2025-11-05 12:31:14 +09:00
|
|
|
// hakmem_tiny_hot_pop.inc.h
|
|
|
|
|
// Phase 2D-1: Hot-path inline functions - Specialized class pop operations
|
|
|
|
|
//
|
|
|
|
|
// This file contains per-class specialized hot-path pop functions for classes 0-3.
|
|
|
|
|
// These functions are extracted from hakmem_tiny.c to improve maintainability and
|
|
|
|
|
// reduce the main file size by approximately 84 lines (lines 407-494).
|
|
|
|
|
//
|
|
|
|
|
// Each function handles:
|
|
|
|
|
// - Optional ultra bump shadow allocation
|
|
|
|
|
// - Fast cache pop with explicit capacity checks
|
|
|
|
|
// - Branchless operation for maximum performance
|
|
|
|
|
|
|
|
|
|
#ifndef HAKMEM_TINY_HOT_POP_INC_H
|
|
|
|
|
#define HAKMEM_TINY_HOT_POP_INC_H
|
|
|
|
|
|
|
|
|
|
#include "hakmem_tiny.h"
|
Phase E3-FINAL: Fix Box API offset bugs - ALL classes now use correct offsets
## Root Cause Analysis (GPT5)
**Physical Layout Constraints**:
- Class 0: 8B = [1B header][7B payload] → offset 1 = 9B needed = ❌ IMPOSSIBLE
- Class 1-6: >=16B = [1B header][15B+ payload] → offset 1 = ✅ POSSIBLE
- Class 7: 1KB → offset 0 (compatibility)
**Correct Specification**:
- HAKMEM_TINY_HEADER_CLASSIDX != 0:
- Class 0, 7: next at offset 0 (overwrites header when on freelist)
- Class 1-6: next at offset 1 (after header)
- HAKMEM_TINY_HEADER_CLASSIDX == 0:
- All classes: next at offset 0
**Previous Bug**:
- Attempted "ALL classes offset 1" unification
- Class 0 with offset 1 caused immediate SEGV (9B > 8B block size)
- Mixed 2-arg/3-arg API caused confusion
## Fixes Applied
### 1. Restored 3-Argument Box API (core/box/tiny_next_ptr_box.h)
```c
// Correct signatures
void tiny_next_write(int class_idx, void* base, void* next_value)
void* tiny_next_read(int class_idx, const void* base)
// Correct offset calculation
size_t offset = (class_idx == 0 || class_idx == 7) ? 0 : 1;
```
### 2. Updated 123+ Call Sites Across 34 Files
- hakmem_tiny_hot_pop_v4.inc.h (4 locations)
- hakmem_tiny_fastcache.inc.h (3 locations)
- hakmem_tiny_tls_list.h (12 locations)
- superslab_inline.h (5 locations)
- tiny_fastcache.h (3 locations)
- ptr_trace.h (macro definitions)
- tls_sll_box.h (2 locations)
- + 27 additional files
Pattern: `tiny_next_read(base)` → `tiny_next_read(class_idx, base)`
Pattern: `tiny_next_write(base, next)` → `tiny_next_write(class_idx, base, next)`
### 3. Added Sentinel Detection Guards
- tiny_fast_push(): Block nodes with sentinel in ptr or ptr->next
- tls_list_push(): Block nodes with sentinel in ptr or ptr->next
- Defense-in-depth against remote free sentinel leakage
## Verification (GPT5 Report)
**Test Command**: `./out/release/bench_random_mixed_hakmem --iterations=70000`
**Results**:
- ✅ Main loop completed successfully
- ✅ Drain phase completed successfully
- ✅ NO SEGV (previous crash at iteration 66151 is FIXED)
- ℹ️ Final log: "tiny_alloc(1024) failed" is normal fallback to Mid/ACE layers
**Analysis**:
- Class 0 immediate SEGV: ✅ RESOLVED (correct offset 0 now used)
- 66K iteration crash: ✅ RESOLVED (offset consistency fixed)
- Box API conflicts: ✅ RESOLVED (unified 3-arg API)
## Technical Details
### Offset Logic Justification
```
Class 0: 8B block → next pointer (8B) fits ONLY at offset 0
Class 1: 16B block → next pointer (8B) fits at offset 1 (after 1B header)
Class 2: 32B block → next pointer (8B) fits at offset 1
...
