diff options
Diffstat (limited to 'deps/jemalloc/include/jemalloc/internal/pa.h')
| -rw-r--r-- | deps/jemalloc/include/jemalloc/internal/pa.h | 243 |
1 files changed, 243 insertions, 0 deletions
diff --git a/deps/jemalloc/include/jemalloc/internal/pa.h b/deps/jemalloc/include/jemalloc/internal/pa.h new file mode 100644 index 000000000..4748a05b6 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/pa.h @@ -0,0 +1,243 @@ +#ifndef JEMALLOC_INTERNAL_PA_H +#define JEMALLOC_INTERNAL_PA_H + +#include "jemalloc/internal/base.h" +#include "jemalloc/internal/decay.h" +#include "jemalloc/internal/ecache.h" +#include "jemalloc/internal/edata_cache.h" +#include "jemalloc/internal/emap.h" +#include "jemalloc/internal/hpa.h" +#include "jemalloc/internal/lockedint.h" +#include "jemalloc/internal/pac.h" +#include "jemalloc/internal/pai.h" +#include "jemalloc/internal/sec.h" + +/* + * The page allocator; responsible for acquiring pages of memory for + * allocations. It picks the implementation of the page allocator interface + * (i.e. a pai_t) to handle a given page-level allocation request. For now, the + * only such implementation is the PAC code ("page allocator classic"), but + * others will be coming soon. + */ + +typedef struct pa_central_s pa_central_t; +struct pa_central_s { + hpa_central_t hpa; +}; + +/* + * The stats for a particular pa_shard. Because of the way the ctl module + * handles stats epoch data collection (it has its own arena_stats, and merges + * the stats from each arena into it), this needs to live in the arena_stats_t; + * hence we define it here and let the pa_shard have a pointer (rather than the + * more natural approach of just embedding it in the pa_shard itself). + * + * We follow the arena_stats_t approach of marking the derived fields. These + * are the ones that are not maintained on their own; instead, their values are + * derived during those stats merges. + */ +typedef struct pa_shard_stats_s pa_shard_stats_t; +struct pa_shard_stats_s { + /* Number of edata_t structs allocated by base, but not being used. */ + size_t edata_avail; /* Derived. */ + /* + * Stats specific to the PAC. For now, these are the only stats that + * exist, but there will eventually be other page allocators. Things + * like edata_avail make sense in a cross-PA sense, but things like + * npurges don't. + */ + pac_stats_t pac_stats; +}; + +/* + * The local allocator handle. Keeps the state necessary to satisfy page-sized + * allocations. + * + * The contents are mostly internal to the PA module. The key exception is that + * arena decay code is allowed to grab pointers to the dirty and muzzy ecaches + * decay_ts, for a couple of queries, passing them back to a PA function, or + * acquiring decay.mtx and looking at decay.purging. The reasoning is that, + * while PA decides what and how to purge, the arena code decides when and where + * (e.g. on what thread). It's allowed to use the presence of another purger to + * decide. + * (The background thread code also touches some other decay internals, but + * that's not fundamental; its' just an artifact of a partial refactoring, and + * its accesses could be straightforwardly moved inside the decay module). + */ +typedef struct pa_shard_s pa_shard_t; +struct pa_shard_s { + /* The central PA this shard is associated with. */ + pa_central_t *central; + + /* + * Number of pages in active extents. + * + * Synchronization: atomic. + */ + atomic_zu_t nactive; + + /* + * Whether or not we should prefer the hugepage allocator. Atomic since + * it may be concurrently modified by a thread setting extent hooks. + * Note that we still may do HPA operations in this arena; if use_hpa is + * changed from true to false, we'll free back to the hugepage allocator + * for those allocations. + */ + atomic_b_t use_hpa; + + /* + * If we never used the HPA to begin with, it wasn't initialized, and so + * we shouldn't try to e.g. acquire its mutexes during fork. This + * tracks that knowledge. + */ + bool ever_used_hpa; + + /* Allocates from a PAC. */ + pac_t pac; + + /* + * We place a small extent cache in front of the HPA, since we intend + * these configurations to use many fewer arenas, and therefore have a + * higher risk of hot locks. + */ + sec_t hpa_sec; + hpa_shard_t hpa_shard; + + /* The source of edata_t objects. */ + edata_cache_t edata_cache; + + unsigned ind; + + malloc_mutex_t *stats_mtx; + pa_shard_stats_t *stats; + + /* The emap this shard is tied to. */ + emap_t *emap; + + /* The base from which we get the ehooks and allocate metadat. */ + base_t *base; +}; + +static inline bool +pa_shard_dont_decay_muzzy(pa_shard_t *shard) { + return ecache_npages_get(&shard->pac.ecache_muzzy) == 0 && + pac_decay_ms_get(&shard->pac, extent_state_muzzy) <= 0; +} + +static inline ehooks_t * +pa_shard_ehooks_get(pa_shard_t *shard) { + return base_ehooks_get(shard->base); +} + +/* Returns