @@ DECLARE_PER_CPU_SHARED_ALIGNED(struct tlb_state, cpu_tlbstate);

static inline void reset_lazy_tlbstate(void)
{
- percpu_write(cpu_tlbstate.state, 0);
- percpu_write(cpu_tlbstate.active_mm, &init_mm);
+ this_cpu_write(cpu_tlbstate.state, 0);
+ this_cpu_write(cpu_tlbstate.active_mm, &init_mm);
}

#endif /* SMP */
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 6218439..10513ed 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -1158,7 +1158,7 @@ void __cpuinit cpu_init(void)
oist = &per_cpu(orig_ist, cpu);

#ifdef CONFIG_NUMA
- if (cpu != 0 && percpu_read(numa_node) == 0 &&
+ if (cpu != 0 && this_cpu_read(numa_node) == 0 &&
early_cpu_to_node(cpu) != NUMA_NO_NODE)
set_numa_node(early_cpu_to_node(cpu));
#endif
diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index 08363b0..adbb598 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -512,7 +512,7 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b)
struct mce m;
int i;

- percpu_inc(mce_poll_count);
+ this_cpu_inc(mce_poll_count);

mce_gather_info(&m, NULL);

@@ -906,7 +906,7 @@ void do_machine_check(struct pt_regs *regs, long error_code)

atomic_inc(&mce_entry);

- percpu_inc(mce_exception_count);
+ this_cpu_inc(mce_exception_count);

if (notify_die(DIE_NMI, "machine check", regs, error_code,
18, SIGKILL) == NOTIFY_STOP)
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index cfa62ec..ca037c0 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -1389,7 +1389,7 @@ perf_event_nmi_handler(struct notifier_block *self,
case DIE_NMI:
break;
case DIE_NMIUNKNOWN:
- this_nmi = percpu_read(irq_stat.__nmi_count);
+ this_nmi = this_cpu_read(irq_stat.__nmi_count);
if (this_nmi != __this_cpu_read(pmu_nmi.marked))
/* let the kernel handle the unknown nmi */
return NOTIFY_DONE;
@@ -1409,7 +1409,7 @@ perf_event_nmi_handler(struct notifier_block *self,
if (!handled)
return NOTIFY_DONE;

- this_nmi = percpu_read(irq_stat.__nmi_count);
+ this_nmi = this_cpu_read(irq_stat.__nmi_count);
if ((handled > 1) ||
/* the next nmi could be a back-to-back nmi */
((__this_cpu_read(pmu_nmi.marked) == this_nmi) &&
diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c
index d90272e..b098092 100644
--- a/arch/x86/kernel/paravirt.c
+++ b/arch/x86/kernel/paravirt.c
@@ -239,16 +239,16 @@ static DEFINE_PER_CPU(enum paravirt_lazy_mode, paravirt_lazy_mode) = PARAVIRT_LA

static inline void enter_lazy(enum paravirt_lazy_mode mode)
{
- BUG_ON(percpu_read(paravirt_lazy_mode) != PARAVIRT_LAZY_NONE);
+ BUG_ON(this_cpu_read(paravirt_lazy_mode) != PARAVIRT_LAZY_NONE);

- percpu_write(paravirt_lazy_mode, mode);
+ this_cpu_write(paravirt_lazy_mode, mode);
}

static void leave_lazy(enum paravirt_lazy_mode mode)
{
- BUG_ON(percpu_read(paravirt_lazy_mode) != mode);
+ BUG_ON(this_cpu_read(paravirt_lazy_mode) != mode);

- percpu_write(paravirt_lazy_mode, PARAVIRT_LAZY_NONE);
+ this_cpu_write(paravirt_lazy_mode, PARAVIRT_LAZY_NONE);
}

void paravirt_enter_lazy_mmu(void)
@@ -265,7 +265,7 @@ void paravirt_start_context_switch(struct task_struct *prev)
{
BUG_ON(preemptible());

- if (percpu_read(paravirt_lazy_mode) == PARAVIRT_LAZY_MMU) {
+ if (this_cpu_read(paravirt_lazy_mode) == PARAVIRT_LAZY_MMU) {
arch_leave_lazy_mmu_mode();
set_ti_thread_flag(task_thread_info(prev), TIF_LAZY_MMU_UPDATES);
}
@@ -287,7 +287,7 @@ enum paravirt_lazy_mode paravirt_get_lazy_mode(void)
if (in_interrupt())
return PARAVIRT_LAZY_NONE;

