Chapp-E Memory Layout

Complete memory map and address allocation for all system components

Overview

Chapp-E uses a carefully organized memory layout that separates the kernel code from system data structures. This page documents every memory address used in the system, critical for debugging and development.

⚠️ CRITICAL: Memory addresses below 0x200000 (2MB) are reserved for the kernel and boot structures. All system data structures must be placed at 0x200000 or higher to avoid conflicts.
🌱 Philosophical Context: This memory layout is Chapp-E's "DNA" - the genetic blueprint that defines where everything lives. See Core Philosophy for how this blueprint "comes to life" through the message loop.

Memory Regions

Kernel Region (0x100000 - 0x1FFFFF)

The kernel loads at 0x100000 (1MB) as required by the Multiboot specification. This region contains:

  • Kernel code and data sections
  • Bootloader structures
  • Multiboot information
  • Page tables and GDT
  • Brain Controller state (0x100000-0x1000FF)
  • Consciousness System state (0x100100-0x100102)

⚠️ WARNING: Writing to addresses in this region can cause system crashes. Only the Brain Controller and Consciousness systems use small data structures here.

Linker Memory Layout: The linker (chappe-ld) defines memory addresses for executables. These are stored as metadata in the `.chappe` header:

  • Code Base: 0x100000 (executable code sections)
  • Data Base: 0x200000 (initialized data sections)
  • BSS Base: 0x300000 (uninitialized data sections - metadata only)
  • Entry Point: 0x100000 (default _start address)
✅ NOTE: The linker's BSS base (0x300000) is stored as metadata in the `.chappe` header, not used for actual loading. Executables are stored as DAG-FS patterns and loaded into Working Memory (0x200000) or other execution regions when run. The conflict is theoretical only - executables are never loaded at 0x300000.

System Data Region (0x200000+)

All system data structures are placed at 0x200000 (2MB) or higher to ensure safe separation from kernel code.

Allocated Systems:

  • 0x200000: Working Memory (Executive Control)
  • 0x201000: Planning System (Executive Control)
  • 0x202000: Neuromodulator System (Limbic System)
  • 0x202100: Blood-Brain Barrier (Hypothalamus)
  • 0x202200: Hippocampus Module
  • 0x202300: Astrocyte Module (Glial Cells)
  • 0x202400: Microglia Module (Glial Cells)
  • 0x203000: Shell Command System
  • 0x204000-0x2FFFFF: Available for future systems

Hardware Memory (0xB8000)

VGA text mode buffer for display output:

  • 0xB8000: VGA text buffer (80×25 characters × 2 bytes = 4KB)

Complete Memory Map

VGA Text Buffer (0xB8000 - 0xB8F9F)

✅ VGA TEXT MODE: Standard VGA text mode buffer for 80×25 character display. Each character uses 2 bytes (character + attribute). This is the primary output device for Chapp-E.
Address Size Purpose File
0xB8000 4KB (80 × 25 × 2 bytes) VGA text mode buffer (80 columns × 25 rows × 2 bytes per character) system/shell/formatting_64.asm

Format: Each character = 2 bytes (ASCII character + attribute byte)

Attribute Byte: Lower 4 bits = foreground color, upper 4 bits = background color

See Also: I/O System, Formatting System

Brain Controller (0x100000 - 0x1000FF)

Address Size Purpose File
0x100000 1 byte Controller state (INIT, READY, PROCESS, SLEEP, ERROR) braincontroller_64.asm
0x100001 1 byte Message queue head pointer braincontroller_64.asm
0x100002 1 byte Message queue tail pointer braincontroller_64.asm
0x100003 253 bytes Message buffer (circular queue) braincontroller_64.asm

Total: 256 bytes (0x100000 - 0x1000FF)

Brain Tick System

The Brain Controller's brain_tick function uses the following memory locations:

Address Size Purpose File
brain_tick_counter (data section) 8 bytes Counter for pseudo-random generation in spontaneous activity braincontroller_64.asm
0x5200000 80KB (10,000 qwords) Temporary input buffer for spontaneous activity (shared with DAG-FS) braincontroller_64.asm
0x52C8000 16KB (2,000 qwords) Temporary output buffer for spontaneous activity forward pass braincontroller_64.asm
💓 Note: The brain_tick function uses temporary buffers at 0x5200000 and 0x52C8000 for spontaneous neural activity. These are shared with DAG-FS operations but used briefly during each tick cycle.