true on error. */ +bool pa_central_init(pa_central_t *central, base_t *base, bool hpa, + hpa_hooks_t *hpa_hooks); + +/* Returns true on error. */ +bool pa_shard_init(tsdn_t *tsdn, pa_shard_t *shard, pa_central_t *central, + emap_t *emap, base_t *base, unsigned ind, pa_shard_stats_t *stats, + malloc_mutex_t *stats_mtx, nstime_t *cur_time, size_t oversize_threshold, + ssize_t dirty_decay_ms, ssize_t muzzy_decay_ms); + +/* + * This isn't exposed to users; we allow late enablement of the HPA shard so + * that we can boot without worrying about the HPA, then turn it on in a0. + */ +bool pa_shard_enable_hpa(tsdn_t *tsdn, pa_shard_t *shard, + const hpa_shard_opts_t *hpa_opts, const sec_opts_t *hpa_sec_opts); + +/* + * We stop using the HPA when custom extent hooks are installed, but still + * redirect deallocations to it. + */ +void pa_shard_disable_hpa(tsdn_t *tsdn, pa_shard_t *shard); + +/* + * This does the PA-specific parts of arena reset (i.e. freeing all active + * allocations). + */ +void pa_shard_reset(tsdn_t *tsdn, pa_shard_t *shard); + +/* + * Destroy all the remaining retained extents. Should only be called after + * decaying all active, dirty, and muzzy extents to the retained state, as the + * last step in destroying the shard. + */ +void pa_shard_destroy(tsdn_t *tsdn, pa_shard_t *shard); + +/* Gets an edata for the given allocation. */ +edata_t *pa_alloc(tsdn_t *tsdn, pa_shard_t *shard, size_t size, + size_t alignment, bool slab, szind_t szind, bool zero, bool guarded, + bool *deferred_work_generated); +/* Returns true on error, in which case nothing changed. */ +bool pa_expand(tsdn_t *tsdn, pa_shard_t *shard, edata_t *edata, size_t old_size, + size_t new_size, szind_t szind, bool zero, bool *deferred_work_generated); +/* + * The same. Sets *generated_dirty to true if we produced new dirty pages, and + * false otherwise. + */ +bool pa_shrink(tsdn_t *tsdn, pa_shard_t *shard, edata_t *edata, size_t old_size, + size_t new_size, szind_t szind, bool *deferred_work_generated); +/* + * Frees the given edata back to the pa. Sets *generated_dirty if we produced + * new dirty pages (well, we always set it for now; but this need not be the + * case). + * (We could make generated_dirty the return value of course, but this is more + * consistent with the shrink pathway and our error codes here). + */ +void pa_dalloc(tsdn_t *tsdn, pa_shard_t *shard, edata_t *edata, + bool *deferred_work_generated); +bool pa_decay_ms_set(tsdn_t *tsdn, pa_shard_t *shard, extent_state_t state, + ssize_t decay_ms, pac_purge_eagerness_t eagerness); +ssize_t pa_decay_ms_get(pa_shard_t *shard, extent_state_t state); + +/* + * Do deferred work on this PA shard. + * + * Morally, this should do both PAC decay and the HPA deferred work. For now, + * though, the arena, background thread, and PAC modules are tightly interwoven + * in a way that's tricky to extricate, so we only do the HPA-specific parts. + */ +void pa_shard_set_deferral_allowed(tsdn_t *tsdn, pa_shard_t *shard, + bool deferral_allowed); +void pa_shard_do_deferred_work(tsdn_t *tsdn, pa_shard_t *shard); +void pa_shard_try_deferred_work(tsdn_t *tsdn, pa_shard_t *shard); +uint64_t pa_shard_time_until_deferred_work(tsdn_t *tsdn, pa_shard_t *shard); + +/******************************************************************************/ +/* + * Various bits of "boring" functionality that are still part of this module, + * but that we relegate to pa_extra.c, to keep the core logic in pa.c as + * readable as possible. + */ + +/* + * These fork phases are synchronized with the arena fork phase numbering to + * make it easy to keep straight. That's why there's no prefork1. + */ +void pa_shard_prefork0(tsdn_t *tsdn, pa_shard_t *shard); +void pa_shard_prefork2(tsdn_t *tsdn, pa_shard_t *shard); +void pa_shard_prefork3(tsdn_t *tsdn, pa_shard_t *shard); +void pa_shard_prefork4(tsdn_t *tsdn, pa_shard_t *shard); +void pa_shard_prefork5(tsdn_t *tsdn, pa_shard_t *shard); +void pa_shard_postfork_parent(tsdn_t *tsdn, pa_shard_t *shard); +void pa_shard_postfork_child(tsdn_t *tsdn, pa_shard_t *shard); + +void pa_shard_basic_stats_merge(pa_shard_t *shard, size_t *nactive, + size_t *ndirty, size_t *nmuzzy); + +void pa_shard_stats_merge(tsdn_t *tsdn, pa_shard_t *shard, + pa_shard_stats_t *pa_shard_stats_out, pac_estats_t *estats_out, + hpa_shard_stats_t *hpa_stats_out, sec_stats_t *sec_stats_out, + size_t *resident); + +/* + * Reads the PA-owned mutex stats into the output stats array, at the + * appropriate positions. Morally, these stats should really live in + * pa_shard_stats_t, but the indices are sort of baked into the various mutex + * prof macros. This would be a good thing to do at some point. + */ +void pa_shard_mtx_stats_read(tsdn_t *tsdn, pa_shard_t *shard, + mutex_prof_data_t mutex_prof_data[mutex_prof_num_arena_mutexes]); + +#endif /* JEMALLOC_INTERNAL_PA_H */ |