Class 6: 512B block → next pointer (8B) fits at offset 1
Class 7: 1024B block → offset 0 for legacy compatibility
```
### Files Modified (Summary)
- Core API: `box/tiny_next_ptr_box.h`
- Hot paths: `hakmem_tiny_hot_pop*.inc.h`, `tiny_fastcache.h`
- TLS layers: `hakmem_tiny_tls_list.h`, `hakmem_tiny_tls_ops.h`
- SuperSlab: `superslab_inline.h`, `tiny_superslab_*.inc.h`
- Refill: `hakmem_tiny_refill.inc.h`, `tiny_refill_opt.h`
- Free paths: `tiny_free_magazine.inc.h`, `tiny_superslab_free.inc.h`
- Documentation: Multiple Phase E3 reports
## Remaining Work
None for Box API offset bugs - all structural issues resolved.
Future enhancements (non-critical):
- Periodic `grep -R '*(void**)' core/` to detect direct pointer access violations
- Enforce Box API usage via static analysis
- Document offset rationale in architecture docs
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-13 06:50:20 +09:00
|
|
|
#include "box/tiny_next_ptr_box.h"
|
2025-11-05 12:31:14 +09:00
|
|
|
#include <stdint.h>
|
|
|
|
|
|
|
|
|
|
// External TLS variables used by hot-path functions
|
|
|
|
|
extern int g_ultra_bump_shadow;
|
|
|
|
|
extern int g_fast_enable;
|
|
|
|
|
extern uint16_t g_fast_cap[TINY_NUM_CLASSES];
|
|
|
|
|
extern __thread void* g_fast_head[TINY_NUM_CLASSES];
|
|
|
|
|
extern __thread uint16_t g_fast_count[TINY_NUM_CLASSES];
|
|
|
|
|
|
|
|
|
|
// Forward declaration for ultra bump shadow support
|
|
|
|
|
static inline void* superslab_tls_bump_fast(int class_idx);
|
|
|
|
|
|
|
|
|
|
// Specialized hot-tier pop (branchless per-class) — used for top hot classes (8/16/32B)
|
|
|
|
|
static inline __attribute__((always_inline)) void* tiny_hot_pop_class0(void) {
|
|
|
|
|
if (__builtin_expect(g_ultra_bump_shadow != 0, 0)) {
|
|
|
|
|
void* bump = superslab_tls_bump_fast(0);
|
|
|
|
|
if (__builtin_expect(bump != NULL, 1)) {
|
|
|
|
|
return bump;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
if (__builtin_expect(!g_fast_enable, 0)) return NULL;
|
|
|
|
|
uint16_t cap = g_fast_cap[0];
|
|
|
|
|
if (__builtin_expect(cap == 0, 0)) return NULL;
|
|
|
|
|
void* head = g_fast_head[0];
|
|
|
|
|
if (__builtin_expect(head == NULL, 0)) return NULL;
|
2025-11-10 18:04:08 +09:00
|
|
|
// Phase 7: header-aware next pointer (C0-C6: base+1, C7: base)
|
Phase E3-FINAL: Fix Box API offset bugs - ALL classes now use correct offsets
## Root Cause Analysis (GPT5)
**Physical Layout Constraints**:
- Class 0: 8B = [1B header][7B payload] → offset 1 = 9B needed = ❌ IMPOSSIBLE
- Class 1-6: >=16B = [1B header][15B+ payload] → offset 1 = ✅ POSSIBLE
- Class 7: 1KB → offset 0 (compatibility)
**Correct Specification**:
- HAKMEM_TINY_HEADER_CLASSIDX != 0:
- Class 0, 7: next at offset 0 (overwrites header when on freelist)