Consciousness System (0x100100 - 0x100102)

Address Size Purpose File
0x100100 1 byte Current consciousness state (AWAKE=0, SLEEP=1, UNCONSCIOUS=2) consciousness_64.asm
0x100101 1 byte Last consciousness state (for state change detection) consciousness_64.asm
0x100102 1 byte Power-off flag (0=normal, 1=power-off requested) consciousness_64.asm

Total: 3 bytes (0x100100 - 0x100102)

Working Memory System (0x200000 - 0x2000C2)

✅ SAFE REGION: This system was moved from 0x100200 to 0x200000 to avoid kernel conflicts. All addresses are writable and tested.
Address Size Purpose File
0x200000 1 byte Working memory state (IDLE=0, ACTIVE=1, OVERLOADED=2) working_memory_64.asm
0x200001 1 byte Number of active memory slots (0-3) working_memory_64.asm
0x200002 64 bytes Memory slot 0 (current task) working_memory_64.asm
0x200042 64 bytes Memory slot 1 (background task) working_memory_64.asm
0x200082 64 bytes Memory slot 2 (interrupt handler) working_memory_64.asm
0x2000C2 3 bytes Priority flags (1 byte per slot: HIGH=3, MEDIUM=2, LOW=1, NONE=0) working_memory_64.asm

Total: 194 bytes (0x200000 - 0x2000C2)

Capacity: 3 slots × 64 bytes = 192 bytes of working memory

Planning System (0x201000 - 0x201163)

✅ SAFE REGION: This system was moved from 0x100300/0x100400 to 0x201000 to avoid kernel conflicts. All addresses are writable and tested.
Address Size Purpose File
0x201000 1 byte Planning state (IDLE=0, PLANNING=1, EXECUTING=2, COMPLETE=3, ERROR=4) planning_64.asm
0x201001 1 byte Number of steps in current plan (0-8) planning_64.asm
0x201002 1 byte Current step being executed (0-7) planning_64.asm
0x201003 256 bytes Plan steps buffer (8 steps × 32 bytes each) planning_64.asm
0x201100 64 bytes Current goal storage planning_64.asm

Total: 355 bytes (0x201000 - 0x201163)

Capacity: 8 steps × 32 bytes = 256 bytes for plan steps, plus 64 bytes for goal

Neuromodulator System (0x202000 - 0x202044)

✅ SAFE ADDRESSES: Neuromodulator System placed at 0x202000 (2MB + 8KB) to avoid conflicts. This system simulates chemical/neuromodulator dynamics with regional distribution across 6 brain regions, enabling targeted chemical modulation.
Address Size Purpose File
0x202000 6 bytes Global chemical levels (Dopamine, Serotonin, Acetylcholine, Cortisol, Noradrenaline, Oxytocin) neuromodulator_64.asm
0x202006 36 bytes Regional chemical levels (6 regions × 6 chemicals = 36 bytes)
Regions: Brainstem, Basal Ganglia, Thalamus, Cortex, Limbic, Cerebellum		 neuromodulator_64.asm
0x20202A 8 bytes Last update cycle counter neuromodulator_64.asm
0x202032 16 bytes Recent event history (last 8 events: REWARD, ERROR, STRESS, CALM) neuromodulator_64.asm
0x202042 1 byte Current history index (0-7, wraps) neuromodulator_64.asm
0x202043 1 byte Current mood state (NEUTRAL=0, POSITIVE=1, NEGATIVE=2, STRESSED=3, CALM=4, FOCUSED=5) neuromodulator_64.asm
0x202044 1 byte Glymphatic clearance active flag (0=inactive, 1=active during SLEEP) neuromodulator_64.asm

