3 files changed, 15 insertions, 7 deletions
diff --git a/modules/processing.py b/modules/processing.py
index 2cd0bf34..b237b3c5 100644
--- a/modules/processing.py
+++ b/modules/processing.py
@@ -117,7 +117,12 @@ class Processed:
 def slerp(val, low, high):
     low_norm = low/torch.norm(low, dim=1, keepdim=True)
     high_norm = high/torch.norm(high, dim=1, keepdim=True)
-    omega = torch.acos((low_norm*high_norm).sum(1))
+    dot = (low_norm*high_norm).sum(1)
+
+    if dot.mean() > 0.9995:
+        return low * val + high * (1 - val)
+
+    omega = torch.acos(dot)
     so = torch.sin(omega)
     res = (torch.sin((1.0-val)*omega)/so).unsqueeze(1)*low + (torch.sin(val*omega)/so).unsqueeze(1) * high
     return res
diff --git a/modules/sd_hijack.py b/modules/sd_hijack.py
index c4450ce4..98bb9b7e 100644
--- a/modules/sd_hijack.py
+++ b/modules/sd_hijack.py
@@ -243,12 +243,12 @@ class StableDiffusionModelHijack:
         model_embeddings.token_embedding = EmbeddingsWithFixes(model_embeddings.token_embedding, self)
         m.cond_stage_model = FrozenCLIPEmbedderWithCustomWords(m.cond_stage_model, self)
 
-        if cmd_opts.opt_split_attention:
+        if cmd_opts.opt_split_attention_v1:
+            ldm.modules.attention.CrossAttention.forward = split_cross_attention_forward_v1
+        elif not cmd_opts.disable_opt_split_attention:
             ldm.modules.attention.CrossAttention.forward = split_cross_attention_forward
             ldm.modules.diffusionmodules.model.nonlinearity = nonlinearity_hijack
             ldm.modules.diffusionmodules.model.AttnBlock.forward = cross_attention_attnblock_forward
-        elif cmd_opts.opt_split_attention_v1:
-            ldm.modules.attention.CrossAttention.forward = split_cross_attention_forward_v1
 
         def flatten(el):
             flattened = [flatten(children) for children in el.children()]
diff --git a/modules/shared.py b/modules/shared.py
index c5742c10..eb4db55a 100644
--- a/modules/shared.py
+++ b/modules/shared.py
@@ -36,8 +36,9 @@ parser.add_argument("--unload-gfpgan", action='store_true', help="does not do an
 parser.add_argument("--precision", type=str, help="evaluate at this precision", choices=["full", "autocast"], default="autocast")
 parser.add_argument("--share", action='store_true', help="use share=True for gradio and make the UI accessible through their site (doesn't work for me but you might have better luck)")
 parser.add_argument("--esrgan-models-path", type=str, help="path to directory with ESRGAN models", default=os.path.join(script_path, 'ESRGAN'))
-parser.add_argument("--opt-split-attention", action='store_true', help="enable optimization that reduce vram usage by a lot for about 10%% decrease in performance")
-parser.add_argument("--opt-split-attention-v1", action='store_true', help="enable older version of --opt-split-attention optimization")
+parser.add_argument("--opt-split-attention", action='store_true', help="does not do anything")
+parser.add_argument("--disable-opt-split-attention", action='store_true', help="disable an optimization that reduces vram usage by a lot")
+parser.add_argument("--opt-split-attention-v1", action='store_true', help="enable older version of split attention optimization that does not consaumes all the VRAM it can find")
 parser.add_argument("--listen", action='store_true', help="launch gradio with 0.0.0.0 as server name, allowing to respond to network requests")
 parser.add_argument("--port", type=int, help="launch gradio with given server port, you need root/admin rights for ports < 1024, defaults to 7860 if available", default=None)
 parser.add_argument("--show-negative-prompt", action='store_true', help="does not do anything", default=False)
@@ -48,9 +49,11 @@ parser.add_argument("--gradio-debug",  action='store_true', help="launch gradio
 parser.add_argument("--gradio-auth", type=str, help='set gradio authentication like "username:password"; or comma-delimit multiple like "u1:p1,u2:p2,u3:p3"', default=None)
 parser.add_argument("--opt-channelslast", action='store_true', help="change memory type for stable diffusion to channels last")
 parser.add_argument("--styles-file", type=str, help="filename to use for styles", default=os.path.join(script_path, 'styles.csv'))
-
 cmd_opts = parser.parse_args()
 
+if cmd_opts.opt_split_attention:
+    print("Information: --opt-split-attention is now the default. To remove this message, remove --opt-split-attention from command line arguments. To disable the optimization, use --disable-opt-split-attention")
+
 device = get_optimal_device()
 
 batch_cond_uncond = cmd_opts.always_batch_cond_uncond or not (cmd_opts.lowvram or cmd_opts.medvram)