summaryrefslogtreecommitdiff
path: root/deps/jemalloc/include/jemalloc/internal/prof_inlines_a.h
diff options
context:
space:
mode:
Diffstat (limited to 'deps/jemalloc/include/jemalloc/internal/prof_inlines_a.h')
-rw-r--r--deps/jemalloc/include/jemalloc/internal/prof_inlines_a.h85
1 files changed, 85 insertions, 0 deletions
diff --git a/deps/jemalloc/include/jemalloc/internal/prof_inlines_a.h b/deps/jemalloc/include/jemalloc/internal/prof_inlines_a.h
new file mode 100644
index 000000000..471d9853c
--- /dev/null
+++ b/deps/jemalloc/include/jemalloc/internal/prof_inlines_a.h
@@ -0,0 +1,85 @@
+#ifndef JEMALLOC_INTERNAL_PROF_INLINES_A_H
+#define JEMALLOC_INTERNAL_PROF_INLINES_A_H
+
+#include "jemalloc/internal/mutex.h"
+
+static inline bool
+prof_accum_add(tsdn_t *tsdn, prof_accum_t *prof_accum,
+ uint64_t accumbytes) {
+ cassert(config_prof);
+
+ bool overflow;
+ uint64_t a0, a1;
+
+ /*
+ * If the application allocates fast enough (and/or if idump is slow
+ * enough), extreme overflow here (a1 >= prof_interval * 2) can cause
+ * idump trigger coalescing. This is an intentional mechanism that
+ * avoids rate-limiting allocation.
+ */
+#ifdef JEMALLOC_ATOMIC_U64
+ a0 = atomic_load_u64(&prof_accum->accumbytes, ATOMIC_RELAXED);
+ do {
+ a1 = a0 + accumbytes;
+ assert(a1 >= a0);
+ overflow = (a1 >= prof_interval);
+ if (overflow) {
+ a1 %= prof_interval;
+ }
+ } while (!atomic_compare_exchange_weak_u64(&prof_accum->accumbytes, &a0,
+ a1, ATOMIC_RELAXED, ATOMIC_RELAXED));
+#else
+ malloc_mutex_lock(tsdn, &prof_accum->mtx);
+ a0 = prof_accum->accumbytes;
+ a1 = a0 + accumbytes;
+ overflow = (a1 >= prof_interval);
+ if (overflow) {
+ a1 %= prof_interval;
+ }
+ prof_accum->accumbytes = a1;
+ malloc_mutex_unlock(tsdn, &prof_accum->mtx);
+#endif
+ return overflow;
+}
+
+static inline void
+prof_accum_cancel(tsdn_t *tsdn, prof_accum_t *prof_accum,
+ size_t usize) {
+ cassert(config_prof);
+
+ /*
+ * Cancel out as much of the excessive prof_accumbytes increase as
+ * possible without underflowing. Interval-triggered dumps occur
+ * slightly more often than intended as a result of incomplete
+ * canceling.
+ */
+ uint64_t a0, a1;
+#ifdef JEMALLOC_ATOMIC_U64
+ a0 = atomic_load_u64(&prof_accum->accumbytes, ATOMIC_RELAXED);
+ do {
+ a1 = (a0 >= SC_LARGE_MINCLASS - usize)
+ ? a0 - (SC_LARGE_MINCLASS - usize) : 0;
+ } while (!atomic_compare_exchange_weak_u64(&prof_accum->accumbytes, &a0,
+ a1, ATOMIC_RELAXED, ATOMIC_RELAXED));
+#else
+ malloc_mutex_lock(tsdn, &prof_accum->mtx);
+ a0 = prof_accum->accumbytes;
+ a1 = (a0 >= SC_LARGE_MINCLASS - usize)
+ ? a0 - (SC_LARGE_MINCLASS - usize) : 0;
+ prof_accum->accumbytes = a1;
+ malloc_mutex_unlock(tsdn, &prof_accum->mtx);
+#endif
+}
+
+JEMALLOC_ALWAYS_INLINE bool
+prof_active_get_unlocked(void) {
+ /*
+ * Even if opt_prof is true, sampling can be temporarily disabled by
+ * setting prof_active to false. No locking is used when reading
+ * prof_active in the fast path, so there are no guarantees regarding
+ * how long it will take for all threads to notice state changes.
+ */
+ return prof_active;
+}
+
+#endif /* JEMALLOC_INTERNAL_PROF_INLINES_A_H */