Total: 69 bytes (0x202000 - 0x202044)

Chemicals: 6 neuromodulators tracked globally + 36 regional values (6 regions × 6 chemicals)

Features: Regional distribution, volume transmission (diffusion), glymphatic clearance during sleep

Blood-Brain Barrier (BBB) System (0x202100 - 0x20212A)

✅ SAFE ADDRESSES: BBB System placed at 0x202100 (2MB + 132KB) to avoid conflicts. This system implements selective filtering of inputs/modulators before they affect brain regions, simulating the real blood-brain barrier's protective function.
Address Size Purpose File
0x202100 1 byte Filter state (ACTIVE=0, BYPASS=1) bbb_64.asm
0x202101 32 bytes Transport log (last 16 transports, 2 bytes per entry: type + timestamp) bbb_64.asm
0x202121 1 byte Current log index (0-15, wraps) bbb_64.asm
0x202122 4 bytes Rejected inputs counter (total rejections) bbb_64.asm
0x202126 4 bytes Accepted inputs counter (total acceptances) bbb_64.asm

Total: 43 bytes (0x202100 - 0x20212A)

Transport Types: Carrier-Mediated, Receptor-Mediated, Adsorptive, Active Efflux

Function: Selective filtering of inputs before they affect brain regions, protecting against "toxins" (errors, overflows)

Hippocampus Module (0x202200 - 0x20224A)

✅ SAFE ADDRESSES: Hippocampus Module placed at 0x202200 (2MB + 136KB) to avoid conflicts. This system implements memory encoding, consolidation, and retrieval - the bridge between working memory and DAG-FS.
Address Size Purpose File
0x202200 1 byte Hippocampus state (IDLE, ENCODING, CONSOLIDATING) hippocampus_64.asm
0x202201 64 bytes Encoding buffer (temporary storage for slot data) hippocampus_64.asm
0x202241 1 byte Number of pending encodings (0-8) hippocampus_64.asm
0x202242 8 bytes Pending encoding list (slot IDs to encode) hippocampus_64.asm

Total: 73 bytes (0x202200 - 0x20224A)

Functions: Memory encoding (working memory → DAG-FS), consolidation (sleep replay), retrieval (DAG-FS → working memory)

See Also: Hippocampus Module, Memory Systems

Astrocyte Module (0x202300 - 0x2023FF)

✅ SAFE ADDRESSES: Astrocyte Module placed at 0x202300 (2MB + 144KB) to avoid conflicts. This system implements energy allocation and resource buffering - the "biological heap allocator" for neural resources.
Address Size Purpose File
0x202300 1 byte Astrocyte state (IDLE, BUFFERING, ALLOCATING) astrocyte_64.asm
0x202301 1 byte Current energy level (0-255) astrocyte_64.asm
0x202302 1 byte Resource buffer size (0-200) astrocyte_64.asm
0x202303 200 bytes Resource buffer (for buffering neural resources) astrocyte_64.asm

Total: 256 bytes (0x202300 - 0x2023FF)

See Also: Memory Systems (Biological Heap Allocator section)

Microglia Module (0x202400 - 0x2024FF)

✅ SAFE ADDRESSES: Microglia Module placed at 0x202400 (2MB + 144KB + 256 bytes) to avoid conflicts. This system implements garbage collection and weak weight pruning - the "immune system" for neural networks.
Address Size Purpose File
0x202400 1 byte Microglia state (IDLE, SCANNING, PRUNING, CLEANING) microglia_64.asm
0x202401 1 byte Activity level (0-255) microglia_64.asm
0x202402 1 byte Prune threshold (weights below this are pruned) microglia_64.asm
0x202403 1 byte Garbage count (number of items collected) microglia_64.asm
0x202404 4 bytes Current scan position in neural network microglia_64.asm
0x202405 250 bytes Work buffer (for scanning and pruning operations) microglia_64.asm

