feat: Safety wrapper and constant-time implementation (#55)

Browse Source

Safety wrapper (llama-turbo.h, llama-turbo.cpp):
- Input validation (dimension must be power of 2, 16-4096)
- Null pointer checks
- Invalid norm detection (NaN/Inf/negative)
- Error codes for all failure modes
- Safe API: polar_quant_encode_turbo4_safe()

Constant-time quantization:
- ct_fabsf: bitwise absolute value (no branches)
- ct_select: bitwise selection (no branches)
- Always examines all 16 centroids
- No data-dependent branches in packing

Metal shader (ggml-metal-turbo.metal):
- Buffer bounds checking on all accesses
- Invalid norm handling (outputs zeros)
- Thread ID validation
- Constant-time dequantization kernel

Tests (tests/test_safety.py):
- 15 tests, all passing
- Power of 2 validation
- Dimension bounds checking
- Buffer size verification
- Packing correctness

Closes #55

...

This commit is contained in:

Alexander Whitestone

2026-04-14 22:14:51 -04:00

parent 7a7ce0e652

commit d750ca4224

5 changed files with 561 additions and 41 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

BIN

tests/__pycache__/test_safety.cpython-312-pytest-9.0.2.pyc Normal file

Binary file not shown.