Adding performance tuning

gnommo/cache.py

@@ -11,10 +11,42 @@ Files are looked up first locally, then in the cache at:
 """
 
 import configparser
+import os
 from pathlib import Path
 from typing import Optional, Tuple
 
 _cache_config: Optional[dict] = None
+_perf_config: Optional[dict] = None
+
+
+def get_ffmpeg_thread_count() -> int:
+    """Return FFmpeg thread count based on [performance] cpu_limit in ~/.gnommo.conf.
+
+    cpu_limit is a fraction of logical CPUs (e.g. 0.8 = 80%).
+    Defaults to 1 when not configured, which is safe on memory-constrained machines.
+
+    Example ~/.gnommo.conf:
+        [performance]
+        cpu_limit = 0.8
+    """
+    global _perf_config
+    if _perf_config is None:
+        config_path = Path.home() / ".gnommo.conf"
+        _perf_config = {}
+        if config_path.exists():
+            cfg = configparser.ConfigParser()
+            cfg.read(config_path)
+            if cfg.has_option("performance", "cpu_limit"):
+                try:
+                    _perf_config["cpu_limit"] = float(cfg.get("performance", "cpu_limit"))
+                except ValueError:
+                    pass
+
+    cpu_limit = _perf_config.get("cpu_limit")
+    if cpu_limit is None:
+        return 1
+    cpu_count = os.cpu_count() or 1
+    return max(1, int(cpu_count * cpu_limit))
 
 
 def load_cache_config() -> Optional[Path]:

gnommo/preprocessor.py

@@ -18,6 +18,12 @@ from .models import (
 )
 from typing import Union, Optional
 
+def _tc() -> str:
+    """Return FFmpeg thread count string from ~/.gnommo.conf [performance] cpu_limit."""
+    from .cache import get_ffmpeg_thread_count
+    return str(get_ffmpeg_thread_count())
+
+
 # Number of parallel workers for chunk processing
 DEFAULT_CHUNK_WORKERS = 1
 
@@ -771,7 +777,7 @@ def apply_combined_video_filters(
     # Build FFmpeg command
     cmd = ["ffmpeg", "-y"]
     # Global options before -i (after -i they become output options and don't limit filter threads)
-    cmd.extend(["-threads", "1", "-filter_threads", "1"])
+    cmd.extend(["-threads", _tc(), "-filter_threads", _tc()])
 
     if take is not None:
         cmd.extend(["-t", str(take)])
@@ -1361,7 +1367,7 @@ def _process_chunk_to_prores4444(
     # After -i they become output-stream options and FFmpeg ignores them for the
     # filter graph — each geq stage then spawns one thread per CPU core (11 on M-series),
     # causing the N-way RGBA frame buffer explosion that OOM-kills the process.
-    cmd.extend(["-threads", "1", "-filter_threads", "1"])
+    cmd.extend(["-threads", _tc(), "-filter_threads", _tc()])
 
     # Seek to start time (before input for fast seeking)
     if start_time > 0:
@@ -1693,7 +1699,7 @@ def apply_chroma_key(
     # ProRes 4444 profile for alpha channel support
     cmd = ["ffmpeg", "-y"]
     # Global options before -i
-    cmd.extend(["-threads", "1", "-filter_threads", "1"])
+    cmd.extend(["-threads", _tc(), "-filter_threads", _tc()])
 
     # Add duration limit if specified (before input for efficiency)
     if take is not None:
@@ -1798,7 +1804,7 @@ def apply_mask(
     # Build FFmpeg command
     cmd = ["ffmpeg", "-y"]
     # Global options before -i
-    cmd.extend(["-threads", "1", "-filter_threads", "1"])
+    cmd.extend(["-threads", _tc(), "-filter_threads", _tc()])
 
     if take is not None:
         cmd.extend(["-t", str(take)])

gnommo/renderer.py

@@ -306,7 +306,9 @@ def build_ffmpeg_command(plan: RenderPlan, output_path: Path) -> list[str]:
     # Global thread limits before any -i. Without this, each format=rgba conversion
     # in the filter graph (one per video layer) spawns one swscaler thread per CPU core,
     # causing OOM on Apple Silicon where av_cpu_count() returns 10-11.
-    cmd.extend(["-threads", "1", "-filter_threads", "1"])
+    from .cache import get_ffmpeg_thread_count
+    _tc = str(get_ffmpeg_thread_count())
+    cmd.extend(["-threads", _tc, "-filter_threads", _tc])
 
     # Resolve paths to absolute
     project_path = plan.project_path.resolve()