Total: 256 bytes (0x202400 - 0x2024FF)

See Also: Memory Systems (Biological Heap Allocator section)

Shell Command System (0x203000 - 0x203FFF)

✅ SAFE ADDRESSES: Shell Command System placed at 0x203000 (2MB + 48KB) to avoid conflicts. This system handles command parsing, execution, and all shell-related data structures. All data uses fixed memory addresses >= 0x200000.
Address Size Purpose File
0x203000 256 bytes Command input buffer shell_64.asm
0x203100 1 byte Buffer position pointer shell_64.asm
0x203101 64 bytes Parsed command name buffer commands_64.asm
0x203141 256 bytes Arguments buffer commands_64.asm
0x203300 ~1KB Command table (command name pointers + handler addresses) commands_64.asm
0x203700 ~2KB Message strings (help text, usage, etc.) commands_64.asm

Total: 4KB (0x203000 - 0x203FFF)

Functions: Command parsing, command execution, argument handling, help system

See Also: Command System

DAG-FS Filesystem (0x300000 - 0x4FFFFF)

✅ DAG-FS FILESYSTEM: Brain-inspired filesystem storing "files" as distributed patterns of synaptic weights. Uses neural network as storage mechanism. See DAG-FS Filesystem for full documentation.
Address Size Purpose File
0x300000 16MB Pattern Storage (pattern metadata: 64 bytes × 1024 patterns) system/filesystem/dagfs_64.asm
0x400000 1MB Index Storage Base (weight snapshots, text storage) system/filesystem/dagfs_64.asm
0x410000 64KB Tag Storage (128 bytes × 512 tags = 64KB) system/filesystem/dagfs_tags.asm
0x420000 - 0x4FFFFF ~14MB Additional Index Storage (weight snapshots, text data) system/filesystem/dagfs_64.asm

Total: 32MB (0x300000 - 0x4FFFFF)

Pattern Structure: 64 bytes per pattern metadata (pattern ID, tags, nodes, weights, strength, access count, state, file type, file flags)

Max Patterns: 1024 patterns supported (0x300000 - 0x304000)

Tag Structure: 128 bytes per tag (starting at 0x410000), up to 512 tags (0x410000 - 0x41FFFF)

Weight Snapshots: 1024 bytes per pattern (16 inputs × 8 hidden × 8 bytes per qword)

Text Storage: 256 bytes per pattern for tag/name strings (0x400000 + pattern_id × 256)

Storage Mechanism: Patterns stored as weight configurations in neural network layers. Weight snapshots stored in index region.

Implementation Status: ✅ C-based implementation complete - Pattern storage, reconstruction, tags, completion, Hebbian learning

File Locations: system/kernel/filesystem/dagfs.h, system/kernel/filesystem/dagfs.c

Kernel Integration: Initialized during kernel boot as part of neural development sequence (Stage 4)

See Also: DAG-FS Filesystem for complete documentation

Ghost File System (0x500000 - 0x50FFFF)

👻 GHOST FILES: Path → Pattern ID mapping system. Every "file path" is a cue (partial activation pattern) that points to a DAG-FS pattern. Multiple cues can point to the same pattern. See File Philosophy for details.
Address Size Purpose File
0x500000 128KB Ghost File Table (128 bytes × 1024 entries = 128KB) system/filesystem/GHOST_FILES.md (planned)

Ghost File Entry Structure (128 bytes):

  • 0-63: Path string (64 bytes) - The cue/command name
  • 64-71: Pattern ID (8 bytes) - Points to DAG-FS pattern
  • 72: File type (1 byte) - FILE_TYPE_EXECUTABLE, FILE_TYPE_DATA, etc.
  • 73: Flags (1 byte) - File-specific flags
  • 74-127: Reserved (54 bytes)

Max Ghost Files: 1024 entries

Status: ⏳ Planned for Phase 2 (Ghost Files + .chappe Header)

See Also: File Philosophy, DAG-FS Filesystem

100,000 Neuron Network (0x600000+)

