scott
8de67c8bd9
Some checks failed
Build ROCm Image / build (push) Failing after 11s
ROCm 7.2 + PyTorch 2.11.0 has a bug where PyTorch passes workspace=0 to MIOpen convolutions, forcing fallback to the slow GemmFwdRest solver. This caused s3gen.inference to take 15-22s instead of <5s, making synthesis 3-4x slower than real-time audio playback. ROCm 6.1 allocates workspace correctly so MIOpen picks fast GEMM solvers without needing torch.compile workarounds. Changes: - Base image: rocm/dev-ubuntu-22.04:7.2 → 6.1 - torch 2.11.0 → 2.5.1 (rocm6.1 wheel index) - Add pytorch_triton_rocm==3.1.0 - transformers 5.2.0 → 4.46.3, safetensors 0.5.3 → 0.4.0 - s3tokenizer unpinned → 0.3.0 - resemble-perth==1.0.1 directly (v1.0.1 is pip-installable; drop stub) - Drop Dockerfile perth_stub steps - Drop torch.compile and timing patches from engine.py (not needed) - Drop multi-pass warmup from main.py (torch JIT warmup not needed) - Drop ROCm 7.2-specific env vars from docker-compose.yml Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
38 lines
936 B
YAML
38 lines
936 B
YAML
services:
|
|
chatterbox-whisper:
|
|
image: git.sdgarren.com/scott/rocm-chatterbox-whisper:latest
|
|
build:
|
|
context: .
|
|
dockerfile: Dockerfile.rocm
|
|
restart: unless-stopped
|
|
ports:
|
|
- "${WYOMING_PORT:-10200}:10200"
|
|
devices:
|
|
- /dev/kfd
|
|
- /dev/dri
|
|
group_add:
|
|
- video
|
|
- render
|
|
ipc: host
|
|
shm_size: 8g
|
|
security_opt:
|
|
- seccomp=unconfined
|
|
ulimits:
|
|
nofile:
|
|
soft: 65536
|
|
hard: 65536
|
|
volumes:
|
|
- ./config.yaml:/app/config.yaml
|
|
- ./voices:/app/voices
|
|
- ./reference_audio:/app/reference_audio
|
|
- hf_cache:/app/hf_cache
|
|
environment:
|
|
- HF_HUB_ENABLE_HF_TRANSFER=1
|
|
# Required for RX 6700 XT (gfx1031) - not natively supported in ROCm.
|
|
- HSA_OVERRIDE_GFX_VERSION=10.3.0
|
|
# Disable MIOpen's SQLite cache — avoids crashes writing benchmark results.
|
|
- MIOPEN_DISABLE_CACHE=1
|
|
|
|
volumes:
|
|
hf_cache:
|