[Audit][Medium] GpuBlockBuffer.getSlotForChunk uses O(n) linear scan instead of O(1) hash lookup

[MichaelFisher1997]

🔍 Module Scanned

(automated audit scan)

📝 Summary

The and functions perform O(n) linear scans through a hashmap's entries to find a chunk by its coordinates. For a buffer supporting up to 16,384 chunks, this causes significant performance degradation during GPU meshing operations where neighbor chunk slots are queried repeatedly.

📍 Location

• File:
• Function/Scope: (line 131) and (line 99)

🔴 Severity: Medium

• Medium: Performance degradation, missing error handling, suboptimal patterns

💥 Impact

During GPU compute meshing in , four neighbor chunk slots are queried per chunk via . Each query performs a full linear scan through up to 16,384 hashmap entries. With hundreds of chunks per frame, this creates a measurable bottleneck in the meshing pipeline.

🔎 Evidence

The underlying storage is which stores as the value. To find a slot by chunk coordinates requires iterating all entries - even though the hashmap is already indexed by slot (a simple usize).

Compare with at line 151 which correctly uses the hashmap's built-in lookup:

🛠️ Proposed Fix

Replace the with two hashmaps:

1. • keep as-is (maps slot → chunk coords)
2. • NEW (maps chunk coords → slot)

Then in :

And in :

And in total used free shared buff/cache available
Mem: 16373460 1459352 7665044 51264 7655004 14914108
Swap: 3145724 0 3145724:

✅ Acceptance Criteria

• All unit tests in pass
• performs a single hashmap lookup (O(1)) instead of iterating all entries (O(n))
• performs a single hashmap lookup instead of iterating all entries
• Memory overhead is minimal: one additional usize per allocated chunk slot
• The fix has been verified with Zig 0.16.0 + SDL3 Dev Environment
  Compiler: 0.16.0
  assets/shaders/vulkan/ui.frag
  assets/shaders/vulkan/shadow.frag
  assets/shaders/vulkan/ssao_blur.frag
  assets/shaders/vulkan/ui_tex.vert
  assets/shaders/vulkan/lpv_inject.comp
  assets/shaders/vulkan/taa.frag
  assets/shaders/vulkan/ui_tex.frag
  assets/shaders/vulkan/bloom_downsample.vert
  assets/shaders/vulkan/bloom_upsample.vert
  assets/shaders/vulkan/debug_shadow.vert
  assets/shaders/vulkan/fxaa.vert
  assets/shaders/vulkan/post_process.vert
  assets/shaders/vulkan/shadow.vert
  assets/shaders/vulkan/sky.vert
  assets/shaders/vulkan/ssao.vert
  assets/shaders/vulkan/taa.vert
  assets/shaders/vulkan/terrain.vert
  assets/shaders/vulkan/ui.vert
  assets/shaders/vulkan/ui_tex.vert
  assets/shaders/vulkan/water.vert
  assets/shaders/vulkan/bloom_downsample.frag
  assets/shaders/vulkan/bloom_upsample.frag
  assets/shaders/vulkan/debug_shadow.frag
  assets/shaders/vulkan/fxaa.frag
  assets/shaders/vulkan/g_pass.frag
  assets/shaders/vulkan/post_process.frag
  assets/shaders/vulkan/shadow.frag
  assets/shaders/vulkan/sky.frag
  assets/shaders/vulkan/ssao.frag
  assets/shaders/vulkan/ssao_blur.frag
  assets/shaders/vulkan/taa.frag
  assets/shaders/vulkan/terrain.frag
  assets/shaders/vulkan/terrain_debug.frag
  assets/shaders/vulkan/ui.frag
  assets/shaders/vulkan/ui_tex.frag
  assets/shaders/vulkan/water.frag
  assets/shaders/vulkan/culling.comp
  assets/shaders/vulkan/depth_pyramid.comp
  assets/shaders/vulkan/lpv_inject.comp
  assets/shaders/vulkan/lpv_propagate.comp
  assets/shaders/vulkan/mesh.comp
  assets/shaders/vulkan/ssao.frag
  assets/shaders/vulkan/ssao.vert
  assets/shaders/vulkan/taa.vert
  assets/shaders/vulkan/sky.frag
  assets/shaders/vulkan/water.vert
  assets/shaders/vulkan/debug_shadow.vert
  assets/shaders/vulkan/debug_shadow.frag
  assets/shaders/vulkan/culling.comp
  assets/shaders/vulkan/depth_pyramid.comp
  assets/shaders/vulkan/shadow.vert
  assets/shaders/vulkan/g_pass.frag
  assets/shaders/vulkan/sky.vert
  assets/shaders/vulkan/terrain.vert
  assets/shaders/vulkan/ui.vert
  assets/shaders/vulkan/water.frag
  assets/shaders/vulkan/lpv_propagate.comp
  assets/shaders/vulkan/mesh.comp
  assets/shaders/vulkan/terrain.frag

📚 References

• defined in with proper hash/eq implementation
• in Zig standard library provides O(1) average-case lookup
• GPU meshing occurs in where neighbor slot lookups are frequent

[github-actions]

The audit finding is accurate. Here's my analysis:

Root Cause Confirmed:

• getSlotForChunk (line 131-138) iterates all slot_to_chunk entries to find a matching ChunkSlot, which is O(n) where n can be up to 16,384.
• freeByChunk (line 99-110) has the same O(n) linear scan issue.
• slot_to_chunk maps slot → ChunkSlot, but there's no reverse index to look up a slot by chunk coordinates.

Call Sites Making This a Bottleneck:

• gpu_mesher.zig:294-297: Four calls to getSlotForChunk per chunk neighbor query
• gpu_acceleration_coordinator.zig:72: Uses getSlotForChunk during chunk upload
• gpu_block_buffer.zig:124: updateBlock uses getSlotForChunk

Proposed Fix is Correct:
Add a chunk_to_slot: std.AutoArrayHashMapUnmanaged(ChunkSlot, usize) to provide O(1) reverse lookup. Memory overhead is one additional usize per allocated slot (~64KB for 16K chunks).

Suggested Implementation Approach:

1. Add chunk_to_slot hashmap in GpuBlockBuffer.init
2. In allocate(): add chunk_to_slot.put(allocator, ChunkSlot{cx, cz}, slot)
3. In free(): add chunk_to_slot.swapRemove(chunk_key)
4. Replace the iteration in getSlotForChunk with a single chunk_to_slot.get() call
5. Replace the iteration in freeByChunk with chunk_to_slot.get() + existing free()

This is a self-contained change in one file (modules/world-meshing/src/gpu_block_buffer.zig) with no cross-subsystem impact.

opencode session  |  github run