🧠 100K NEURON ARCHITECTURE: Chapp-E's core neural network with 100,000 neurons, organized in a brain-inspired hierarchical structure with sparse connectivity. This is Chapp-E's "cortex" - the foundation of all learning and memory.
✅ SAFE ADDRESSES: All 100k network addresses are placed at 0x600000 (6MB) or higher, well above the DAG-FS region (0x300000-0x4FFFFF) and kernel region (0x100000-0x1FFFFF). No conflicts.
Address Size Purpose File
0x600000+ ~784MB Synaptic weights (sparse connectivity: 100k neurons × 1000 connections × 8 bytes)
Layer breakdown:
- Layer 1→2: 10,000 × 1000 × 8 = 80MB
- Layer 2→3: 30,000 × 1000 × 8 = 240MB
- Layer 3→4: 30,000 × 1000 × 8 = 240MB
- Layer 4→5: 28,000 × 1000 × 8 = 224MB		 network_100k_64.asm
0x5000000 - 0x50C7FFF 800KB Neuron activations (100,000 neurons × 8 bytes) network_100k_64.asm
0x50C8000 - 0x5197FFF 800KB Neuron biases (100,000 neurons × 8 bytes) network_100k_64.asm
0x5190000+ ~800MB Sparse connection indices (100k neurons × 1000 connections × 8 bytes) network_100k_64.asm
0x5200000 80KB Temporary input buffer (10,000 qwords = 80KB, shared with DAG-FS and brain_tick) dagfs_64.asm, braincontroller_64.asm
0x52C8000 16KB Temporary output buffer (2,000 qwords = 16KB, shared with DAG-FS and brain_tick) dagfs_64.asm, braincontroller_64.asm

Architecture:

  • Layer 1 (Input): 10,000 neurons (sensory input)
  • Layer 2 (Hidden 1): 30,000 neurons (sparse: ~1000 connections each)
  • Layer 3 (Hidden 2): 30,000 neurons (sparse: ~1000 connections each)
  • Layer 4 (Hidden 3): 28,000 neurons (sparse: ~1000 connections each)
  • Layer 5 (Output): 2,000 neurons (motor/action output)
  • Total: 100,000 neurons

Sparse Connectivity: Each neuron connects to ~1000 others (not fully connected), mimicking real brain connectivity patterns. This reduces memory requirements from ~80GB (fully connected) to ~784MB (sparse).

Functions: Forward pass, Hebbian learning, pattern storage and retrieval

Initialization: Network initialized at boot with small random weights and sparse connectivity patterns. Weights modified through Hebbian learning during remember operations.

See Also: DAG-FS Filesystem for how patterns are stored in this network

Neuron Agent System (0x6000000 - 0x65FFFFF)

✅ NEURON AGENTS: Neuron Agent System placed at 0x6000000 (96MB) to avoid conflicts. This system provides RING-0 access for neurons to interact directly with DAG-FS and system components. Neurons are created during kernel boot as part of the neural development sequence.
Address Size Purpose File
0x6000000 - 0x65FFFFF 96MB Neuron agent control structures (100,000 neurons × 16 bytes = 1.6MB, expandable to 96MB) system/kernel/neural/neuron.c
0x6000000 + (neuron_id × 16) 16 bytes Per-neuron agent data (expandable) system/kernel/neural/neuron.c
0x5000000 Variable Neural network activations system/kernel/neural/neuron.c
0x50C8000 Variable Neural network biases system/kernel/neural/neuron.c
0x5190000 Variable Neural network connection indices system/kernel/neural/neuron.c
0x600000 ~784MB Neural network weight matrices (sparse: ~1000 connections per neuron) system/kernel/neural/neuron.c

See Also: Neuron Agent System

Memory Layout History & Debugging

The Great Memory Conflict (2024)

During the implementation of the Executive Control layer, we discovered a critical memory mapping issue that caused system crashes during kernel loading.