- Class 1-6: next at offset 1 (after header)
- HAKMEM_TINY_HEADER_CLASSIDX == 0:
- All classes: next at offset 0
**Previous Bug**:
- Attempted "ALL classes offset 1" unification
- Class 0 with offset 1 caused immediate SEGV (9B > 8B block size)
- Mixed 2-arg/3-arg API caused confusion
## Fixes Applied
### 1. Restored 3-Argument Box API (core/box/tiny_next_ptr_box.h)
```c
// Correct signatures
void tiny_next_write(int class_idx, void* base, void* next_value)
void* tiny_next_read(int class_idx, const void* base)
// Correct offset calculation
size_t offset = (class_idx == 0 || class_idx == 7) ? 0 : 1;
```
### 2. Updated 123+ Call Sites Across 34 Files
- hakmem_tiny_hot_pop_v4.inc.h (4 locations)
- hakmem_tiny_fastcache.inc.h (3 locations)
- hakmem_tiny_tls_list.h (12 locations)
- superslab_inline.h (5 locations)
- tiny_fastcache.h (3 locations)
- ptr_trace.h (macro definitions)
- tls_sll_box.h (2 locations)
- + 27 additional files
Pattern: `tiny_next_read(base)` → `tiny_next_read(class_idx, base)`
Pattern: `tiny_next_write(base, next)` → `tiny_next_write(class_idx, base, next)`
### 3. Added Sentinel Detection Guards
- tiny_fast_push(): Block nodes with sentinel in ptr or ptr->next
- tls_list_push(): Block nodes with sentinel in ptr or ptr->next
- Defense-in-depth against remote free sentinel leakage
## Verification (GPT5 Report)
**Test Command**: `./out/release/bench_random_mixed_hakmem --iterations=70000`
**Results**:
- ✅ Main loop completed successfully
- ✅ Drain phase completed successfully
- ✅ NO SEGV (previous crash at iteration 66151 is FIXED)
- ℹ️ Final log: "tiny_alloc(1024) failed" is normal fallback to Mid/ACE layers
**Analysis**:
- Class 0 immediate SEGV: ✅ RESOLVED (correct offset 0 now used)
- 66K iteration crash: ✅ RESOLVED (offset consistency fixed)
- Box API conflicts: ✅ RESOLVED (unified 3-arg API)
## Technical Details
### Offset Logic Justification
```
Class 0: 8B block → next pointer (8B) fits ONLY at offset 0
Class 1: 16B block → next pointer (8B) fits at offset 1 (after 1B header)
Class 2: 32B block → next pointer (8B) fits at offset 1
...
Class 6: 512B block → next pointer (8B) fits at offset 1
Class 7: 1024B block → offset 0 for legacy compatibility
```
### Files Modified (Summary)
- Core API: `box/tiny_next_ptr_box.h`
- Hot paths: `hakmem_tiny_hot_pop*.inc.h`, `tiny_fastcache.h`
- TLS layers: `hakmem_tiny_tls_list.h`, `hakmem_tiny_tls_ops.h`
- SuperSlab: `superslab_inline.h`, `tiny_superslab_*.inc.h`
- Refill: `hakmem_tiny_refill.inc.h`, `tiny_refill_opt.h`
- Free paths: `tiny_free_magazine.inc.h`, `tiny_superslab_free.inc.h`
- Documentation: Multiple Phase E3 reports
## Remaining Work
None for Box API offset bugs - all structural issues resolved.
