perf_event, powerpc: Fix compilation after big perf_counter rename
authorPaul Mackerras <paulus@samba.org>
Mon, 21 Sep 2009 23:48:08 +0000 (09:48 +1000)
committerIngo Molnar <mingo@elte.hu>
Tue, 22 Sep 2009 07:30:40 +0000 (09:30 +0200)
This fixes two places in the powerpc perf_event (perf_counter) code
where 'list_entry' needs to be changed to 'group_entry', but were
missed in commit 65abc865 ("perf_counter: Rename list_entry ->
group_entry, counter_list -> group_list").

This also changes 'event' back to 'counter' in a couple of
contexts:

* Field and function names that deal with the limited-function
  counters: it's really the hardware counters whose function is
  limited, not the events that they count.  Hence:

  MAX_LIMITED_HWEVENTS -> MAX_LIMITED_HWCOUNTERS
  limited_event -> limited_counter
  freeze/thaw_limited_events -> freeze/thaw_limited_counters

* The machine-specific PMU description struct (struct power_pmu): this
  renames 'n_event' back to 'n_counter' since it really describes how
  many hardware counters the machine has.  (Renaming this back avoids
  a compile error in each of the machine-specific PMU back-ends where
  they initialize their power_pmu struct.)

Signed-off-by: Paul Mackerras <paulus@samba.org>
Cc: linuxppc-dev@ozlabs.org
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <19128.4280.813369.589704@cargo.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
arch/powerpc/include/asm/perf_event.h
arch/powerpc/kernel/perf_event.c

index 2499aaadaeb94de9e54d402dacc603a7e309ad96..3288ce3997e0d380e299575883ccf37588f72c96 100644 (file)
@@ -14,7 +14,7 @@
 
 #define MAX_HWEVENTS           8
 #define MAX_EVENT_ALTERNATIVES 8
-#define MAX_LIMITED_HWEVENTS   2
+#define MAX_LIMITED_HWCOUNTERS 2
 
 /*
  * This struct provides the constants and functions needed to
@@ -22,7 +22,7 @@
  */
 struct power_pmu {
        const char      *name;
-       int             n_event;
+       int             n_counter;
        int             max_alternatives;
        unsigned long   add_fields;
        unsigned long   test_adder;
index 197b7d958796687e6f045e6aa8f8b5397e8d7eca..bbcbae183e92fce861fd800bdfa3647dae8ff31e 100644 (file)
@@ -30,8 +30,8 @@ struct cpu_hw_events {
        u64 events[MAX_HWEVENTS];
        unsigned int flags[MAX_HWEVENTS];
        unsigned long mmcr[3];
-       struct perf_event *limited_event[MAX_LIMITED_HWEVENTS];
-       u8  limited_hwidx[MAX_LIMITED_HWEVENTS];
+       struct perf_event *limited_counter[MAX_LIMITED_HWCOUNTERS];
+       u8  limited_hwidx[MAX_LIMITED_HWCOUNTERS];
        u64 alternatives[MAX_HWEVENTS][MAX_EVENT_ALTERNATIVES];
        unsigned long amasks[MAX_HWEVENTS][MAX_EVENT_ALTERNATIVES];
        unsigned long avalues[MAX_HWEVENTS][MAX_EVENT_ALTERNATIVES];
@@ -253,7 +253,7 @@ static int power_check_constraints(struct cpu_hw_events *cpuhw,
        unsigned long addf = ppmu->add_fields;
        unsigned long tadd = ppmu->test_adder;
 
-       if (n_ev > ppmu->n_event)
+       if (n_ev > ppmu->n_counter)
                return -1;
 
        /* First see if the events will go on as-is */
@@ -426,7 +426,7 @@ static int is_limited_pmc(int pmcnum)
                && (pmcnum == 5 || pmcnum == 6);
 }
 
-static void freeze_limited_events(struct cpu_hw_events *cpuhw,
+static void freeze_limited_counters(struct cpu_hw_events *cpuhw,
                                    unsigned long pmc5, unsigned long pmc6)
 {
        struct perf_event *event;
@@ -434,7 +434,7 @@ static void freeze_limited_events(struct cpu_hw_events *cpuhw,
        int i;
 
        for (i = 0; i < cpuhw->n_limited; ++i) {
-               event = cpuhw->limited_event[i];
+               event = cpuhw->limited_counter[i];
                if (!event->hw.idx)
                        continue;
                val = (event->hw.idx == 5) ? pmc5 : pmc6;
@@ -445,7 +445,7 @@ static void freeze_limited_events(struct cpu_hw_events *cpuhw,
        }
 }
 