The Problem

Initially, Working Memory was placed at 0x100200 and Planning System at 0x100300/0x100400. These addresses were too close to the kernel at 0x100000, causing:

  • ELF loading crashes in GRUB
  • Only 45 bytes writable from 0x100202
  • Complete failure at 0x100242 (WM_SLOT_1)
  • System would crash before any debug messages appeared

The Investigation

Through systematic binary search debugging, we discovered:

  1. 45-byte boundary: Only 45 bytes were writable from 0x100202
  2. 46-byte crash: Writing the 46th byte caused immediate system crash
  3. Invalid region: 0x100242 (WM_SLOT_1) was completely invalid
  4. Root cause: Memory addresses too close to kernel code region

The Solution

We moved all system data structures to 0x200000 (2MB) or higher:

  • Working Memory: 0x100200 → 0x200000
  • Planning System: 0x100300/0x100400 → 0x201000

This provides a full 1MB buffer between the kernel and system data, ensuring complete safety.

Lessons Learned

📚 MEMORY RULE: In a multiboot kernel environment, always place system data structures at least 1MB away from the kernel load address (0x100000). Use 0x200000 (2MB) or higher for all data structures.

System Integration Points

How Systems Interconnect

The neuromodulator system integrates with all major brain systems:

Consciousness System Integration

  • SLEEP State Trigger: When consciousness enters SLEEP, glymphatic clearance activates
  • Mood → State: High cortisol can force SLEEP state (overload protection)
  • State → Chemicals: AWAKE state allows normal chemical dynamics, SLEEP accelerates clearance

Executive Control Integration

  • Working Memory: Acetylcholine levels affect focus and capacity
  • Planning System: Dopamine levels influence exploration vs. exploitation
  • Regional Modulation: Cortex region chemicals directly affect executive functions

Brain Controller Integration

  • BBB Filtering: All inputs pass through BBB before reaching Brain Controller
  • Event Recording: Brain Controller events (rewards, errors) update neuromodulators
  • System ID Routing: BBB uses system IDs for receptor-mediated transport
  • Brain Tick Heartbeat: The brain_tick function continuously calls neuromodulator diffusion and update, ensuring chemicals flow even when idle
  • Spontaneous Activity: Brain tick generates sparse random neural activity to keep the network "breathing"

Brain Tick System Integration

The brain_tick function is the central heartbeat that integrates all subsystems:

  • Neuromodulator Dynamics: Every tick calls neuromodulator_diffuse and neuromodulator_update for continuous chemical flow
  • Neural Network Activity: Generates spontaneous input and runs forward pass through 100k network
  • Glial Maintenance: Calls microglia_prune_weak_weights and astrocyte_buffer_resource for continuous maintenance
  • Memory Consolidation: Calls hippocampus_process_pending to encode pending memories
  • Sleep Activities: During SLEEP state, consolidates memories and flushes waste via glymphatic system
  • Shell Integration: Shell main loop calls brain_tick when idle, ensuring continuous activity even while waiting for input
💓 Continuous Symbiotic Flow: The brain_tick system ensures Chapp-E never truly "sits idle" - it maintains background neural activity, processes memories, evolves mood, and keeps all subsystems interacting continuously, just like a biological brain.

Future Integration Points

  • Attention Network: Acetylcholine will modulate attention filtering
  • Salience Network: Noradrenaline will affect interrupt handling
  • Cerebellum: Error correction will update cortisol levels
  • Reward System: Will directly boost dopamine in reward pathways

Memory Safety Guidelines

Do's ✅

  • Place all new system data structures at 0x200000 or higher
  • Keep data structures in the same region (e.g., all Executive Control at 0x200000+)
  • Document all memory addresses in this page
  • Test memory writes incrementally when adding new systems
  • Use clear separation between different systems (at least 4KB apart)

Don'ts ❌

  • Never place data structures below 0x200000 (except Brain Controller and Consciousness)
  • Don't assume contiguous memory is available near the kernel
  • Don't write to addresses without testing incrementally first
  • Don't overlap memory regions between systems
  • Don't use addresses in the 0x100000-0x1FFFFF range for new systems

Memory Allocation Strategy

For future systems, use this allocation pattern:

0x200000 - 0x200FFF: Working Memory (1KB)
0x201000 - 0x201FFF: Planning System (4KB)
0x202000 - 0x202FFF: [Next System] (4KB)
0x203000 - 0x203FFF: [Next System] (4KB)
...