Future enhancements (non-critical):
- Periodic `grep -R '*(void**)' core/` to detect direct pointer access violations
- Enforce Box API usage via static analysis
- Document offset rationale in architecture docs
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-13 06:50:20 +09:00
|
|
|
g_fast_head[0] = tiny_next_read(0, head);
|
2025-11-05 12:31:14 +09:00
|
|
|
uint16_t count = g_fast_count[0];
|
|
|
|
|
if (count > 0) {
|
|
|
|
|
g_fast_count[0] = (uint16_t)(count - 1);
|
|
|
|
|
} else {
|
|
|
|
|
g_fast_count[0] = 0;
|
|
|
|
|
}
|
2025-11-10 18:04:08 +09:00
|
|
|
// No C7 here (class 0), just return base
|
2025-11-05 12:31:14 +09:00
|
|
|
return head;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static inline __attribute__((always_inline)) void* tiny_hot_pop_class1(void) {
|
|
|
|
|
if (__builtin_expect(g_ultra_bump_shadow != 0, 0)) {
|
|
|
|
|
void* bump = superslab_tls_bump_fast(1);
|
|
|
|
|
if (__builtin_expect(bump != NULL, 1)) {
|
|
|
|
|
return bump;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
if (__builtin_expect(!g_fast_enable, 0)) return NULL;
|
|
|
|
|
uint16_t cap = g_fast_cap[1];
|
|
|
|
|
if (__builtin_expect(cap == 0, 0)) return NULL;
|
|
|
|
|
void* head = g_fast_head[1];
|
|
|
|
|
if (__builtin_expect(head == NULL, 0)) return NULL;
|
2025-11-10 18:04:08 +09:00
|
|
|
// Phase 7: header-aware next pointer (C0-C6: base+1)
|
Phase E3-FINAL: Fix Box API offset bugs - ALL classes now use correct offsets
## Root Cause Analysis (GPT5)
**Physical Layout Constraints**:
- Class 0: 8B = [1B header][7B payload] → offset 1 = 9B needed = ❌ IMPOSSIBLE
- Class 1-6: >=16B = [1B header][15B+ payload] → offset 1 = ✅ POSSIBLE
- Class 7: 1KB → offset 0 (compatibility)
**Correct Specification**:
- HAKMEM_TINY_HEADER_CLASSIDX != 0:
- Class 0, 7: next at offset 0 (overwrites header when on freelist)
- Class 1-6: next at offset 1 (after header)
- HAKMEM_TINY_HEADER_CLASSIDX == 0:
- All classes: next at offset 0
**Previous Bug**:
- Attempted "ALL classes offset 1" unification
- Class 0 with offset 1 caused immediate SEGV (9B > 8B block size)
- Mixed 2-arg/3-arg API caused confusion
## Fixes Applied
### 1. Restored 3-Argument Box API (core/box/tiny_next_ptr_box.h)
```c
// Correct signatures
void tiny_next_write(int class_idx, void* base, void* next_value)
void* tiny_next_read(int class_idx, const void* base)
// Correct offset calculation
size_t offset = (class_idx == 0 || class_idx == 7) ? 0 : 1;
```
### 2. Updated 123+ Call Sites Across 34 Files
- hakmem_tiny_hot_pop_v4.inc.h (4 locations)
- hakmem_tiny_fastcache.inc.h (3 locations)
- hakmem_tiny_tls_list.h (12 locations)
- superslab_inline.h (5 locations)
- tiny_fastcache.h (3 locations)
- ptr_trace.h (macro definitions)
- tls_sll_box.h (2 locations)
- + 27 additional files
Pattern: `tiny_next_read(base)` → `tiny_next_read(class_idx, base)`
Pattern: `tiny_next_write(base, next)` → `tiny_next_write(class_idx, base, next)`
### 3. Added Sentinel Detection Guards
- tiny_fast_push(): Block nodes with sentinel in ptr or ptr->next
- tls_list_push(): Block nodes with sentinel in ptr or ptr->next
- Defense-in-depth against remote free sentinel leakage
## Verification (GPT5 Report)
**Test Command**: `./out/release/bench_random_mixed_hakmem --iterations=70000`
**Results**:
- ✅ Main loop completed successfully
- ✅ Drain phase completed successfully
- ✅ NO SEGV (previous crash at iteration 66151 is FIXED)
- ℹ️ Final log: "tiny_alloc(1024) failed" is normal fallback to Mid/ACE layers
**Analysis**:
- Class 0 immediate SEGV: ✅ RESOLVED (correct offset 0 now used)
- 66K iteration crash: ✅ RESOLVED (offset consistency fixed)
- Box API conflicts: ✅ RESOLVED (unified 3-arg API)
## Technical Details
### Offset Logic Justification
```
Class 0: 8B block → next pointer (8B) fits ONLY at offset 0
Class 1: 16B block → next pointer (8B) fits at offset 1 (after 1B header)
Class 2: 32B block → next pointer (8B) fits at offset 1
...
