fc1c47043c
実装内容: - Phase 1a: Page size macro化 - TINY_C7_ULTRA_PAGE_SHIFT (16) を定義 - tiny_c7_ultra_page_of で division → bit shift に変更 - refill/free での seg_end 計算を multiplication → bit shift に最適化 - Phase 1b: Segment learning を移動 - segment learning を free初回 → alloc refill時に移動 - free側での unlikely segment_from_ptr call を削除 - normal pattern (alloc → free) での segment既学習を前提 ベンチマーク結果（Mixed 16-1024B, 1M iter, ws=400）: - Baseline: 39.5M ops/s - Phase 1a: 39.5M ops/s (誤差範囲) - Phase 1b: 42.3M ops/s - 最終平均: 43.9M ops/s (+11.1% = +4.4M ops/s) tiny_c7_ultra_page_of は計測では同じ値だが、実際には以下が改善: - division コスト削減（数cycle/call） - free時のsegment learning削除（per-thread 1回削減） - refill での計算簡素化 これにより全体の refill パス最適化が達成できました。
109 lines
3.2 KiB
C
109 lines
3.2 KiB
C
// tiny_c7_ultra_segment.c - C7 ULTRA 専用セグメント管理（UF-3）
|
||
|
||
#include "box/tiny_c7_ultra_segment_box.h"
|
||
|
||
#include <errno.h>
|
||
#include <stdint.h>
|
||
#include <stdlib.h>
|
||
#include <string.h>
|
||
#include <sys/mman.h>
|
||
|
||
// 2MiB セグメントを 64KiB ページに分割（C7 専用、pow2 で mask しやすく）
|
||
#define TINY_C7_ULTRA_SEG_SIZE ((size_t)(2 * 1024 * 1024))
|
||
#define TINY_C7_ULTRA_PAGE_SIZE ((size_t)(64 * 1024))
|
||
#define TINY_C7_ULTRA_PAGE_SHIFT 16 // 64KiB = 2^16 (for O(1) bit shift instead of division)
|
||
|
||
static __thread tiny_c7_ultra_segment_t* g_ultra_seg;
|
||
|
||
static inline void tiny_c7_ultra_segment_clear(tiny_c7_ultra_segment_t* seg) {
|
||
if (!seg) return;
|
||
seg->base = NULL;
|
||
seg->seg_size = 0;
|
||
seg->page_size = 0;
|
||
seg->num_pages = 0;
|
||
seg->pages = NULL;
|
||
}
|
||
|
||
tiny_c7_ultra_segment_t* tiny_c7_ultra_segment_acquire(void) {
|
||
if (g_ultra_seg) {
|
||
return g_ultra_seg;
|
||
}
|
||
|
||
tiny_c7_ultra_segment_t* seg =
|
||
(tiny_c7_ultra_segment_t*)calloc(1, sizeof(tiny_c7_ultra_segment_t));
|
||
if (!seg) {
|
||
return NULL;
|
||
}
|
||
seg->seg_size = TINY_C7_ULTRA_SEG_SIZE;
|
||
seg->page_size = TINY_C7_ULTRA_PAGE_SIZE;
|
||
seg->num_pages = (uint32_t)(seg->seg_size / seg->page_size);
|
||
|
||
seg->pages = (tiny_c7_ultra_page_meta_t*)calloc(seg->num_pages,
|
||
sizeof(tiny_c7_ultra_page_meta_t));
|
||
if (!seg->pages) {
|
||
free(seg);
|
||
return NULL;
|
||
}
|
||
|
||
void* base = mmap(NULL, seg->seg_size, PROT_READ | PROT_WRITE,
|
||
MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
|
||
if (base == MAP_FAILED) {
|
||
int saved = errno;
|
||
free(seg->pages);
|
||
free(seg);
|
||
(void)saved;
|
||
return NULL;
|
||
}
|
||
|
||
seg->base = base;
|
||
g_ultra_seg = seg;
|
||
return seg;
|
||
}
|
||
|
||
void tiny_c7_ultra_segment_release(tiny_c7_ultra_segment_t* seg) {
|
||
if (!seg) return;
|
||
if (seg->base && seg->seg_size) {
|
||
munmap(seg->base, seg->seg_size);
|
||
}
|
||
free(seg->pages);
|
||
if (seg == g_ultra_seg) {
|
||
g_ultra_seg = NULL;
|
||
}
|
||
free(seg);
|
||
}
|
||
|
||
tiny_c7_ultra_segment_t* tiny_c7_ultra_segment_from_ptr(void* p) {
|
||
tiny_c7_ultra_segment_t* seg = g_ultra_seg;
|
||
if (!seg || !seg->base || seg->seg_size == 0) return NULL;
|
||
uintptr_t base = (uintptr_t)seg->base;
|
||
uintptr_t addr = (uintptr_t)p;
|
||
if (addr < base || addr >= base + seg->seg_size) {
|
||
return NULL;
|
||
}
|
||
return seg;
|
||
}
|
||
|
||
tiny_c7_ultra_page_meta_t* tiny_c7_ultra_page_of(void* p,
|
||
tiny_c7_ultra_segment_t** out_seg,
|
||
uint32_t* out_page_idx) {
|
||
tiny_c7_ultra_segment_t* seg = tiny_c7_ultra_segment_from_ptr(p);
|
||
if (!seg) {
|
||
return NULL;
|
||
}
|
||
uintptr_t base = (uintptr_t)seg->base;
|
||
uintptr_t addr = (uintptr_t)p;
|
||
size_t offset = (size_t)(addr - base);
|
||
// Phase PERF-ULTRA-REFILL-OPT-1a: Replace division with bit shift for O(1) lookup
|
||
uint32_t idx = (uint32_t)(offset >> TINY_C7_ULTRA_PAGE_SHIFT);
|
||
if (idx >= seg->num_pages) {
|
||
return NULL;
|
||
}
|
||
if (out_seg) {
|
||
*out_seg = seg;
|
||
}
|
||
if (out_page_idx) {
|
||
*out_page_idx = idx;
|
||
}
|
||
return &seg->pages[idx];
|
||
}
|