- return percpu_read(paravirt_lazy_mode);
+ return this_cpu_read(paravirt_lazy_mode);
}

void arch_flush_lazy_mmu_mode(void)
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 7a3b651..f286be1 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -373,7 +373,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
if (prev->gs | next->gs)
lazy_load_gs(next->gs);

- percpu_write(current_task, next_p);
+ this_cpu_write(current_task, next_p);

return prev_p;
}
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index f693e44..35fdc14 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -73,7 +73,7 @@ EXPORT_SYMBOL_GPL(idle_notifier_unregister);

void enter_idle(void)
{
- percpu_write(is_idle, 1);
+ this_cpu_write(is_idle, 1);
atomic_notifier_call_chain(&idle_notifier, IDLE_START, NULL);
}

@@ -336,7 +336,7 @@ start_thread_common(struct pt_regs *regs, unsigned long new_ip,
load_gs_index(0);
regs->ip = new_ip;
regs->sp = new_sp;
- percpu_write(old_rsp, new_sp);
+ this_cpu_write(old_rsp, new_sp);
regs->cs = _cs;
regs->ss = _ss;
regs->flags = X86_EFLAGS_IF;
@@ -470,11 +470,11 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
/*
* Switch the PDA and FPU contexts.
*/
- prev->usersp = percpu_read(old_rsp);
- percpu_write(old_rsp, next->usersp);
- percpu_write(current_task, next_p);
+ prev->usersp = this_cpu_read(old_rsp);
+ this_cpu_write(old_rsp, next->usersp);
+ this_cpu_write(current_task, next_p);

- percpu_write(kernel_stack,
+ this_cpu_write(kernel_stack,
(unsigned long)task_stack_page(next_p) +
THREAD_SIZE - KERNEL_STACK_OFFSET);

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 84a28ea..f46cb15 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -5024,15 +5024,15 @@ static DEFINE_PER_CPU(struct kvm_vcpu *, current_vcpu);

static int kvm_is_in_guest(void)
{
- return percpu_read(current_vcpu) != NULL;
+ return this_cpu_read(current_vcpu) != NULL;
}

static int kvm_is_user_mode(void)
{
int user_mode = 3;

- if (percpu_read(current_vcpu))
- user_mode = kvm_x86_ops->get_cpl(percpu_read(current_vcpu));
+ if (this_cpu_read(current_vcpu))
+ user_mode = kvm_x86_ops->get_cpl(this_cpu_read(current_vcpu));

return user_mode != 0;
}
@@ -5041,8 +5041,8 @@ static unsigned long kvm_get_guest_ip(void)
{
unsigned long ip = 0;

- if (percpu_read(current_vcpu))
- ip = kvm_rip_read(percpu_read(current_vcpu));
+ if (this_cpu_read(current_vcpu))
+ ip = kvm_rip_read(this_cpu_read(current_vcpu));

return ip;
}
@@ -5055,13 +5055,13 @@ static struct perf_guest_info_callbacks kvm_guest_cbs = {

void kvm_before_handle_nmi(struct kvm_vcpu *vcpu)
{
- percpu_write(current_vcpu, vcpu);
+ this_cpu_write(current_vcpu, vcpu);
}
EXPORT_SYMBOL_GPL(kvm_before_handle_nmi);

void kvm_after_handle_nmi(struct kvm_vcpu *vcpu)
{
- percpu_write(current_vcpu, NULL);
+ this_cpu_write(current_vcpu, NULL);
}
EXPORT_SYMBOL_GPL(kvm_after_handle_nmi);

diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index d6c0418..3804471 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -61,10 +61,10 @@ static DEFINE_PER_CPU_READ_MOSTLY(int, tlb_vector_offset);
*/
void leave_mm(int cpu)
{
- if (percpu_read(cpu_tlbstate.state) == TLBSTATE_OK)
+ if (this_cpu_read(cpu_tlbstate.state) == TLBSTATE_OK)
BUG();
cpumask_clear_cpu(cpu,
- mm_cpumask(percpu_read(cpu_tlbstate.active_mm)));
+ mm_cpumask(this_cpu_read(cpu_tlbstate.active_mm)));
load_cr3(swapper_pg_dir);
}
EXPORT_SYMBOL_GPL(leave_mm);
@@ -152,8 +152,8 @@ void smp_invalidate_interrupt(struct pt_regs *regs)
* BUG();
*/