This provides 4KB (page-aligned) separation between systems, making it easy to track and debug.

Memory Map Summary

Region Address Range Size System Status
VGA Text Buffer 0xB8000 - 0xB8F9F 4KB VGA Text Mode ✅ Active
Kernel Code 0x100000 - 0x1FFFFF 1MB Multiboot Kernel ✅ Reserved
Brain Controller 0x100000 - 0x1000FF 256 bytes Brain Controller ✅ Active
Consciousness 0x100100 - 0x100102 3 bytes Consciousness System ✅ Active
Linker Code Base 0x100000+ Variable Executable Code Sections (.text) ✅ Active
Working Memory 0x200000 - 0x2000C2 194 bytes Executive Control ✅ Active
Planning System 0x201000 - 0x201163 355 bytes Executive Control ✅ Active
Neuromodulator System 0x202000 - 0x202044 69 bytes Limbic System ✅ Active
Blood-Brain Barrier 0x202100 - 0x20212A 43 bytes Hypothalamus (Diencephalon) ✅ Active
Hippocampus 0x202200 - 0x20224A 73 bytes Hippocampus Module ✅ Active
Astrocyte 0x202300 - 0x2023FF 256 bytes Glial Cells ✅ Active
Microglia 0x202400 - 0x2024FF 256 bytes Glial Cells ✅ Active
Shell Command System 0x203000 - 0x203FFF 4KB Shell System ✅ Active
Available 0x204000 - 0x2FFFFF ~1MB Future Systems 🟡 Available
Linker Data Base 0x200000+ Variable Initialized Data Sections (.data) ✅ Active
Linker BSS Base 0x300000+ Variable (metadata only) Uninitialized Data Sections (.bss) - Stored as metadata in .chappe header, not loaded at this address ✅ Metadata Only
DAG-FS Pattern Storage 0x300000 - 0x3FFFFF 16MB DAG-FS Filesystem ✅ Active
DAG-FS Index Storage 0x400000 - 0x40FFFF 1MB DAG-FS Filesystem (weight snapshots, text storage) ✅ Active
DAG-FS Tag Storage 0x410000 - 0x41FFFF 64KB DAG-FS Filesystem (512 tags × 128 bytes) ✅ Active
DAG-FS Additional Index 0x420000 - 0x4FFFFF ~14MB DAG-FS Filesystem (additional weight snapshots, text data) ✅ Active
Ghost File Table 0x500000 - 0x50FFFF 128KB Ghost File System (1024 entries × 128 bytes) ⏳ Planned
100k Neuron Network - Weights 0x600000+ ~784MB Neural Network (sparse synaptic weights) ✅ Active
100k Neuron Network - Activations 0x5000000 - 0x50C7FFF 800KB Neural Network (100,000 neurons × 8 bytes) ✅ Active
100k Neuron Network - Biases 0x50C8000 - 0x5197FFF 800KB Neural Network (100,000 neurons × 8 bytes) ✅ Active
100k Neuron Network - Connections 0x5190000+ ~800MB Neural Network (sparse connection indices) ✅ Active
Brain Tick - Spontaneous Input Buffer 0x5200000 80KB (temporary, shared with DAG-FS) Brain Controller (spontaneous activity) ✅ Active
Brain Tick - Spontaneous Output Buffer 0x52C8000 16KB (temporary) Brain Controller (spontaneous activity) ✅ Active
Neuron Agent System 0x6000000+ Variable Neuron Agent Control Structures (96MB+) ✅ Active
Available 0x30000000+ Unlimited Future Systems 🟢 Available