Class 6: 512B block → next pointer (8B) fits at offset 1
Class 7: 1024B block → offset 0 for legacy compatibility
```
### Files Modified (Summary)
- Core API: `box/tiny_next_ptr_box.h`
- Hot paths: `hakmem_tiny_hot_pop*.inc.h`, `tiny_fastcache.h`
- TLS layers: `hakmem_tiny_tls_list.h`, `hakmem_tiny_tls_ops.h`
- SuperSlab: `superslab_inline.h`, `tiny_superslab_*.inc.h`
- Refill: `hakmem_tiny_refill.inc.h`, `tiny_refill_opt.h`
- Free paths: `tiny_free_magazine.inc.h`, `tiny_superslab_free.inc.h`
- Documentation: Multiple Phase E3 reports
## Remaining Work
None for Box API offset bugs - all structural issues resolved.
Future enhancements (non-critical):
- Periodic `grep -R '*(void**)' core/` to detect direct pointer access violations
- Enforce Box API usage via static analysis
- Document offset rationale in architecture docs
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-13 06:50:20 +09:00
|
|
|
g_fast_head[1] = tiny_next_read(1, head);
|
2025-11-05 12:31:14 +09:00
|
|
|
uint16_t count = g_fast_count[1];
|
|
|
|
|
if (count > 0) {
|
|
|
|
|
g_fast_count[1] = (uint16_t)(count - 1);
|
|
|
|
|
} else {
|
|
|
|
|
g_fast_count[1] = 0;
|
|
|
|
|
}
|
|
|
|
|
return head;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static inline __attribute__((always_inline)) void* tiny_hot_pop_class2(void) {
|
|
|
|
|
if (__builtin_expect(g_ultra_bump_shadow != 0, 0)) {
|
|
|
|
|
void* bump = superslab_tls_bump_fast(2);
|
|
|
|
|
if (__builtin_expect(bump != NULL, 1)) {
|
|
|
|
|
return bump;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
if (__builtin_expect(!g_fast_enable, 0)) return NULL;
|
|
|
|
|
uint16_t cap = g_fast_cap[2];
|
|
|
|
|
if (__builtin_expect(cap == 0, 0)) return NULL;
|
|
|
|
|
void* head = g_fast_head[2];
|
|
|
|
|
if (__builtin_expect(head == NULL, 0)) return NULL;
|
2025-11-10 18:04:08 +09:00
|
|
|
// Phase 7: header-aware next pointer (C0-C6: base+1)
|
Phase E3-FINAL: Fix Box API offset bugs - ALL classes now use correct offsets
## Root Cause Analysis (GPT5)
**Physical Layout Constraints**:
- Class 0: 8B = [1B header][7B payload] → offset 1 = 9B needed = ❌ IMPOSSIBLE
- Class 1-6: >=16B = [1B header][15B+ payload] → offset 1 = ✅ POSSIBLE
- Class 7: 1KB → offset 0 (compatibility)
**Correct Specification**:
- HAKMEM_TINY_HEADER_CLASSIDX != 0:
- Class 0, 7: next at offset 0 (overwrites header when on freelist)
- Class 1-6: next at offset 1 (after header)
- HAKMEM_TINY_HEADER_CLASSIDX == 0:
- All classes: next at offset 0
**Previous Bug**:
- Attempted "ALL classes offset 1" unification
- Class 0 with offset 1 caused immediate SEGV (9B > 8B block size)
- Mixed 2-arg/3-arg API caused confusion
## Fixes Applied
### 1. Restored 3-Argument Box API (core/box/tiny_next_ptr_box.h)
```c
// Correct signatures
void tiny_next_write(int class_idx, void* base, void* next_value)
void* tiny_next_read(int class_idx, const void* base)
// Correct offset calculation
size_t offset = (class_idx == 0 || class_idx == 7) ? 0 : 1;
```
### 2. Updated 123+ Call Sites Across 34 Files
- hakmem_tiny_hot_pop_v4.inc.h (4 locations)
- hakmem_tiny_fastcache.inc.h (3 locations)
- hakmem_tiny_tls_list.h (12 locations)
- superslab_inline.h (5 locations)
- tiny_fastcache.h (3 locations)
- ptr_trace.h (macro definitions)
- tls_sll_box.h (2 locations)
- + 27 additional files
Pattern: `tiny_next_read(base)` → `tiny_next_read(class_idx, base)`
Pattern: `tiny_next_write(base, next)` → `tiny_next_write(class_idx, base, next)`
### 3. Added Sentinel Detection Guards
- tiny_fast_push(): Block nodes with sentinel in ptr or ptr->next
- tls_list_push(): Block nodes with sentinel in ptr or ptr->next
- Defense-in-depth against remote free sentinel leakage
## Verification (GPT5 Report)
**Test Command**: `./out/release/bench_random_mixed_hakmem --iterations=70000`
**Results**:
- ✅ Main loop completed successfully
- ✅ Drain phase completed successfully
- ✅ NO SEGV (previous crash at iteration 66151 is FIXED)
- ℹ️ Final log: "tiny_alloc(1024) failed" is normal fallback to Mid/ACE layers
**Analysis**:
- Class 0 immediate SEGV: ✅ RESOLVED (correct offset 0 now used)
- 66K iteration crash: ✅ RESOLVED (offset consistency fixed)
- Box API conflicts: ✅ RESOLVED (unified 3-arg API)
## Technical Details
### Offset Logic Justification
```
Class 0: 8B block → next pointer (8B) fits ONLY at offset 0
Class 1: 16B block → next pointer (8B) fits at offset 1 (after 1B header)
Class 2: 32B block → next pointer (8B) fits at offset 1
...