- if (f->flush_mm == percpu_read(cpu_tlbstate.active_mm)) {
- if (percpu_read(cpu_tlbstate.state) == TLBSTATE_OK) {
+ if (f->flush_mm == this_cpu_read(cpu_tlbstate.active_mm)) {
+ if (this_cpu_read(cpu_tlbstate.state) == TLBSTATE_OK) {
if (f->flush_va == TLB_FLUSH_ALL)
local_flush_tlb();
else
@@ -322,7 +322,7 @@ void flush_tlb_page(struct vm_area_struct *vma, unsigned long va)
static void do_flush_tlb_all(void *info)
{
__flush_tlb_all();
- if (percpu_read(cpu_tlbstate.state) == TLBSTATE_LAZY)
+ if (this_cpu_read(cpu_tlbstate.state) == TLBSTATE_LAZY)
leave_mm(smp_processor_id());
}

diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index 2d69617..c4e362a 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -776,11 +776,11 @@ static DEFINE_PER_CPU(unsigned long, xen_cr0_value);

static unsigned long xen_read_cr0(void)
{
- unsigned long cr0 = percpu_read(xen_cr0_value);
+ unsigned long cr0 = this_cpu_read(xen_cr0_value);

if (unlikely(cr0 == 0)) {
cr0 = native_read_cr0();
- percpu_write(xen_cr0_value, cr0);
+ this_cpu_write(xen_cr0_value, cr0);
}

return cr0;
@@ -790,7 +790,7 @@ static void xen_write_cr0(unsigned long cr0)
{
struct multicall_space mcs;

- percpu_write(xen_cr0_value, cr0);
+ this_cpu_write(xen_cr0_value, cr0);

/* Only pay attention to cr0.TS; everything else is
ignored. */
diff --git a/arch/x86/xen/irq.c b/arch/x86/xen/irq.c
index 8bbb465..1573376 100644
--- a/arch/x86/xen/irq.c
+++ b/arch/x86/xen/irq.c
@@ -26,7 +26,7 @@ static unsigned long xen_save_fl(void)
struct vcpu_info *vcpu;
unsigned long flags;

- vcpu = percpu_read(xen_vcpu);
+ vcpu = this_cpu_read(xen_vcpu);

/* flag has opposite sense of mask */
flags = !vcpu->evtchn_upcall_mask;
@@ -50,7 +50,7 @@ static void xen_restore_fl(unsigned long flags)
make sure we're don't switch CPUs between getting the vcpu
pointer and updating the mask. */
preempt_disable();
- vcpu = percpu_read(xen_vcpu);
+ vcpu = this_cpu_read(xen_vcpu);
vcpu->evtchn_upcall_mask = flags;
preempt_enable_no_resched();

@@ -72,7 +72,7 @@ static void xen_irq_disable(void)
make sure we're don't switch CPUs between getting the vcpu
pointer and updating the mask. */
preempt_disable();
- percpu_read(xen_vcpu)->evtchn_upcall_mask = 1;
+ this_cpu_read(xen_vcpu)->evtchn_upcall_mask = 1;
preempt_enable_no_resched();
}
PV_CALLEE_SAVE_REGS_THUNK(xen_irq_disable);
@@ -86,7 +86,7 @@ static void xen_irq_enable(void)
the caller is confused and is trying to re-enable interrupts
on an indeterminate processor. */

- vcpu = percpu_read(xen_vcpu);
+ vcpu = this_cpu_read(xen_vcpu);
vcpu->evtchn_upcall_mask = 0;

/* Doesn't matter if we get preempted here, because any
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index 3dd53f9..6a7aaac 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -1106,14 +1106,14 @@ static void drop_other_mm_ref(void *info)
struct mm_struct *mm = info;
struct mm_struct *active_mm;

- active_mm = percpu_read(cpu_tlbstate.active_mm);
+ active_mm = this_cpu_read(cpu_tlbstate.active_mm);

- if (active_mm == mm && percpu_read(cpu_tlbstate.state) != TLBSTATE_OK)
+ if (active_mm == mm && this_cpu_read(cpu_tlbstate.state) != TLBSTATE_OK)
leave_mm(smp_processor_id());

