arena: Refactor tcache_fill to batch fill from slab

Refactor tcache_fill, introducing a new function arena_slab_reg_alloc_batch,
which will fill multiple pointers from a slab.

There should be no functional changes here, but allows future optimization
on reg_alloc_batch.
This commit is contained in:
Dave Watson 2018-10-18 09:49:45 -07:00
parent 57553c3b1a
commit 4b82872ebf

View File

@ -268,6 +268,27 @@ arena_slab_reg_alloc(extent_t *slab, const bin_info_t *bin_info) {
return ret; return ret;
} }
static void
arena_slab_reg_alloc_batch(extent_t *slab, const bin_info_t *bin_info,
unsigned cnt, void** ptrs) {
arena_slab_data_t *slab_data = extent_slab_data_get(slab);
assert(extent_nfree_get(slab) > 0);
assert(!bitmap_full(slab_data->bitmap, &bin_info->bitmap_info));
size_t regind = 0;
for (unsigned i = 0; i < cnt; i++) {
void *ret;
regind = bitmap_sfu(slab_data->bitmap, &bin_info->bitmap_info);
ret = (void *)((uintptr_t)extent_addr_get(slab) +
(uintptr_t)(bin_info->reg_size * regind));
extent_nfree_dec(slab);
*(ptrs + i) = ret;
}
}
#ifndef JEMALLOC_JET #ifndef JEMALLOC_JET
static static
#endif #endif
@ -1286,7 +1307,7 @@ arena_bin_malloc_hard(tsdn_t *tsdn, arena_t *arena, bin_t *bin,
void void
arena_tcache_fill_small(tsdn_t *tsdn, arena_t *arena, tcache_t *tcache, arena_tcache_fill_small(tsdn_t *tsdn, arena_t *arena, tcache_t *tcache,
cache_bin_t *tbin, szind_t binind, uint64_t prof_accumbytes) { cache_bin_t *tbin, szind_t binind, uint64_t prof_accumbytes) {
unsigned i, nfill; unsigned i, nfill, cnt;
bin_t *bin; bin_t *bin;
assert(tbin->ncached == 0); assert(tbin->ncached == 0);
@ -1297,33 +1318,44 @@ arena_tcache_fill_small(tsdn_t *tsdn, arena_t *arena, tcache_t *tcache,
bin = &arena->bins[binind]; bin = &arena->bins[binind];
malloc_mutex_lock(tsdn, &bin->lock); malloc_mutex_lock(tsdn, &bin->lock);
for (i = 0, nfill = (tcache_bin_info[binind].ncached_max >> for (i = 0, nfill = (tcache_bin_info[binind].ncached_max >>
tcache->lg_fill_div[binind]); i < nfill; i++) { tcache->lg_fill_div[binind]); i < nfill; i += cnt) {
extent_t *slab; extent_t *slab;
void *ptr;
if ((slab = bin->slabcur) != NULL && extent_nfree_get(slab) > if ((slab = bin->slabcur) != NULL && extent_nfree_get(slab) >
0) { 0) {
ptr = arena_slab_reg_alloc(slab, &bin_infos[binind]); unsigned tofill = nfill - i;
cnt = tofill < extent_nfree_get(slab) ?
tofill : extent_nfree_get(slab);
arena_slab_reg_alloc_batch(
slab, &bin_infos[binind], cnt,
tbin->avail - nfill + i);
} else { } else {
ptr = arena_bin_malloc_hard(tsdn, arena, bin, binind); cnt = 1;
} void *ptr = arena_bin_malloc_hard(tsdn, arena, bin,
if (ptr == NULL) { binind);
/* /*
* OOM. tbin->avail isn't yet filled down to its first * OOM. tbin->avail isn't yet filled down to its first
* element, so the successful allocations (if any) must * element, so the successful allocations (if any) must
* be moved just before tbin->avail before bailing out. * be moved just before tbin->avail before bailing out.
*/ */
if (ptr == NULL) {
if (i > 0) { if (i > 0) {
memmove(tbin->avail - i, tbin->avail - nfill, memmove(tbin->avail - i,
tbin->avail - nfill,
i * sizeof(void *)); i * sizeof(void *));
} }
break; break;
} }
if (config_fill && unlikely(opt_junk_alloc)) {
arena_alloc_junk_small(ptr, &bin_infos[binind], true);
}
/* Insert such that low regions get used first. */ /* Insert such that low regions get used first. */
*(tbin->avail - nfill + i) = ptr; *(tbin->avail - nfill + i) = ptr;
} }
if (config_fill && unlikely(opt_junk_alloc)) {
for (unsigned j = 0; j < cnt; j++) {
void* ptr = *(tbin->avail - nfill + i + j);
arena_alloc_junk_small(ptr, &bin_infos[binind],
true);
}
}
}
if (config_stats) { if (config_stats) {
bin->stats.nmalloc += i; bin->stats.nmalloc += i;
bin->stats.nrequests += tbin->tstats.nrequests; bin->stats.nrequests += tbin->tstats.nrequests;