Class 6: 512B block → next pointer (8B) fits at offset 1
Class 7: 1024B block → offset 0 for legacy compatibility
```
### Files Modified (Summary)
- Core API: `box/tiny_next_ptr_box.h`
- Hot paths: `hakmem_tiny_hot_pop*.inc.h`, `tiny_fastcache.h`
- TLS layers: `hakmem_tiny_tls_list.h`, `hakmem_tiny_tls_ops.h`
- SuperSlab: `superslab_inline.h`, `tiny_superslab_*.inc.h`
- Refill: `hakmem_tiny_refill.inc.h`, `tiny_refill_opt.h`
- Free paths: `tiny_free_magazine.inc.h`, `tiny_superslab_free.inc.h`
- Documentation: Multiple Phase E3 reports
## Remaining Work
None for Box API offset bugs - all structural issues resolved.
Future enhancements (non-critical):
- Periodic `grep -R '*(void**)' core/` to detect direct pointer access violations
- Enforce Box API usage via static analysis
- Document offset rationale in architecture docs
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-13 06:50:20 +09:00
|
|
|
g_fast_head[2] = tiny_next_read(2, head);
|
2025-11-05 12:31:14 +09:00
|
|
|
uint16_t count = g_fast_count[2];
|
|
|
|
|
if (count > 0) {
|
|
|
|
|
g_fast_count[2] = (uint16_t)(count - 1);
|
|
|
|
|
} else {
|
|
|
|
|
g_fast_count[2] = 0;
|
|
|
|
|
}
|
|
|
|
|
return head;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static inline __attribute__((always_inline)) void* tiny_hot_pop_class3(void) {
|
|
|
|
|
if (__builtin_expect(g_ultra_bump_shadow != 0, 0)) {
|
|
|
|
|
void* bump = superslab_tls_bump_fast(3);
|
|
|
|
|
if (__builtin_expect(bump != NULL, 1)) {
|
|
|
|
|
return bump;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
if (__builtin_expect(!g_fast_enable, 0)) return NULL;
|
|
|
|
|
uint16_t cap = g_fast_cap[3];
|
|
|
|
|
if (__builtin_expect(cap == 0, 0)) return NULL;
|
|
|
|
|
void* head = g_fast_head[3];
|
|
|
|
|
if (__builtin_expect(head == NULL, 0)) return NULL;
|
2025-11-10 18:04:08 +09:00
|
|
|
// Phase 7: header-aware next pointer (C0-C6: base+1)
|
Phase E3-FINAL: Fix Box API offset bugs - ALL classes now use correct offsets
## Root Cause Analysis (GPT5)
**Physical Layout Constraints**:
- Class 0: 8B = [1B header][7B payload] → offset 1 = 9B needed = ❌ IMPOSSIBLE
- Class 1-6: >=16B = [1B header][15B+ payload] → offset 1 = ✅ POSSIBLE
- Class 7: 1KB → offset 0 (compatibility)
**Correct Specification**:
- HAKMEM_TINY_HEADER_CLASSIDX != 0:
- Class 0, 7: next at offset 0 (overwrites header when on freelist)
- Class 1-6: next at offset 1 (after header)
- HAKMEM_TINY_HEADER_CLASSIDX == 0:
- All classes: next at offset 0
**Previous Bug**:
- Attempted "ALL classes offset 1" unification
- Class 0 with offset 1 caused immediate SEGV (9B > 8B block size)
- Mixed 2-arg/3-arg API caused confusion
## Fixes Applied
### 1. Restored 3-Argument Box API (core/box/tiny_next_ptr_box.h)
```c
// Correct signatures
void tiny_next_write(int class_idx, void* base, void* next_value)
void* tiny_next_read(int class_idx, const void* base)