/* If this cpu still has a stale cr3 reference, then make sure
it has been flushed. */
- if (percpu_read(xen_current_cr3) == __pa(mm->pgd))
+ if (this_cpu_read(xen_current_cr3) == __pa(mm->pgd))
load_cr3(swapper_pg_dir);
}

@@ -1220,17 +1220,17 @@ static void __init xen_pagetable_setup_done(pgd_t *base)

static void xen_write_cr2(unsigned long cr2)
{
- percpu_read(xen_vcpu)->arch.cr2 = cr2;
+ this_cpu_read(xen_vcpu)->arch.cr2 = cr2;
}

static unsigned long xen_read_cr2(void)
{
- return percpu_read(xen_vcpu)->arch.cr2;
+ return this_cpu_read(xen_vcpu)->arch.cr2;
}

unsigned long xen_read_cr2_direct(void)
{
- return percpu_read(xen_vcpu_info.arch.cr2);
+ return this_cpu_read(xen_vcpu_info.arch.cr2);
}

static void xen_flush_tlb(void)
@@ -1313,12 +1313,12 @@ static void xen_flush_tlb_others(const struct cpumask *cpus,

static unsigned long xen_read_cr3(void)
{
- return percpu_read(xen_cr3);
+ return this_cpu_read(xen_cr3);
}

static void set_current_cr3(void *v)
{
- percpu_write(xen_current_cr3, (unsigned long)v);
+ this_cpu_write(xen_current_cr3, (unsigned long)v);
}

static void __xen_write_cr3(bool kernel, unsigned long cr3)
@@ -1341,7 +1341,7 @@ static void __xen_write_cr3(bool kernel, unsigned long cr3)
xen_extend_mmuext_op(&op);

if (kernel) {
- percpu_write(xen_cr3, cr3);
+ this_cpu_write(xen_cr3, cr3);

/* Update xen_current_cr3 once the batch has actually
been submitted. */
@@ -1357,7 +1357,7 @@ static void xen_write_cr3(unsigned long cr3)

/* Update while interrupts are disabled, so its atomic with
respect to ipis */
- percpu_write(xen_cr3, cr3);
+ this_cpu_write(xen_cr3, cr3);

__xen_write_cr3(true, cr3);

diff --git a/arch/x86/xen/multicalls.h b/arch/x86/xen/multicalls.h
index dee79b7..9c2e74f 100644
--- a/arch/x86/xen/multicalls.h
+++ b/arch/x86/xen/multicalls.h
@@ -47,7 +47,7 @@ static inline void xen_mc_issue(unsigned mode)
xen_mc_flush();

/* restore flags saved in xen_mc_batch */
- local_irq_restore(percpu_read(xen_mc_irq_flags));
+ local_irq_restore(this_cpu_read(xen_mc_irq_flags));
}

/* Set up a callback to be called when the current batch is flushed */
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index 041d4fe..449f868 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -76,7 +76,7 @@ static void __cpuinit cpu_bringup(void)
xen_setup_cpu_clockevents();

set_cpu_online(cpu, true);
- percpu_write(cpu_state, CPU_ONLINE);
+ this_cpu_write(cpu_state, CPU_ONLINE);
wmb();

/* We can take interrupts now: we're officially "up". */
diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index 9ca008f..aad7e1c 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -165,59 +165,6 @@ extern phys_addr_t per_cpu_ptr_to_phys(void *addr);
#define alloc_percpu(type) \
(typeof(type) __percpu *)__alloc_percpu(sizeof(type), __alignof__(type))

