aboutsummaryrefslogtreecommitdiffstats
path: root/arch/blackfin/kernel
diff options
context:
space:
mode:
authorChristoph Lameter <cl@linux.com>2014-08-17 13:30:52 -0400
committerTejun Heo <tj@kernel.org>2014-08-26 13:45:55 -0400
commit7e788ab11d73fbb617973c12a9b3f84f93721e67 (patch)
tree9c709266f7f754f56fdc809631ff4109f6ad2720 /arch/blackfin/kernel
parent81829a96869c8bad74a582705617e75758c4152d (diff)
blackfin: Replace __get_cpu_var uses
__get_cpu_var() is used for multiple purposes in the kernel source. One of them is address calculation via the form &__get_cpu_var(x). This calculates the address for the instance of the percpu variable of the current processor based on an offset. Other use cases are for storing and retrieving data from the current processors percpu area. __get_cpu_var() can be used as an lvalue when writing data or on the right side of an assignment. __get_cpu_var() is defined as : #define __get_cpu_var(var) (*this_cpu_ptr(&(var))) __get_cpu_var() always only does an address determination. However, store and retrieve operations could use a segment prefix (or global register on other platforms) to avoid the address calculation. this_cpu_write() and this_cpu_read() can directly take an offset into a percpu area and use optimized assembly code to read and write per cpu variables. This patch converts __get_cpu_var into either an explicit address calculation using this_cpu_ptr() or into a use of this_cpu operations that use the offset. Thereby address calculations are avoided and less registers are used when code is generated. At the end of the patch set all uses of __get_cpu_var have been removed so the macro is removed too. The patch set includes passes over all arches as well. Once these operations are used throughout then specialized macros can be defined in non -x86 arches as well in order to optimize per cpu access by f.e. using a global register that may be set to the per cpu base. Transformations done to __get_cpu_var() 1. Determine the address of the percpu instance of the current processor. DEFINE_PER_CPU(int, y); int *x = &__get_cpu_var(y); Converts to int *x = this_cpu_ptr(&y); 2. Same as #1 but this time an array structure is involved. DEFINE_PER_CPU(int, y[20]); int *x = __get_cpu_var(y); Converts to int *x = this_cpu_ptr(y); 3. Retrieve the content of the current processors instance of a per cpu variable. DEFINE_PER_CPU(int, y); int x = __get_cpu_var(y) Converts to int x = __this_cpu_read(y); 4. Retrieve the content of a percpu struct DEFINE_PER_CPU(struct mystruct, y); struct mystruct x = __get_cpu_var(y); Converts to memcpy(&x, this_cpu_ptr(&y), sizeof(x)); 5. Assignment to a per cpu variable DEFINE_PER_CPU(int, y) __get_cpu_var(y) = x; Converts to __this_cpu_write(y, x); 6. Increment/Decrement etc of a per cpu variable DEFINE_PER_CPU(int, y); __get_cpu_var(y)++ Converts to __this_cpu_inc(y) CC: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Christoph Lameter <cl@linux.com> Signed-off-by: Tejun Heo <tj@kernel.org>
Diffstat (limited to 'arch/blackfin/kernel')
-rw-r--r--arch/blackfin/kernel/perf_event.c10
1 files changed, 5 insertions, 5 deletions
diff --git a/arch/blackfin/kernel/perf_event.c b/arch/blackfin/kernel/perf_event.c
index ea2032013cc2..1e9c8b0bf486 100644
--- a/arch/blackfin/kernel/perf_event.c
+++ b/arch/blackfin/kernel/perf_event.c
@@ -300,7 +300,7 @@ again:
300 300
301static void bfin_pmu_stop(struct perf_event *event, int flags) 301static void bfin_pmu_stop(struct perf_event *event, int flags)
302{ 302{
303 struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); 303 struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
304 struct hw_perf_event *hwc = &event->hw; 304 struct hw_perf_event *hwc = &event->hw;
305 int idx = hwc->idx; 305 int idx = hwc->idx;
306 306
@@ -318,7 +318,7 @@ static void bfin_pmu_stop(struct perf_event *event, int flags)
318 318
319static void bfin_pmu_start(struct perf_event *event, int flags) 319static void bfin_pmu_start(struct perf_event *event, int flags)
320{ 320{
321 struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); 321 struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
322 struct hw_perf_event *hwc = &event->hw; 322 struct hw_perf_event *hwc = &event->hw;
323 int idx = hwc->idx; 323 int idx = hwc->idx;
324 324
@@ -335,7 +335,7 @@ static void bfin_pmu_start(struct perf_event *event, int flags)
335 335
336static void bfin_pmu_del(struct perf_event *event, int flags) 336static void bfin_pmu_del(struct perf_event *event, int flags)
337{ 337{
338 struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); 338 struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
339 339
340 bfin_pmu_stop(event, PERF_EF_UPDATE); 340 bfin_pmu_stop(event, PERF_EF_UPDATE);
341 __clear_bit(event->hw.idx, cpuc->used_mask); 341 __clear_bit(event->hw.idx, cpuc->used_mask);
@@ -345,7 +345,7 @@ static void bfin_pmu_del(struct perf_event *event, int flags)
345 345
346static int bfin_pmu_add(struct perf_event *event, int flags) 346static int bfin_pmu_add(struct perf_event *event, int flags)
347{ 347{
348 struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); 348 struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
349 struct hw_perf_event *hwc = &event->hw; 349 struct hw_perf_event *hwc = &event->hw;
350 int idx = hwc->idx; 350 int idx = hwc->idx;
351 int ret = -EAGAIN; 351 int ret = -EAGAIN;
@@ -421,7 +421,7 @@ static int bfin_pmu_event_init(struct perf_event *event)
421 421
422static void bfin_pmu_enable(struct pmu *pmu) 422static void bfin_pmu_enable(struct pmu *pmu)
423{ 423{
424 struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); 424 struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
425 struct perf_event *event; 425 struct perf_event *event;
426 struct hw_perf_event *hwc; 426 struct hw_perf_event *hwc;
427 int i; 427 int i;