-static void thaw_limited_events(struct cpu_hw_events *cpuhw,
+static void thaw_limited_counters(struct cpu_hw_events *cpuhw,
                                  unsigned long pmc5, unsigned long pmc6)
 {
        struct perf_event *event;
@@ -453,7 +453,7 @@ static void thaw_limited_events(struct cpu_hw_events *cpuhw,
        int i;
 
        for (i = 0; i < cpuhw->n_limited; ++i) {
-               event = cpuhw->limited_event[i];
+               event = cpuhw->limited_counter[i];
                event->hw.idx = cpuhw->limited_hwidx[i];
                val = (event->hw.idx == 5) ? pmc5 : pmc6;
                atomic64_set(&event->hw.prev_count, val);
@@ -495,9 +495,9 @@ static void write_mmcr0(struct cpu_hw_events *cpuhw, unsigned long mmcr0)
                       "i" (SPRN_PMC5), "i" (SPRN_PMC6));
 
        if (mmcr0 & MMCR0_FC)
-               freeze_limited_events(cpuhw, pmc5, pmc6);
+               freeze_limited_counters(cpuhw, pmc5, pmc6);
        else
-               thaw_limited_events(cpuhw, pmc5, pmc6);
+               thaw_limited_counters(cpuhw, pmc5, pmc6);
 
        /*
         * Write the full MMCR0 including the event overflow interrupt
@@ -653,7 +653,7 @@ void hw_perf_enable(void)
                        continue;
                idx = hwc_index[i] + 1;
                if (is_limited_pmc(idx)) {
-                       cpuhw->limited_event[n_lim] = event;
+                       cpuhw->limited_counter[n_lim] = event;
                        cpuhw->limited_hwidx[n_lim] = idx;
                        ++n_lim;
                        continue;
@@ -702,7 +702,7 @@ static int collect_events(struct perf_event *group, int max_count,
                flags[n] = group->hw.event_base;
                events[n++] = group->hw.config;
        }
-       list_for_each_entry(event, &group->sibling_list, list_entry) {
+       list_for_each_entry(event, &group->sibling_list, group_entry) {
                if (!is_software_event(event) &&
                    event->state != PERF_EVENT_STATE_OFF) {
                        if (n >= max_count)
@@ -742,7 +742,7 @@ int hw_perf_group_sched_in(struct perf_event *group_leader,
                return 0;
        cpuhw = &__get_cpu_var(cpu_hw_events);
        n0 = cpuhw->n_events;
-       n = collect_events(group_leader, ppmu->n_event - n0,
+       n = collect_events(group_leader, ppmu->n_counter - n0,
                           &cpuhw->event[n0], &cpuhw->events[n0],
                           &cpuhw->flags[n0]);
        if (n < 0)
@@ -764,7 +764,7 @@ int hw_perf_group_sched_in(struct perf_event *group_leader,
        cpuctx->active_oncpu += n;
        n = 1;
        event_sched_in(group_leader, cpu);
-       list_for_each_entry(sub, &group_leader->sibling_list, list_entry) {
+       list_for_each_entry(sub, &group_leader->sibling_list, group_entry) {
                if (sub->state != PERF_EVENT_STATE_OFF) {
                        event_sched_in(sub, cpu);
                        ++n;
@@ -797,7 +797,7 @@ static int power_pmu_enable(struct perf_event *event)
         */
        cpuhw = &__get_cpu_var(cpu_hw_events);
        n0 = cpuhw->n_events;
-       if (n0 >= ppmu->n_event)
+       if (n0 >= ppmu->n_counter)
                goto out;
        cpuhw->event[n0] = event;
        cpuhw->events[n0] = event->hw.config;
@@ -848,11 +848,11 @@ static void power_pmu_disable(struct perf_event *event)
                }
        }
        for (i = 0; i < cpuhw->n_limited; ++i)
-               if (event == cpuhw->limited_event[i])
+               if (event == cpuhw->limited_counter[i])
                        break;
        if (i < cpuhw->n_limited) {
                while (++i < cpuhw->n_limited) {
-                       cpuhw->limited_event[i-1] = cpuhw->limited_event[i];
+                       cpuhw->limited_counter[i-1] = cpuhw->limited_counter[i];
                        cpuhw->limited_hwidx[i-1] = cpuhw->limited_hwidx[i];
                }
                --cpuhw->n_limited;
@@ -1078,7 +1078,7 @@ const struct pmu *hw_perf_event_init(struct perf_event *event)
         */
        n = 0;
        if (event->group_leader != event) {
-               n = collect_events(event->group_leader, ppmu->n_event - 1,
+               n = collect_events(event->group_leader, ppmu->n_counter - 1,
                                   ctrs, events, cflags);
                if (n < 0)
                        return ERR_PTR(-EINVAL);
@@ -1230,7 +1230,7 @@ static void perf_event_interrupt(struct pt_regs *regs)
        int nmi;
 
        if (cpuhw->n_limited)
-               freeze_limited_events(cpuhw, mfspr(SPRN_PMC5),
+               freeze_limited_counters(cpuhw, mfspr(SPRN_PMC5),
                                        mfspr(SPRN_PMC6));
 
        perf_read_regs(regs);
@@ -1260,7 +1260,7 @@ static void perf_event_interrupt(struct pt_regs *regs)
         * Any that we processed in the previous loop will not be negative.
         */
        if (!found) {
-               for (i = 0; i < ppmu->n_event; ++i) {
+               for (i = 0; i < ppmu->n_counter; ++i) {
                        if (is_limited_pmc(i + 1))
                                continue;
                        val = read_pmc(i + 1);