-/*
- * Optional methods for optimized non-lvalue per-cpu variable access.
- *
- * @var can be a percpu variable or a field of it and its size should
- * equal char, int or long. percpu_read() evaluates to a lvalue and
- * all others to void.
- *
- * These operations are guaranteed to be atomic w.r.t. preemption.
- * The generic versions use plain get/put_cpu_var(). Archs are
- * encouraged to implement single-instruction alternatives which don't
- * require preemption protection.
- */
-#ifndef percpu_read
-# define percpu_read(var) \
- ({ \
- typeof(var) *pr_ptr__ = &(var); \
- typeof(var) pr_ret__; \
- pr_ret__ = get_cpu_var(*pr_ptr__); \
- put_cpu_var(*pr_ptr__); \
- pr_ret__; \
- })
-#endif
-
-#define __percpu_generic_to_op(var, val, op) \
-do { \
- typeof(var) *pgto_ptr__ = &(var); \
- get_cpu_var(*pgto_ptr__) op val; \
- put_cpu_var(*pgto_ptr__); \
-} while (0)
-
-#ifndef percpu_write
-# define percpu_write(var, val) __percpu_generic_to_op(var, (val), =)
-#endif
-
-#ifndef percpu_add
-# define percpu_add(var, val) __percpu_generic_to_op(var, (val), +=)
-#endif
-
-#ifndef percpu_sub
-# define percpu_sub(var, val) __percpu_generic_to_op(var, (val), -=)
-#endif
-
-#ifndef percpu_and
-# define percpu_and(var, val) __percpu_generic_to_op(var, (val), &=)
-#endif
-
-#ifndef percpu_or
-# define percpu_or(var, val) __percpu_generic_to_op(var, (val), |=)
-#endif
-
-#ifndef percpu_xor
-# define percpu_xor(var, val) __percpu_generic_to_op(var, (val), ^=)
-#endif

/*
* Branching function to split up a function into a set of functions that
diff --git a/include/linux/topology.h b/include/linux/topology.h
index fc839bf..bf64eac 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -235,7 +235,7 @@ static inline int cpu_to_node(int cpu)
#ifndef set_numa_node
static inline void set_numa_node(int node)
{
- percpu_write(numa_node, node);
+ this_cpu_write(numa_node, node);
}
#endif

@@ -270,7 +270,7 @@ DECLARE_PER_CPU(int, _numa_mem_);
#ifndef set_numa_mem
static inline void set_numa_mem(int node)
{
- percpu_write(_numa_mem_, node);
+ this_cpu_write(_numa_mem_, node);
}
#endif

diff --git a/net/netfilter/xt_TEE.c b/net/netfilter/xt_TEE.c
index 5f054a0..e78103d 100644
--- a/net/netfilter/xt_TEE.c
+++ b/net/netfilter/xt_TEE.c
@@ -90,7 +90,7 @@ tee_tg4(struct sk_buff *skb, const struct xt_action_param *par)
const struct xt_tee_tginfo *info = par->targinfo;
struct iphdr *iph;

- if (percpu_read(tee_active))
+ if (this_cpu_read(tee_active))
return XT_CONTINUE;
/*
* Copy the skb, and route the copy. Will later return %XT_CONTINUE for
@@ -127,9 +127,9 @@ tee_tg4(struct sk_buff *skb, const struct xt_action_param *par)
ip_send_check(iph);

if (tee_tg_route4(skb, info)) {
- percpu_write(tee_active, true);
+ this_cpu_write(tee_active, true);
ip_local_out(skb);
- percpu_write(tee_active, false);
+ this_cpu_write(tee_active, false);
} else {
kfree_skb(skb);
}
@@ -170,7 +170,7 @@ tee_tg6(struct sk_buff *skb, const struct xt_action_param *par)
{
const struct xt_tee_tginfo *info = par->targinfo;

- if (percpu_read(tee_active))
+ if (this_cpu_read(tee_active))
return XT_CONTINUE;
skb = pskb_copy(skb, GFP_ATOMIC);
if (skb == NULL)
@@ -188,9 +188,9 @@ tee_tg6(struct sk_buff *skb, const struct xt_action_param *par)
--iph->hop_limit;
}
if (tee_tg_route6(skb, info)) {
- percpu_write(tee_active, true);
+ this_cpu_write(tee_active, true);
ip6_local_out(skb);
- percpu_write(tee_active, false);
+ this_cpu_write(tee_active, false);
} else {
kfree_skb(skb);
}
diff --git a/net/socket.c b/net/socket.c
index ffe92ca..4b62ca9 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -479,7 +479,7 @@ static struct socket *sock_alloc(void)
inode->i_uid = current_fsuid();
inode->i_gid = current_fsgid();

- percpu_add(sockets_in_use, 1);
+ this_cpu_add(sockets_in_use, 1);
return sock;
}

@@ -522,7 +522,7 @@ void sock_release(struct socket *sock)
if (rcu_dereference_protected(sock->wq, 1)->fasync_list)
printk(KERN_ERR "sock_release: fasync list not empty!\n");

- percpu_sub(sockets_in_use, 1);
+ this_cpu_sub(sockets_in_use, 1);
if (!sock->file) {
iput(SOCK_INODE(sock));
return;
--
1.7.5.1







--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/