// Correct offset calculation
size_t offset = (class_idx == 0 || class_idx == 7) ? 0 : 1;
```
### 2. Updated 123+ Call Sites Across 34 Files
- hakmem_tiny_hot_pop_v4.inc.h (4 locations)
- hakmem_tiny_fastcache.inc.h (3 locations)
- hakmem_tiny_tls_list.h (12 locations)
- superslab_inline.h (5 locations)
- tiny_fastcache.h (3 locations)
- ptr_trace.h (macro definitions)
- tls_sll_box.h (2 locations)
- + 27 additional files
Pattern: `tiny_next_read(base)` → `tiny_next_read(class_idx, base)`
Pattern: `tiny_next_write(base, next)` → `tiny_next_write(class_idx, base, next)`
### 3. Added Sentinel Detection Guards
- tiny_fast_push(): Block nodes with sentinel in ptr or ptr->next
- tls_list_push(): Block nodes with sentinel in ptr or ptr->next
- Defense-in-depth against remote free sentinel leakage
## Verification (GPT5 Report)
**Test Command**: `./out/release/bench_random_mixed_hakmem --iterations=70000`
**Results**:
- ✅ Main loop completed successfully
- ✅ Drain phase completed successfully
- ✅ NO SEGV (previous crash at iteration 66151 is FIXED)
- ℹ️ Final log: "tiny_alloc(1024) failed" is normal fallback to Mid/ACE layers
**Analysis**:
- Class 0 immediate SEGV: ✅ RESOLVED (correct offset 0 now used)
- 66K iteration crash: ✅ RESOLVED (offset consistency fixed)
- Box API conflicts: ✅ RESOLVED (unified 3-arg API)
## Technical Details
### Offset Logic Justification
```
Class 0: 8B block → next pointer (8B) fits ONLY at offset 0
Class 1: 16B block → next pointer (8B) fits at offset 1 (after 1B header)
Class 2: 32B block → next pointer (8B) fits at offset 1
...
Class 6: 512B block → next pointer (8B) fits at offset 1
Class 7: 1024B block → offset 0 for legacy compatibility
```
### Files Modified (Summary)
- Core API: `box/tiny_next_ptr_box.h`
- Hot paths: `hakmem_tiny_hot_pop*.inc.h`, `tiny_fastcache.h`
- TLS layers: `hakmem_tiny_tls_list.h`, `hakmem_tiny_tls_ops.h`
- SuperSlab: `superslab_inline.h`, `tiny_superslab_*.inc.h`
- Refill: `hakmem_tiny_refill.inc.h`, `tiny_refill_opt.h`
- Free paths: `tiny_free_magazine.inc.h`, `tiny_superslab_free.inc.h`
- Documentation: Multiple Phase E3 reports
## Remaining Work
None for Box API offset bugs - all structural issues resolved.
Future enhancements (non-critical):
- Periodic `grep -R '*(void**)' core/` to detect direct pointer access violations
- Enforce Box API usage via static analysis
- Document offset rationale in architecture docs
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-13 06:50:20 +09:00
|
|
|
g_fast_head[3] = tiny_next_read(3, head);
|
2025-11-05 12:31:14 +09:00
|
|
|
uint16_t count = g_fast_count[3];
|
|
|
|
|
if (count > 0) {
|
|
|
|
|
g_fast_count[3] = (uint16_t)(count - 1);
|
|
|
|
|
} else {
|
|
|
|
|
g_fast_count[3] = 0;
|
|
|
|
|
}
|
|
|
|
|
return head;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#endif // HAKMEM_TINY_HOT_POP_INC_H
|