Add optimization --cuda-stream

See also the readme for more details

README.md

@@ -88,6 +88,8 @@ Without any cmd flag, Forge can run SDXL with 4GB vram and SD1.5 with 2GB vram.
 
 3. `--cuda-malloc` (This flag will make things **faster** but more risky). This will ask pytorch to use *cudaMallocAsync* for tensor malloc. On some profilers I can observe performance gain at millisecond level, but the real speed up on most my devices are often unnoticed (about or less than 0.1 second per image). This cannot be set as default because many users reported issues that the async malloc will crash the program. Users need to enable this cmd flag at their own risk.
 
+4. `--cuda-stream` (This flag will make things **faster** but more risky). This will use pytorch CUDA streams (a special type of thread on GPU) to move models and compute tensors simultaneously. This can almost eliminate all model moving time, and speed up SDXL on 30XX/40XX devices with small VRAM (eg, RTX 4050 6GB, RTX 3060 Laptop 6GB, etc) by about 15\% to 25\%. However, this unfortunately cannot be set as default because I observe higher possibility of pure black images (Nan outputs) on 2060, and higher chance of OOM on 1080 and 2060. When the resolution is large, there is a chance that the computation time of one single attention layer is longer than the time for moving entire model to GPU. When that happens, the next attention layer will OOM since the GPU is filled with the entire model, and no remaining space is available for computing another attention layer. Most overhead detecting methods are not robust enough to be reliable on old devices (in my tests). Users need to enable this cmd flag at their own risk.
+
 If you really want to play with cmd flags, you can additionally control the GPU with:
 
 (extreme VRAM cases)